$ Ace; MO! -L^^SJ^. AN INVESTIGATION OF THE ABSORPTION SPECTRA OF SOME FISH 0'IIS. AND OTHER EXPERIMENTS THOMAS GERALD HOW A Thesis Submitted f o r the Degree of M A S T E R O F A R T S in the Department of. ' PHYSICS ' THE UNIVERSITY OF BRITISH COLUMBIA APRIL, 1935 G 0 H E 1 I . S An Investigation of the Absorption Spectra of Some F i s h Olla Introduction' Page 1 Experimental 2 Results 15 •Discussion of Results 20 The Ion Content of The. Air Introduction 24 Experimental 24 Results 50 F I G U R E S AND TABLES An Investigation of the Absorption JA'vSpigtotra ^ o g v jjome^.-Fish.y§i%k < > F i g . 1 Page 3. F i g . 2 » 4. Fi g . 8 " 5 . F i g . 4 " 8 . F i g . 5 and F i g . 6 " 9. F i g . 7 » 10. F i g . 8 " 11, F i g . 9 " 12. F i g . 9a facing page 14. Fig.10 " 15. Table I. " 16. Fig.11 » 17. Sable II. " 1 8 . Fig.12 & F i g , 13 " 19. Table I I I . " 20; The Ion Content.of the A i r F i g . 1 Page 25. Fig, 2 26. F i g . 3 27, .Fig,. 4 28. 6 To face page30. *** AH INVESTIGATION OF THE ABSORPTION SPECTRA —. .' OP 'SOME PISS OTIS ". ~~ INTRODUCTION. Since i t has been d e f i n i t e l y established that the determination of the intensity of absorption of Vitamin A concentrates and l i v e r o i l s of high Vitamin A potency i n the u l t r a v i o l e t at 3280 A 0 gives values which agree with the b i o l o g i c a l t e s t 1 , It i s possible that the method might somehow be extended to o i l s of lower potency without any loss of accuracy. This report deals c h i e f l y with such an attempt, the o i l under investigation being B r i t i s h Columbia Pilchard body o i l . Several methods are outlined f o r obtaining the extinction c o e f f i c i e n t at 3280 A 0; the absorption spectrum of the o i l i s obtained and investigated. The discrepancy between b i o l o g i c a l tests and physical tests i s discussed. Included in the repart are the d e t a i l s of a method fo r obtaining the u l t r a v i o l e t absorption spectrum of any solution with only one photographic exposure. 1. Coward, Dyer & Morton. (1932) Biochem. J, 26, 1593. « Page 2 «*• . EXPERIMENTAL The extinction c o e f f i c i e n t u i s defined by the equation I = I p 1 0 ~ u x , where I Q i s the l i g h t incident on the absorbing solution and I that which i s transmitted through x om. of the solution. The value of t h i s c o e f f i c i e n t at 3280 A 0 was determined by two d i s t i n c t methods. £P The f i r s t i s a v i s u a l method using a fluorescent screen; the other the photographic procedure tff comparison of i n t e n s i t i e s . In the methods given below, the l a s t three are photographic. Method I. The most convenient way of determining u includes the use of a copper arc, a f i l t e r , a fluorescent screen and same mechanism f o r c o n t r o l l i n g the in t e n s i t y of the incident l i g h t . Such a method has been patented by Adam Hilger Ltd. and the apparatus named a "Vltameter A". Following t h i s p r i n c i p l e an instrument was designed and. used f o r the Vitamin A t e s t s . It i s shown in diagram i n F i g . l . A i s a copper arc run at 3 amperes, on a dir e c t current c i r c u i t of 110 v o l t s j and B g are two small c i r c u l a r apertures; G i s a l i g h t f i l t e r which transmits only the radiation near 3280 A 0. D i s a c e l l containing the solution unter examination; E i s an adjustable sector and F a fluorescent screen of Canafcy glass* Page 3 -A balancing glass plate G is placed i n the beam which does not pass through the solution. of the l i g h t passing through B can be varied and the two images of F can thus be brought to the same brightness^ From the angular setting of the sector log 1 0 can be found and hence u. The balancing plate G was necessary to compensate for the loss of radiation by r e f l e c t i o n and absorption caused by the windows of the c e l l . While quartz windows would transmit more l i g h t at 3280 A 0, i t was found that glass windows made from photographic plate would decrease the intensity very l i t t l e . Moreover, G corrects for any error that might arise because of the F i g . l . By varying the adjustable sector the inteHsity - Page 4 -absorption. The f i l t e r C was a quartz window, sufficient*, l y s i l v e r e d so that one could just see the outline of a window when looking through the f i l t e r . If the f i l m i s too thick i t cute down th* inte n s i t y i n the region of investigation. If i t i s too thin i t i s not s u f f i c i e n t l y s e l e c t i v e . Pig.2 shows the e f f e c t of the f i l t e r on th* radiation from a copper arc. The sector was run at a speed, of 30 revolu-tions per seoond. s u f f i c i e n t l y high so that there would be no f l i o k e r and Talbot's law would then hold. M'g* 2 2, "The apparent tateamity of an intermittent l i g h t i s proportional to the time of each individual f l a s h , provided there i s no f l i c k e r . " «? Page 5 -Method I I . She photographic method f o r comparison of i n t e n s i t i e s requires the " c a l i b r a t i o n " of each photo* graphic plate because the r e l a t i o n between the intensity of the incident l i g h t and the blackening of the plat® i s so complex that i t can not be determined by formula. It i s necessary, therefore, to obtain experimentally a curve showing t h i s r e l a t i o n , (See F i g . S): Knowing the density or blackening produced by an unknown inten s i t y , the l a t t e r can be d i r e c t l y read from the graph. F i g . 3. - Page 6 * Since the r e l a t i o n mentioned above i s a function of \ , the wavelength, i t i s necessary to have ca l i b r a t i o n curves f o r every wave length f o r which u i s to be determined. This method, however, concerns only monochromatic l i g h t of \ = 3280 A 0. A small Hilger quartz spectrograph was used as a monochromatic illuminator by masking the dark-slide opening so that only the radiation of the two strong l i n e s i n the copper arc at 3247 and 3274 A 0 would be exposed to the photographic plate. For c a l i b r a t i o n purposes, a steady l i g h t i s very necessary, unless the c a l i b r a t i o n i s done in.one exposure (see Method IV,) For this reason,an argon glow lamp of the type which i s used i n the study of minerals by fluorescence, was employed. An image of the source was focussed on the s l i t of the spectrograph by means of a quartz condensing lens. Intensities were varied i n known r a t i o s by means of meshes placed adjacent to the condensing lens. When placed at an appreciable distance from the lens, i t was found that they caused shadow effects i n the illumination of the s l i t . There was no trace of such a disturbing e f f e c t when the screens were placed close to the lens. The transmitting power of the screens was measured by placing a photo-electric c e l l i n the position occupied by the s l i t of the spectrograph. i5he c e l l was •» Page 7 -f i r s t tested to see i f i t gave a l i n e a r r e l a t i o n between the photo-*eleotrio current and intensity by using the inverse square law. The transmitting powers of the screens were; #1, 56.8$; #S, 41.2%; #3,20.4$; #4,5.71$, Using these singly and i n combinations i t was possible to cut the o r i g i n a l i n t e n s i t y , a r b i t r a r i l y designated as 100 down to 1 i n eight or nine steps. Fi g . 4 shows a t y p i c a l plate. The radiation used f o r c a l i b r a t i o n was that at the head of a band i n the spectrum of the glow lamp at approximately 3350 A 0, The lower s t r i p showing exposures of the main l i n e s in the copper arc were obtained by placing the absorption c e l l immediately i n front of the upper half of the s l i t . So that the radiation f a l l i n t on this part of the s l i t would have passed through the absorbing solution. Intensities of the absorbed and unabsorbed radiation and hence u, were then determined from the c a l i b r a t i o n curve. A l l exposures were f o r a constant time, 30 sees. Eastman 33 plates were used and were dev-eloped i n Bodlnal which had less tendency to fog the plate than the ordinary pyro developers 2 . While i n the developer the plate was kept moving vigorously. In using a oamel^hair brush to obtain even detelopment i t 3. G.R.Harrison, J,Opt. Soc. of Amer. Mar.1934. - Page 8 -was found that the emulsion was usually scratched i n the process. Density measurements were made with a Moll recording mierophotometer. F i g . 4, Method: I I I . A sector photometer (Adam Hilger l t d . ) 4 was used to deter nine the absorption spectrum and in p a r t i c u l a r the absorption c o e f f i c i e n t at 3280 A 0. A tungsten-steel spark was used as a l i g h t source. 4. Catalogue of the Manufactures of Adam Hilger Ltd • Page 9 -F i g . 5 shows a ppeotrogram taken by this method. Method IV. Fig.5, This remaining method involves the use of u II 1 = Fig.6 Page 10 « a step-sector f o r c a l i b r a t i o n purposes. Fig.6 i l l u s t r a t e s how the absorption speotrum of a solution and c a l i b r a t i o n i n t e n s i t i e s oan be obtained simultaneously on a photographic plate. l i g h t from a point source was condensed upon the s l i t of a quartz spectrograph (Hilger El) by means of a sphero-cylindrical lens suoh that l i g h t remained p a r a l l e l i n the v e r t i c a l plane. Therefore, when the c e l l was placed adjacent to the s l i t in the path of the l i g h t , the horizontal portion of the wall of the c e l l east a d i s t i n c t shadow on the s l i t . The speotrum was therefore divided into two portions, the upper part being the l i g h t which passed through the c e l l and the lower part being the unabsorbed spectrum, as i n Fig.7. It i s necessary i n this method to have even illumination a l l along the s l i t , except where the shadow of the c e l l wall i s , before the absorb-ing material i s introduced into the o e l l . Cyclo*hexane was the solvent - Page 11 -used and since i t has negligible absorption i n the region under investigation, and since quartz windows were used for the c e l l , (Fig.8) any difference of Illumination between the two portions of the spectra would be due to Fig. 8. r e f l e c t i o n on the c e l l windows. Whereas i n the upper beam, when the c e l l i s loaded, there i s only one r e f l e c t i n g surface of importance, while i n the lower path there i s two, there w i l l undoubtedly be an error a r i s i n g which can be overcome by methods suggested below. To ascertain whether or not t h i s error was of importance a photograph was taken with the c e l l - Page 12 " f i l l e d with solvent only and the densities of the two spectra compared on the microphotometer. Fig.9 shows the negligible r e s u l t . o 0 » »-If V Fig.9. However, to overcome even t h i s small error i t would be necessary to design a double c e l l which had two compartments, one f o r the solution, one f o r the solvent. This would then compensate f o r the r e f l e c t i o n . The introduction of a step-sector^ between the s l i t and the c e l l , and the adjustment of the c e l l so 5. G.R.Harrison, J, Opt. Soo. of Amer. Mar,1934. - .Page IS • • that about 2/3 of the s l i t length can be used f o r c a l i b r a t i o n i n t e n s i t i e s , gives a f i n a l photograph similar to F i g , 7. Using the iron aro as a source, i t i s possible to obtain u f o r a s u f f i c i e n t number of wave-lengths to plot the absorption spectrum, A small,sharp absorption band of width less that 25 A 0 might Abe detected In some regions. Using a continuous souroe, t h i s d i f f i c u l t y would disappear. The windows of the ©ell i n Fig,8 ar® much larger than necessary. Rather than c u t t i n g them into a more convenient s i z e , they were cemented on as they were, Different cements were t r i e d f o r t h i s purpose. As the c e l l had to be dismantled frequently to elean i t , i t was found best to use " l i q u i d solder" which does not harm the polished surface of the window* This cement i s also insoluble i n chloroform, ether and ojrclo-hexane. In preparing the solutions, several precautions were found necessary, Pippettes were re~ calibrated because of the viscous o i l . Solutions should'be allowed to stand overnight. When t h i s pre* oaution was not taken, that i s , when readings of the extinction c o e f f i c i e n t of a solution were taken soon after i t s preparation, i t was found that different portions of the solution had different densities. The top portion in the test-tube had a smaller c o e f f i c i e n t than the next, -Page 14 * while the solution at the bottom of the tube was very mueh more dense* This resulted even though th® solution had been shaken immediately aft e r i t s preparation. To be certain there was no contamination from the corks, solutions were always prepared i n SO c c , glassy-stoppered b o t t l e s . Microphotometer readings were done in the usual way using D - log * for the d e f i n i t i o n of intensity. +o The step-sector used i n method IY, f o r c a l i b r a t i o n i s shown In Pig. 9a. It was constructed so that both the l i g h t beams would be intermittent. The use of sectors f o r the reduction of in t e n s i t y f o r 6 photographic work has been proved to cause no errors. The oxidation of the o i l samples was per-formed by bubbling a i r through-the 611 f o r 30 hours. In oxidizing the sample to obtain curve d, Pig*13, a i r was bubbled through the o i l at 60° C. f o r s i x hours and exposed to the radiation from a copper arc for eight hours. To extract the pigment of the pilchard o i l , the o i l was f i r s t dissolved in cyclohexane, a l i t t l e diatomaceous earth added and the mixture shaken. Absorption measurements were taken aft e r f i l t e r i n g . The extraction of the unsaponifiable f r a c t i o n 7 followed the procedure given by Baumann and Steenbeck. 6. G. E. Harrison, J. Opt. Soe. of Amer. Mar. 1934. ?• C. A* Baumann & H, Steenbock, J. B i o l , Chem., 101, 547-60,1933. «• Page 15 ~ RESULTS Method I. Several samples of pilchard o i l were studied; one or two reference cod-liver o i l s , and a standard solution of Vitamin A obtained from B r i t i s h Drug Houses Ltd. To obtain the Vitamin A content i n terms of 1934. International Vitamin A u n i t s , the extinction c o e f f i c i e n t f o r a one-percent solution of 1 cm. length was multiplied by the conversion factor 1600, To prove the consistency of the measure-ments, readings f o r different concentrations of the same o i l were taken and a graph plotted to i l l u s t r a t e Beer's law, which postulates a l i n e a r r e l a t i o n between the extinction c o e f f i c i e n t and concentration. F i g . 10 i s such a graph. Fig.10. • <* Pag© 16 ri Table I. Sector Absorption Vitamin O i l % Concentration Reading Coefficient Content Standard .045 18.5 .69 11,700 CI 1 1.0 14* .81 620 flPC-1 1.0 15i° .76 580 UP § .45 g 80v ) UP 5 .83 21°\ ) ,80* 610 it • • . 1.11 11,5° .48 39° ) tt . •*71. ) • •n- • ' .96 16° ) ,77* 590 tt 1.20 11° ) it .83 20° •) n 1.12 12° ) .81* 620 tt 1.37 6° ) li.Q6 •" 14?. .77:. 590 NP4 .83 27°; ) it 1.10 19gV ) .58* 440 tt 1.40 1 3 ^ ) n 1,05 23° 450 HP 7 1.05 26^°? .53 385 » .75 35® •). n 1.05 ) • .50* 380 tt 1.29 18° ) it .83 32° ) .52* 400 it 1.40 16° •) HP 12 1.0 21° • 61' 460 n 1.0 22° .59 450 -',*>'' 1.0 22° 450 1.39 19® )• it 1.04 28° ) .48* 370 n • 1,05 26° .52 400 it 1.05 27° .50 380 WCI-1 14i° (oxidized) 1,0 .79 600 wCI-1 (in room) 1.0 22° ,61 460 NPC-l(oxid.)1.0 15° .77 590 Absorption c o e f f i c i e n t values i n Table I. are f o r a c e l l of length 2.1 cm. Those marked * are^verage values f o r 1% solutions obtained from graphs similar to F i g . 10. - Page 17 • Method I I . This method was not used for Vitamin A measurements because of i t s inconvenience compared with Method I, However, absorption measurements were taken which showed the accuracy of the procedure. As i n Method I . f a graph was obtained f o r two di f f e r e n t samples of cod-liver o i l , showing the li n e a r r e l a t i o n between concentration and absorption. See P i g . 11, Fig.11. Method I I I . This procedure was adopted mainly f o r the purpose of checking the results of Method I. The - Page 18 -following table w i l l show how the two different experiments give the same r e s u l t s . Table I I . O i l Plate m, % Concentration UP 7 1 1 1 £ • 7.6 1.0 .43 .45 Extinction Coefficient MethT~rnT¥iwrii-.32 .49 .22 .23 .54 .51 .21 .21 METHOD 17. In the manner described i t was found possible to record conveniently on a photographic plate the absorption spectra of several samples of o i l . For reasons pointed out below, the following absorption spectra were obtained; a. Pilchard O i l . N.P.C.I. b. " ,f n - pigment extracted with diatomaceous earth. c. Unsaponifiable f r a c t i o n of Pilchard oil,W.C.I.I d. " of oxidized Pilchard O i l •- i.a,i>i These are shown i n Figs. 12 & 13. • Page 19 -» F i g . 13. Page 20 -DISCUSSION OP RESULTS. • . . . ' i . .,' . . . . . j . ' , .i - . . ' • i . ' j . i . . . . Because the f i r s t procedure described in the e a r l i e r part of the report involves no photography and gives readings d i r e c t , i t i s by f a r the most convenient method. It not only gives consistent results (see Fig.10)„ hut also checks with the photographic method according to Table I I , However, i f the values of the Vitamin A content obtained by Method I are compared with b i o l o g i c a l assays made by the B i o l o g i c a l Board of Canada, i t i s found there Is a large discrepancy between the results of the physical and b i o l o g i c a l methods, as shown In Table I I I , TABLE I I I . Spectroscopic B i o l o g i c a l W.0,1.-1 380 1200 N.P.C.-1 • • 58© 175 Furthermore, the usual value of pilchard o i l , from rat tests, ranges from 100 - 200 Int. units of Vitamin A per gram. Therefore Method I, and indeed any spectroscopic measurement, since Method I has been shown to give readings consistent with other methods, gives incorrect values of the Vitamin A content. This i s to be expected since in o i l s of low potency the contribution to the t o t a l absorption by other * 'Page 21 © materials i n the region of 5280 A° i s of considerable importance. Since i t i s hardly possible that t h i s irrelevant absorption i s related to the Vitamin A content or that i s constant i n a l l pilchard o i l s , no correction factor,can be used to allow f o r i t . In an e f f o r t to explain t h i s extra absorption i t was considered that i t was due to the free f a t t y acids In the o i l , the pigment, or some other uasaponifiable substance. The examination of the two spectra in F i g . 12 shows that while extraction of the pigment cuts down the absorption i n the v i s i b l e by 80$, there i s not an equal diminution of the c o e f f i c i e n t at 5280 A°. This proves that the pigment i s not responsible f o r the bulk of the absorption i n this l a t t e r region. In decolorising the o i l , the vitamin may have been p a r t i a l l y extracted, which would cause a decrease in the absorption i n the ultra*? v i o l e t ; the diatomaceous earth might have absorbed a l i t t l e of the o i l , which i s more possible since the decrease i s not confined to the Vitamin A region; or the pigment may have some s l i g h t absorption in the u l t r a - v i o l e t . Fig.13 shows the results of an investigation to determine the importance of absorption by the unsaponi-f i a b l e f r a c t i o n . From curve "c" i t i s seen that the Vitamin A band i s on the edge of a large absorption band * Page 22 . «*» which extends down into the u l t r a - v i o l e t . It Is this hand that causes the increased absorption at 328QA0. This has been found i n other o i l s of low potency. 8 Measurement of the Vitamin A content of pilchard o i l by a direct reading of i t s extinction c o e f f i c i e n t i s therefore impossible. However, i f the Vitamin A could be eliminated completely from the o i l and the extinc-tion coefficientsbefore and aft e r removal, compared,ta -. quantitative determination should be able to be made. — - — | With this i n mind, the o i l was oxidised, as described, and the unsaponifiable f r a c t i o n studied* Curve "d*?, ••Fig.. 13, shows the r e s u l t . While there i s a difference i n the absorption c o e f f i c i e n t at 3280A0 between curves "c" and "d w equivalent to 180 Vitamin A units, there has also been a decided change i n the shape of the whole curve. This must be due to the destruction of the unsaponifiable materials other than Vitamin A. Because of t h i s i t i s d i f f i c u l t to make any d e f i n i t e conclusion concerning the Vitamin A content. Before the spectro-scopic method of Vitamin A determination in Pilchard O i l can be used, i t w i l l be necessary to devise means f o r removing the Vitamin A completely without destroying any of the other unsaponifiable materials in the o i l . » Page 23 '-• s h o u l d be noted here that absorption measurements of the oxidized o i l i t s e l f are higher than that of the fresh o i l . (see Table I,) An o i l which had been kept at room temperature f o r one month instead of i n a r e f r i g e r a t o r showed a 20$ increase i n absorption although the container was tight corked. • This increase i s due to increase of th® f a t t y acid content of the o i l . Measurements o'f a eofi.-liver o i l and a Vitamin A standard showed the reading of the extinction c e e f f i c i e n t was a r e l i a b l e method f o r Vitamin A assay. The Vitamin A standard was guaranteed to contain 12*500 Int. units per gram. The value according to Method I was 11,700. This i s as correct as opuld as could be expected. The cod-l i v e r o i l had been tested b i o l o g i c a l l y and shoved a value of 600 Int. un i t s . These two results oompared with the previous ones on pilchard body o i l show the necessity f o r a new method f o r a physical determination of the Vitamin A content of low potency o i l s . THE? ION CONTENT OF THE AIR INTRODUCTIOI Many experimenters have found that the measurement of very small direct currents by amplification with thermionic tubes i s more convenient and just as accurate as the old style electrometer method. This report shows how the GenerallEle-ctr 1 c •*FP*54Pliotxon^-has been used f o r the measurement of extremely small direct currents in the study- of the Ion content of the atmosphere. The method Is fundamentally the same as that of K o l l e r 1 but with several modifications, The tube was also used to measure a capacity of the order 30 x 10 farads. Included i n this paper are the readings taken over a period of one month to show the variation of the ion content with atmospheric conditions, EXPERIMENTAL The usual method f o r measuring the ion 2 content of the a i r i s the "aspiration" method. It consists of an aspirator through which the a i r i s drawn at a known speed. An e l e c t r i c f i e l d in the tube forces the negative or positive ions on a c o l l e c t o r and the 1. K o l l e r - Jour. Franklin Inst. Nov.1932. 8. "The E l e c t r i c a l Conductivity of the Atmosphere and . i t s Causes" - Victor F. Hess. Constable, London.(1928) p.24, - Page 25 -quantity of e l e c t r i c i t y thus col l e c t e d i s measured fey the change i n voltage of the c o l l e c t o r . In using the FP-54 the c o l l e c t o r i s connected to the control g r i d and allowed to ''float" f o r SO or 60 seconds. The change i n voltage can he determined from the tube c h a r a c t e r i s t i c , She tube specifications given "by the manufacturers are as follows; Filament - 2,5 volts 90.09 amperes, Plate «* 6 v o l t s , Central g r i d - «4 v o l t s . Space charge g r i d - 4 v o l t s . • Input resistance ..«» 10ifc> ohms. Plate current •«. 40 microamps. JuuUUUUU F i g . l - Page 26 «* The c i r c u i t used i s shown i n F i g , l . It was found that the potential drop across the filament was 2,24 volts when the filament current I f was ,090 amps0 Eg must then he 40,7 ohms to make the plate voltage 6v with respect to the negative side of the filament, B 4 was made twice Eg so that the potential of the space charge g r i d would be * 4 v o l t s . So that the current flowing through R 4 and Eg would be negligible E^ was made equal to 1000 ohms and Eg 500 ohms, Th.e t o t a l current going through Rg w i l l therefore be ,094 amps. To get a g r i d voltage of -4 volts Eg should be 42,5 ohms. The g r i d voltage-plat© current c h a r a c t e r i s t i c was obtained by applying different g r i d voltages with a potentiometer* The curve obtained i s shown i n Fig, 2, - s i - * » " Si. F i g , 2 " Page £ 7 . . Pig 3 shows how the iom c o l l e c t o r wag connected to the g r i d of the tube. A clockwork automatically set the g r i d c i r c u i t at -4 vo l t s every 50 seconds, The plate current was measured with a mirror .galvanometer of low s e n s i t i v i t y * 'The regular current of 40 microamperes was "balanced out with a dry c e l l and variable resistance. C J Pig, 3. In l a t e r work the c o l l e c t o r i n Pig* 3 was redesigned so that there would be no potential barrier at the entrance of the aspirator. The method follows that of Swann ®. Pig, 4 shows the arrangement, 3. Terr, Magn. 19, 171 (1914) -> Page 28 --Fig,; 4. ' The pliotron. i t s e l f was used to measure the capacity of the g r i d c i r c u i t , 0 g. Standard condensers of capacities 48,5 x 10~ 1 2 farads and 15.2 x 1CT 1 2 farads were b u i l t . These were of the coaxial cylinder type, the outer cylinder was 10 cm. long and had a radius of 0.54 em. Sulphur was used as a d i e l e c t r i c and a l l tests were made while the sulphur was s t i l l quite fresh. Capacities were calculated d i r e c t l y from the formula, rraog; r a / r i •; The method of finding Gg was as followss The standard condenser of capacity was charged and then • Page 29 •? Connected to the g r i d . The resultant voltage Y was r determined from the plate current. I f C i s the r e s u l t -ant capacity when the g r i d i s connected to the standard pondenser, we have Using t h i s formula, th© average value of C g was between 31 and 32 x 1Q~ 1 2 farads,. Two different standardscondensers were used to obtain a check on the method They were mounted v e r t i c a l l y and could be connected to the g r i d by swinging, them about t h e i r own axes u n t i l a small wire«arm Joined to the inner cylinder touched i t . A i r was drawn through the c o l l e c t o r by means of a fan driven by an e l e c t r i c motor at a ve l o c i t y of 3150 cc per sec. This value was determined with an anemometer. Taking the mutual conductance M from the ch a r a c t e r i s t i c curve as .1811 microamp/volt t then we can calculate the number of ions per cc, per microamp, change i n I per 60 sees. P . . . This number 9 C x 3 x (o p\ x 3 i s © x tc>* unit \» - Page 30 o RESULTS Daily readings of humidity and ion content sh&w a close r e l a t i o n to each other. This i s shown i n F i g . 5. Because of the presence of a large number of droplets i n th© a i r when the humidity Is high, ions w i l l tend to c o l l e c t on these and become "large ions" of low mobility which are not collected by the apparatus, '..Daily, readings beside an open window were taken over a period of two months, M© r e l a t i o n between any weather conditions, except humidity, and the ion content eould be found. The average ion content over the period was 520 ions per cc. The values ranged between 180 eo and 1600 cc. The presence of the experimenter i n the room i n which readings were taken would tend to decrease the ion content. Room readings were always considerably lower when the window was shut than when i t was open except on exceedingly moist days.