Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A biochemical study of the oils of the Pacific salmon (oncorhynchus) Bailey, Basil Edwin 1936

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1936_A7_B2_B5.pdf [ 19.3MB ]
Metadata
JSON: 831-1.0061972.json
JSON-LD: 831-1.0061972-ld.json
RDF/XML (Pretty): 831-1.0061972-rdf.xml
RDF/JSON: 831-1.0061972-rdf.json
Turtle: 831-1.0061972-turtle.txt
N-Triples: 831-1.0061972-rdf-ntriples.txt
Original Record: 831-1.0061972-source.json
Full Text
831-1.0061972-fulltext.txt
Citation
831-1.0061972.ris

Full Text

A BIOCHEMICAL' STUDY OF THE OILS OF THE PACIFIC SALMON (ONCORHYNCHUS) By Basil E. Bailey A Thesis submitted for the degree of MASTER OF APPLIED SCIENCE in the Department of Chemistry. THE UNIVERSITY OF BRITISH COLUMBIA 1956 - TABLE OF CONTENTS -Page General I n t r o d u c t i o n • . 1 Salmon Body O i l s 2 Sampling 2 Preparation of o i l s 3 P h y s i c a l and chemical c h a r a c t e r i s t i c s 3 Methods 4 Discussion of p h y s i c a l and chemical c h a r a c t e r i s t i c s 7 Vitamin A determinations 9 C o l o r i m e t r i c assay f o r v i t a m i n A 9 B i o l o g i c a l assay f o r vitamin A 10 Discussion of vitamin A assays 13 Vitamin D assay 14 . Discussion of vitamin D assays . 15 Salmon L i v e r O i l s .16 Sampling 16 Preparation of o i l s 17 P h y s i c a l and chemical c h a r a c t e r i s t i c s 17 Discussion of p h y s i c a l and chemical c h a r a c t e r i s t i c s 18 Vitamin A determinations 19 Bio-assays f o r vitamin A 19 C o l o r i m e t r i c assays f o r vitamin A 20 Discussion of vitamin A values 21 Vitamin D assay . 23 Pigments of the Red Muscle of the Sockeye Salmon 23 E x t r a c t i o n and p u r i f i c a t i o n of pigments 24 Examination of pigment f r a c t i o n s . 25 Examination of pigments i n ether l a y e r 25 Examination of p r e c i p i t a t e d red pigment. . . . . . 28 Examination of pigments i n soap s o l u t i o n 28 Discussion of pigment studies. . . . . 29 General Summary •. 29 Acknowledgements 30 Bibliography 31 Copies of p u b l i c a t i o n s dealing w i t h the work of t h i s i n v e s t i g a t i o n . . pocket i n s i d e back cover. - LIST OF TABLES -Page I . P h y s i c a l and chemical constants of salmon body o i l s 7a I I . Bio-assay f o r vitamin A i n salmon body o i l s 13 I I I . Bio-assay f o r v i t a m i n D i n salmon body o i l s 15a IV. D e s c r i p t i o n and analyses of salmon l i v e r samples 17 V. P h y s i c a l and chemical constants of salmon l i v e r o i l s 18a VI. Bio-assay f o r vitamin A i n salmon l i v e r o i l s 20 V I I . Vitamin A potencies of salmon l i v e r o i l s as determined by the c o l o r i m e t r i c t e s t 21 V I I I . Average colours of salmon body o i l s by species 22 IX. Vitamin D assay of salmon l i v e r o i l 23 - LIST OF FIGURES -(Facing) Page I . Map of B r i t i s h Columbia showing areas sampled . . 2. I I . Diagrammatic i l l u s t r a t i o n of the l i n e t e s t . . . . 14. - LIST OF PLATES -I . P o s i t i v e c o n t r o l l i n e t e s t 1 u n i t standard v i t a m i n D . 1 5 . I I . P o s i t i v e c o n t r o l l i n e t e s t 1 u n i t standard vitamin D . . . . . 15. I I I . Negative c o n t r o l l i n e t e s t 15. IV. Skeena Ri v e r sockeye 20 mg. l i n e t e s t . . . . . . 15. V. Rivers I n l e t sockeye 15 mg. l i n e t e s t 15. VI. Fraser R i v e r sockeye 15 mg. l i n e t e s t 15. V I I . Butedale pink 20 mg. l i n e t e s t 15. V I I I . Johnstone S t r a i t s pinks 15 mg. l i n e t e s t . . . . 15 IX. Fraser River pinks 15 mg. l i n e t e s t . 15. X. (a) Ether-soluble pigment 28. (b) Amorphous pigment "salmon a c i d " 28. (c) C r y s t a l l i n e "salmon a c i d " 28. A BIC€Hi_„ICAL STUDY OF _ i _ OILS OF THE PACIFIC SALtoOH (ONCOKHYliCriUS). By B. E. Bailey. The Pacific salmon (genus Oncorhynchus) i s caught in the coastal waters and rivers of the Pacific coast from California to the Bering Sea. There are five species belonging to this genus, the spring salmon, Oncor-hynchus tschawytscha; the sockeye, Oncorhynchus nerka; the coho, Oncor-hynchus kisutchj the humpback or pink salmon, Oncorhynchus gorbuscha; and the chum or dog salmon, Oncorhynchus keta. Although one representative of the genus Salmo, the steelhead, Salmo gardneri, i s also caught in Pacific coast waters, the amount, compared with the Oncorhynchus, i s very small. Since only f i sh from the genus Oncor-hynchus were studied in this work, the word "salmon" w i l l be used to desig-nate only f ish from that genus, and for brevity the common rather than the scienti f ic names for the various species-, w i l l be used. A l l of the f ish used in this investigation were caught in Bri t i sh Columbia waters. The salmon fishery i s by far the most important of Br i t i sh Columbia's f isheries. The total landings for the years 1930, 1931, and 1952 were re-spectively 2,296,231 ewt., 1,287,041 ewt., and 1, 166,671 ewt. The greater part of the fish landed are canned, although some are frozen, some sold fresh and some preserved by salting and smoking. The work here reported includes a study of the potency in vitamins A and D and of some physical and chemical characteristics of the body and l iver o i l s of the salmon. A preliminary study has also been made of the body o i l pigments. Particular attention has been given, i n the case of the body o i l s , to the determination of vitamin D and, i n the l i v e r o i l s , to the determination of vitamin A. The physical and chemical data for the o i l s w i l l be discussed from a biochemical viewpoint. This research was carried out at the Pacific Fisheries Experimental Station at Prince Rupert, B. C. - 2 -and forms a part of a general Investigation of the properties of North Pacific coast f ish o i l s . SALMON BODY OILS In the salmon, the muscle i s the main storage depot for fat. As with most f ish o i l s , this o i l i s mostly l iqu id at room temperature, though i t may contain a small proportion of sol id glycerides. I t i s usually called the body o i l , but might quite as correctly be called the flesh or muscle o i l . While the flesh of the spring salmon varies from white to bright red, the flesh of the other species has each a f a i r l y definite colouration. That of the sockeye is bright red; of the coho, reddish though somewhat paler than the sockeye; and of the pink salmon, or pinks as they are frequently called, orange pink. The flesh of the chum has only a slight yellowish pigmentation. The colours are due to o i l soluble pigments, which give characteristic co l -ours to the body o i l s . The l iver o i l s , on the other hand, are a l l dark coloured, due possibly to extractive matter. The body o i l s used in this work were mainly from two species, sock-eye and pink salmonj the work was planned with a view to studying the var-iations between the o i l s of each of these species from different d i s t r i c t s . The samples of o i l were obtained from canned salmon, which contains only the flesh o i l . SAMPLING Samples of canned, sockeye salmon from the Skeena River, Rivers Inlet, and the Fraser River were secured, and of pink salmon from Butedale, Johnstone Straits , and the Fraser River. Each consisted of £4 one-lb. cans taken at random from the pack of a cannery in the d i s t r i c t represented. The canned sockeye from Rivers Inlet was packed in 1930, a l l of the other samples in 1931. In figure I a map of Brit ish Columbia i s given, showing approximate location and range of the fishing areas sampled. (To face page 2) F I G U R E I - 3 -For the data of this type of investigation s t r i c t ly to represent the season's pack in any d i s t r i c t , the samples should be carefully selected to represent the pack of the season. Unfortunately, when the present work was commenced, no such samples were available, so one half case was selected at random from the pack of a cannery in each of the dis tr icts chosen for study; while the data obtained are representative of the samples, they can not be taken as the average for the d i s t r ic t s studied. The data, however, are very useful in showing the gross differences between body o i l s from sockeye and from pink salmon. ; PREPARATION OF OILS The free o i l was prepared by pressing the canned salmon cold in a small hydraulic press, separating the o i l from the press l iquor, and drying i t with anhydrous sodium sulphate. After pressing, the press cakes were broken up and extracted with ethyl ether. The solutions thus obtained were dehydrated with anhydrous sodium sulphate, and the o i l recovered by removing the solvent i n vacuo. After separate determinations of colour and acid value had been made on the pressed and extracted o i l s , they were combined for the remainder of the physical and chemical tests, and for the vitamin assays. PHYSICAL AND CHEMICAL CHAMCIERISIICS Several chemical studies of salmon body oi l s have been reported in the l i terature . These include comparisons of the chemical characteristics of o i l s from restricted d i s t r i c t s . Bailey and Johnston (1) studied the iodine and hexabromide values of o i l s from canned salmon of a l l five species of the genus Oncorhynchus. Data regarding the free acidity of coho and spring salmon oi l s were given by Brocklesby (2), while Brocklesby and Den-stedt (3) gave a tabular summary of the variation in acid value, iodine value, saponification value, and refractive index for twenty samples of red spring salmon o i l . The f ish used by Bailey and Johnston were a l l caught i n - 4 -the coastal waters of Washington, Oregon and Alaska, while those of the l a t -terjtwo investigations were from northern Bri t i sh Columbia waters. As' i n -dicated above, in the description of the samples, the present investigation included a study of the physical and chemical characteristics of body oi ls of salmon from representative fishing areas in Bri t i sh Columbia. In the present work, the colour, acid value, and percent unsap-onifiable matter were determined, and a study of the unsaturation made by measurements of iodine value, refractive index, and percent ether and chlor-oform insoluble bromides. Determinations of percent ether and chloroform insoluble bromides were made on the fatty acids, a l l other determinations on the natural o i l s . Methods Colour:-Colours were measured in Lovibond units on a Rosenheim-Schuster tintometer, using a 1.0 cm. c e l l . Free ac id i ty : -From 2.0 to 5.0 gnu o i l were weighed out} 25.0 cc. alcohol added and the mixture heated to boiling to dissolve the fatty acids. The result-ing solution was titrated with N/20 aqueous potassium hydroxide, using phenol red together with a small percent of bromthymol blue as an indicator. With this indicator, much sharper end-points could be obtained i n t i t ra t ing the red o i l s than with phenolphthalein. Unsaponifiable matter:-Five gm. o i l were weighed into a 200-cc. Erlenmeyer flask, 50.0 cc. of 95$ ethyl alcohol and"3 cc. of 50$ K0H were added, and the mixture re-fluxed under a slow stream of carbon dioxide for 1 hour. After cooling, the soaps were diluted with 25 cc. N/5 potassium hydroxide, shaken with - 5 -50 cc. ethyl ether and transferred to a separating funnel. The flask was then rinsed with two successive 50 cc. portions of ethyl ether and f ina l ly with 25.0 cc. N/5 potassium hydroxide. The mixture In the separating funnel was then shaken gently and allowed to stand u n t i l the two layers separated. The soaps were drawn off and the ether layer washed, f i r s t with 50 cc. potassium hydroxide to remove the dissolved soaps, and then with successive 30 cc. por-tions of d i s t i l l e d water, u n t i l the wash water was neutral to phenolphthalein. After dehydrating the ether solution with anhydrous sodium sulphate, the bulk of the solvent was d i s t i l l e d off and the residue transferred to a small tared Erlenmeyer flask from which the remaining solvent was removed in vacuo. Iodine value:-0.1500 + 0#0050 gm. o i l were weighed out in a small glass cup and placed i n a glass-stoppered flask. 25.0 cc. of carbon tetrachloride (which had been purified by reduction with Fehling's solution, dried with sodium sulphate and dis t i l led) were added and flask shaken to dissolve the o i l . Twenty-five cubic centimetres Wijs solution were added, the stopper moisten-ed with potassium iodide solution, flask shaken gently and placed i n refr ig-erator. After standing in the refrigerator for at least 30 minutes, 20 cc. of 10$ potassium iodide were added, the flask shaken gently, and 100 cc. d i s t i l l ed water added. The free iodine was then titrated immediately with N/10 sodium thiosulphate to pale straw yellow, using starch indicator for the exact end point. Blanks were run in para l le l , one being titrated f i r s t and one la s t . Refractive index:- Determined with an Abbe refractometer at 25 ° C. Insoluble bromides:-1. Preparation of total fatty acids:- Approximately 5.0 gm. o i l was placed in a 125 cc. Erlenmeyer flask, and saponified by refluxing under inert gas for one half hour with 4 cc. of 50% potassium hydroxide and a small - 6 -amount of 95% ethyl alcohol. The alcohol was d i s t i l l ed off under part ia l vacuum, the soaps diluted with 50 cc. hot d i s t i l l ed water and transferred to a separating funnel. The fatty acids were then liberated with hydrof-chloric acid, and washed free of acid with d i s t i l l ed water. F i f ty cc. ethyl ether were added to dissolve the fatty acids, and solution dehydrated with anhydrous sodium sulphate. F ina l ly , solvent was removed by d i s t i l l -ation i n vacuo. 2. Determination of hexa bromides:- The centrifuge cup and s t i r r -ing rod were tared together and 1.00 gm. i '0.05 gm. fatty acids weighed out into the cup. The sample was dissolved in 10 cc. ethyl ether which, at the temperature of an ice bath, had already been saturated with bromides from another sample of the same o i l . The tube was then placed in the ice bath to cool . When cool, the brominating mixture (prepared by adding 5.00 cc. bromine to 25.00 cc. g lac ia l acetic acid) was added uniformly u n t i l solution was reddish orange, and an excess of 0.5 cc. added. The reaction mixture was stirred well and allowed to stand i n the ice bath for 15 minutes. The tube was then removed from the ice bath and C P . amylene added unt i l the excess bromine was a l l taken up, as indicated by decolorization. The mixture was centrifuged and the supernatant l iqu id decanted and washed with the pre-pared ethyl ether at least three times or more, i f necessary, depending on the colour of the precipitate. The bromides were dried at a low temperature in the vacuum oven, and weighed. The results were expressed as percent of the total fatty acids. 3. Determinatinn of Octa bromides:- Procedure the same as that for hexa bromides, but chloroform, saturated with bromides, used as a solvent i n place of ethyl ether. Since both octa and hexa bromides are insoluble in chloroform, the percent octa bromides i s found by substracting the percent ether insoluble bromides from the percent chloroform insoluble bromides. - 7 -•s. The data for the physical and chemical characteristics are given in Table I. TVi R rtii BB inn nf physin.fl.l and nhaml nal constants. The above data disclose several interesting facts regarding the natural properties of the salmon body o i l s examined. F i r s t , the free acid-i t i e s of the pressed o i l s are a l l low, and are closely grouped for each spe-cies, those of sockeye lying between 0.38 to 0.50 and of the pinks, 0.84 to 0.88. The acid values of the extracted o i l s , on the other hand are very high and vary between wide l imit s , those of the pinks showing decidedly the highest free acidity. This high and variable acidity of ethyl ether ex-tracted salmon body oi ls has since been shown by Brocklesby, i n these labor-atories, (personal communication) to be due to water soluble acidic substances. It can easily be seen that some organic acids occurring in the flesh would be extracted in appreciable amounts by ethyl ether, while the amount extracted by the natural free o i l would be negligible. Lactic acid, for example, i s miscible with both ether and water, but soluble to only a small extent i n o i l . The presence of lac t ic acid in the muscle after death would probably^be due to the breakdown of carbohydrate during muscular exertion in the death struggle of the f i sh . The iodine value is very useful as a measure of the degree of un-saturation of an o i l . When the o i l i s free from oxidized or polymerized t r i -glycerides the iodine value should have an approximately direct; relationship to the refractive index. Departures from a f a i r l y close parallelism of these two factors i s indicative of the fact that part ia l oxidation or polymerization of the o i l has taken place. The approximate parallelism found here i s useful as an indication of the fact that no oxidation or polymerization had taken place in the oi l s during the pressure cooking of the canned f i sh . The percent hexa and octa bromides give an indication of the pro-tO H •H O >> o pp o a H cd CQ <H o co P 03 H - P to W rt t-3 o cq O EH H 03 o 0 CD o rt 03 cd o •H PH •p cn o H LO CO o B to H to CO 03 O O CD • • • • ft • FH CO CD CM H o CM w Oi to to to 03 X • co H Oi H CD 03 in g n to to H CX> to U O CD • • ft • 6 e co LO CD o O CJ) H H LO LO 1—1 O i> o z> z> CD 03 CO CO Oi Q z> £> f> <tf • • • • e • CI) H H H H H H a CD CD •H ? {in to <> to • T3 H F » • • • • O 03 |H LO LO H o LO M > to CD CD H H H H H H CD 1 H o) cd FH to -H CD H O in in CO H Oi H to 13 - H - P • • • • « • cd H H H o o H SR. O a CD • 3 fH CO W z> o CD • H H - P r l H H CO CO O CD X -H • • • • • • <tj > CD O CV H o LO H CD n 3 CD CO o O LO 00 • H H CD H to LO CO CO 00 a cd fn vH • • « • ft <3) > <4H O o o o o o o H • H CO CD CD CD CO o o • • e • • CD o Ol LO 3 . H CA2 CM w O fn H - P « W H 00 to C - Oi O X • • e * e • O &q H CO w o in to to to H H ^ H CO Oi LO * - H • • • ft • • fH O CO CO CD P w w w w Oi H O CD H CD H to to Oi O O FH • • e • « • O &H LO H H H to • • H - P rt Oi CD to LO o rt cd • • « • H CD o CO LO LO to • H - P O rt CM ... O o fH - P fH CD CD CD CD > H t> > • H rt • H CD to •rH >> « H PH CD rt P PH 4 j .p H O -H •H ,rt cd to FH 03 - P cd FH H ttO fH CD T3 to FH CD 03 3 CD CD to CD rt - P to O 0) CD > cd P CQ cd o o • H FH o FH CQ cq 1-3 PH 38Toedg a^9?[oog to p> • -H to rt p ^ •H rt & ^ o H i—I CD CD Pi X rt fl o o •H -H t> > O O rl II II cr; - 8 -portion of fatty acids containing three and four double bonds in the carbon chain. It must be pointed out, however, that the values obtained are not, for several reasons, a s t r i c t ly quantitative measure of the amount of fatty acids containing three and four double bonds. In the f i r s t place the unsat-urated carbon atoms are not completely saturated by addition of bromine, and the solvent appears to have some effect on the amount of bromine absorbed. Secondly the separation of the various bromides depends upon their so lubi l i ty in different solvents. While a re lat ively large amount of one or more may be dissolved, the separation is not absolutely quantitative. The fatty acids, rather than the natural o i l s , were used not only to avoid any discrepancy which might arise due to the presence of triglycerides containing different fatty acids, but also to eliminate the unsaponifiable matter. For both the sockeye and pink salmon o i l s examined there i s a regular decrease in the degree of unsaturation with loca l i ty of catching, from north to south. It i s well known that animal and vegetable o i l s from higher latitudes are, in general, more highly unsaturated than oi l s from the same species of animal or plant from more tropical latitudes. This i s due to the necessity, on the part of the organism, for a more mobile o i l in the colder environment. The regular variation of iodine value with north and south distribution of the samples of this Investigation i s remarkable, howev-er, since difference i n latitude i s not great. The maximum variation i n iodine value of the sockeye o i l was 15.8 while that of the pink salmon o i l was only 3.9. Both pink and sockeye o i l s showed a greater proportion of octa-bromides than hexa bromides in the fatty acids. The higher percentage of octa than hexa bromides was not due simply to the two extra atoms of bromine in the molecule of the la t ter . This can readily be seen from the fact that the molecular weight ratio of the octa to hexa bromides of a fatty acid having - 9 -18 carbon atoms i s approximately whereas the lowest ratio of octa to hexa bromidesin the data given in table I i s This ratio i s practical ly identical for both Butedale and Johnstone Straits pinks—the proportion of octabromides being greater i n the other four samples. In the oi l s of both the sockeye and the pinks there i s a decrease (from north to south) in the proportion of octabromides in the fatty acids. This i s true in the case of the pink salmon oi l s for the hexa bromides i n the fatty acids, as indic-ated by the percent hexa bromides, though not in the case of the sockeye o i l s . VIIAMIH A Hiia'fihMMIIUi.iS Studies on the vitamin potency of salmon body o i l have already been reported by Holmes and Pigott (4), S. and S. Schmidt-Nielsen (5), Ahmad and Drummond (6), Truesdail and Boynton (7), and Tolle and Nelson (8). Trues-da i l and Boynton, in a study of the vitamin A potency of the body o i l s from five species of salmon, found the amount of vitamin A present to be relat ively small. Tolle and Nelson found that the o i l s from canned salmon, though relat ively potent in vitamin D, contained l i t t l e vitamin A. No work has yet been reported on the vitamin potency of the body o i l s of salmon caught i n Brit i sh Columbia waters. Col Qri metric Afifsa.y for Vitamin A. While the biological assay i s at present the most rel iable method of testing o i l s for their potency in vitamin A, the colorimetric test of Carr and Price (9) has proved very useful i n grading o i l s of a similar nature for their potency in this vitamin. In this test a saturated solution of antimony trichloride in chloroform i s added to the solution of the o i l in chloroform. When vitamin A i s present, a blue colour i s developed which, i n the absence of interfering-substances, i s approximately proportional to the amount of vitamin i n the o i l . This colour i s usually measured i n Lovi-bond units on a Rosenheim-Schuster tintometer. It has been found, however, - 10 -that various substances naturally occurring in the o i l s interfere with the test either by inhibit ing the development of the blue colour from the v i t -amin A, or by themselves giving a blue or bluish colour with the antimony t r i -chloride. For this reason, and also because of the great variations i n v i t -amin A potency of different natural o i l s , the technique of the test must be carefully studied when testing a new kind of o i l . Paral le l determinations of vitamin A by both colorimetric and biological methods must be made on several samples. The colorimetric test has been most frequently used with cod l iver o i l s ; Norris and Danielson (10) applied this test to the salmon body o i l s assayed biological ly by Truesdail and Boynton (7). Their colorimetric blue values although low, agreed f a i r l y well with the results of the Biological assays. It was found that with salmon body o i l s satisfactory measurements of the blue colour values could not be made, using the natural o i l s , since the vitamin A potency i s very low and with the unsaponifiable matter, the blue colour was masked by red and yellow. Determination of the vitamin A potency of salmon body o i l by the colorimetric test hence did not seem re l iable . Bjologinal assay fcvr vitamin A. The technique of the biological assay for vitamin A was as follows. Young albino rats were used for the test. They were bred from a homozygous strain receiving a stock ration of 66$ ground whole wheat 55$ milk 1$ iodized salt with a small amount of fresh greens. Corn meal was occasionally substituted for the wheat. From the tenth day after the birth of the l i t t e r s , the mothers were fed a vitamin-restricted diet consisting of: - 11 -34$ yellow corn meal 33$ ground whole wheat 21$ dry skim milk 7$ linseed meal 4$ powdered yeast 0.5$ iodized salt 0.5$ calcium carbonate. The young rats were weaned when 21 days old, and kept on the restricted diet for a further period of 4 to 8 days, after which time they were placed on the vitamin A-free diet . Only those rats weighing between 37 and 45 gm. when from 25 to 29 days old were used. The composition of the vitamin A-free diet was: 18$ extracted casein 63$ corn starch 8% yeast 4$ salt mixture (McCollum #185) 2$ agar 5$ fat (Crisco)* ^Irradiated ergosterol was added to the diet so that every 10 gm. of diet (the approximate daily consumption) contained 2j International Units of v i t -amin D, to meet the antirachitic requirements of the animals. When the rats had been maintained on this diet for 4 to 5 weeks, L they ceased to grow, and began to lose weight. After they had showed constant or declining weight for a period of one week they were given the o i l to be tested in suitable amounts. The o i l s were fed for a period of 8 weeks; growth data were recrpded twice weekly. Test o i l s were administered dai ly, separate from the diet , by means of specially-made oil-feeding pipettes, calibrated to deliver 20 mg. of o i l . - 12 -From preliminary qualitative colorimetric tests i t appeared that the salmon body o i l s were very low in vitamin A. To avoid the consequently large dosage which would be necessary i f the pure o i l were fed, the unsapon-if iable matter was prepared, and dissolved i n a small amount of vitamin-free o i l (Wesson Oil) for feeding. The method of preparation was.as follows:-Several 5.0 to 8.0 portions of o i l were weighed out in 250 cc. flasks. For each gram of o i l , 3.0 cc. alcohol and 1.0 cc. hot concentrated potassium hydroxide were added. The flasks were then f i l l e d with carbon dioxide, stoppered, and shaken to saponify the o i l . The soaps were diluted with 50 cc. N/5 potassium hydroxide and extracted in a separatory funnel with 150 to 200 cc . ethyl ether. Another 50 cc. N/5 K0H was then added to break the emulsion, and the two layers were drawn off separately. After washing the ethyl ether solution of the unsaponifiable matter, and dehydrating i t by means of anhydrous sodium sulphate, a few drops of Wesson o i l were added. When this was not added, considerable d i f f i cu l ty was encountered in dissolv-ing the unsaponifiable matter i n o i l after removal of the volat i le solvent. Also, when exposed in the unsaponifiable matter, vitamin A i s known to be very l ab i l e . This procedure should thus reduce the poss ibi l i ty of part ial oxidation of the vitamin. Most of the solvent was then d i s t i l l ed off and the residue transferred to a small v i a l . The v i a l was warmed s l ight ly , a stream of C0g passed over the surface to carry off the rest of the solvent, and f ina l ly , i t was placed under high vacuum for one hour. Wesson o i l was added to the residue so that the unsaponifiable matter was contained in exactly one tenth the or iginal volume of o i l . .411 manipulations were carried out i n an atmosphere of inert gas. The positive controls received 2 mg. daily of a 500 unit (unit as defined on page 13) cod l iver o i l which was chosen as a standard, made up to 40 mg. with Wesson o i l . - 13 -Biological determinations of vitamin A were made on three o i l s , Skeena River sockeye, Eraser River sockeye, and a composite sample of the three pink salmon o i l s . The data obtained are given in table II . TABLE II Imnn body n i l s O i l used i n test f T OP viT.au O i l feeding 1(=>V<=1 n n A i n No. of - r a t s ;Skeena River Sockeye II II it II II II n n II 150 mg. 200 250 350 4 6 6 5 Fraser " « II n II II II II 350 mg. 400 500 6 5 6 Composite pink salmon 400 mg. 7 Zod l i ve r o i l 2 mg. 7 Negative controls 7 Number surviving test period* None 1 2 4 None 5 5 1 7 None Av. gain per week i n gm. -4.0 -1.0 3.1 4.8 5.6 -4.5 7.1 JDisnuRsi.QTi nf Vitamin k, Aasayfu From the data in table II i t would appear that some of the levels of salmon o i l supplied the animals with sufficient vitamin A for normal growth. The average growth rate of the positive controls was 7.1 gm. per week; of the rats on the 500 mg. level of Fraser River sockeye o i l , 5.6 gm. per week; and of those on the 350 mg. level of Skeena River sockeye o i l , 3.1 gm. per week. However, while a l l of the animals on the 350 mg. level of Fraser River sockeye died, four of those on 350 mg. Skeena River sockeye survived and showed, as already stated, a small gain. Five rats receiving 400 mg. Fraser sockeye o i l showed a gain of 4.8 gm. per week. Thus the Skeena River sockeye o i l was more potent than the Fraser River sockeye o i l . They contained approximately 2.5 and 2.0 units of vitamin A per gm. respectively. The term "unit of vitamin A" being here defined as the minimum amount of growth-promoting substance which, administered daily to vitamin A depleted rats, causes a resumption of growth at a rate equivalent to that of the pos-i t ive controls. The latter received an amount of cod l iver o i l sufficient - 14 -cause resumption of growth at approximately normal rate. Seven rats receiveing daily the vitamin A from 400 mg. of the composite pink salmon o i l showed no VITAMIN D iiSSA I ( response. For the determination of vitamin D, the technique recommended by the American Drug Manufacturers Association (11) was used. Young albino rats were raised by the same method, and from the same strain, as those for vitamin A tests. M others were given the restricted diet from the tenth day after birth of the l i t t e r s , and the latter were weaned when 21 days o ld . To produce rickets, Steenbock's diet #2985 was used. This con-sisted of: 76% yellow corn meal Z0% gluten flour Z% calcium carbonate 1% sodium chloride The animals were given this diet when from 29 to 32 days old. They were selected according to weight. Those chosen in one l i t t e r did not vary more than 10 gm. and the extreme weight l imits for any vitamin D test rats when started on the dist were 45 to 60 gm. Experience had shown in these laboratories that a satisfactory degree of rickets could be produced by this diet in 23 days. Accordingly, at the end of the 23rd day on the diet, one rat from each l i t t e r was k i l l ed and the le f t t i b i a examined by means of the l ine test. This test , i l l u s -trated diagrammatically in figure II, consists essentially in a visual grad-ing of the amount of healing as shown by the width of the provisional zone of calci f icat ion in the t ib ia of the test rat . The bones are sp l i t , immersed in s i lver nitrate, and exposed to a bright l i ght . If the animal was found to be sufficiently rachit ic , the remaining animals of the l i t t e r were immed-iately placed in separate cages and given the o i l s to be tested. The oi ls were administered by the same type of pipettes as were used for the vitamin A dosages. When the test level was not an integral multiple of 20 mg., the (To f a e e page 14) Pr FIGURE II - 15 -amount for which the pipettes were calibrated, the dosage was adjusted with Wesson o i l . Irradiated ergosterol (synthetic Vitamin D) dissolved in corn o i l , prepared by the Fleischmann Co., was given to the positive controls. This was standardized by the standard vitamin D supplied by the National In-stitute for Medical Research, Hampstead, England. One rat unit (0.0001 mg.) of the latter was found to be equal to 0.0002 mg. of the Fleischmann i r rad-iated ergosterol in antirachitic potency. Both were diluted with Wesson o i l to give the necessary amount in a 20 mg. dosage. The test period was six days, At the end of the sixth day a l l of the rats were k i l l ed and the t ibias exam-ined by means of the McCollum l ine test. A l l six samples of salmon body o i l which had been obtained for the investigation were tested for vitamin D. As a preliminary test, Rivers Inlet sockeye o i l was tested at a large number of o i l feeding levels . The other o i l s were tested subsequently using a smaller number of levels . The results are given in table III. (See page 15a for table III) •Discussion of Vitamin D Assays., In interpreting the results of the vitamin D test, the minimum curative dosage was selected as that amount of o i l giving most nearly the same average healing as was produced by one unit of standard vitamin D (0.0001 rag. irradiated ergosterol). This amount of o i l then contained one unit of vitamin D. As shown in table III, the vitamin D potency of the oi l s varied, in the case of both sockeye and pinks, from 50 to 67 units per gram. The number of samples tested was not great, but the variation was small. Representative l ine test photographs are shown in plates I to IX and dia-grammatic i l lus t ra t ion of the l ine test in figure II . It would thus appear that the body oi ls from both sockeye and pink salmon have a moderately high vitamin D-potency. Since the vitamin D potency of cod l i v e r o i l i s from 75 to co H • H O >> XS o O a H cd co fl •rl •3' H w m EH a • H s CO P • H > fn o cti co to al o •rl C P r l CD ft . S t o o j i H b o to CO - p H to 0 SH - P co H 0 - P - P 0 SH bO • r l CD i H hH" > H t o o LO 2> CD CD O LO C D CD CO Cv2 H H W H H H CVJ H W CVJ W H C\2 CVJ CvJ tO H H H K t D t O CV2 LO W CV2 H "sH CD CVJ HC\2 SH •rl O H 3 cd (D O W p • a) O PH • r l Xi H CD CD CD > C H <D H H co L O cvj CD CVJ CvJ H <tf H LO CO LO LO co co co C O l > - C D C V J Z > CD LO CO CD CD CD CD CD CvJ CD CD CD CD O LO O O H r-t CvJ "tf O LO O LO O O H H W WtO-# O LO O LO o o O LO H H LO O H CO H C\J O O o o o o H o CD CD >» >> CD CD CD £: H o ft o o S co co Ctf C O fH P 0 CD i> e • c H s •H fH • H cd to SH U CD 0 CD e > e )^ • H C O P H ^5 e c e r: c C C C r: CD SH >> o CD S Mi c r—1 S o cd o to to JH CD • H > c fte • r l fH CD H PH cd CD t o CD cd c - P s u 3 P H cq CO SH a CD o P a s CO H o cd bO to o SH P H H c H c • H c cd O ••d t o PH CD ft CD P - P Mi - P cd to to • H H •H •H O o cd ft bO cd PH fn t: fn fn P P fH CD PH to CD M O > C • d o CD • r l CD • C fH - p CO 0 o cd 0 > p fH • H P H • H to CD Xi P c CO cd • cd cd e fH b o o fH fH CD 0 i - s P H M s; (To face page 15) PLATS I Positive Control Line Test 1 unit Standard Vitamin D. (To face page 15) PLATE I I Positive Control Line Test 1 unit Standard Vitamin D. (To face page 15) P L A T E I I I Negative Control Line Test. (To face page 15) PLATE IV Skeena River Sockeye 20 mg. Line Test (To face page 15) PLATE V Rivers Inlet Sockeye 15 Mg. Line Test. (To face page 15) PLAXK V I . Fraser River Sockeye 15 Mg. Line Test. (To face Page 15) PLATE VII Butedale Pink 20 Mg. Line Test (To face page 15). PLATE V I I I Johnstone Straits Pinks 15 Mg. Line Test. (To face page 15) PLAlii I i Fraser River Pinks 15 Mg. Line Test. - 16 -100 units per gram, i t i s evident that the body o i l of the salmon is an ex-cellent antirachitic agent. This i s important from a nutrit ional viewpoint, since this vitamin is of least common occurrence i n natural foodstuffs. SALMON LIVER OILS No comprehensive study of the vitamin A potency of salmon l iver o i l and no work whatever on B.C. salmon l iver o i l had been reported at the time that this investigation was undertaken. S. and S. Schmidt-Nielsen (5) and Ahmad and Drummond (6) have both reported finding relat ively large amounts of vitamin A in samples of salmon l i v e r o i l . Recently Lee and Tolle (12) of the United States Bureau of Fisheries reported a study of the vitamin potency of salmon l i v e r and salmon egg o i l s , and the results obtained were substantally the same as those of the present investigation, although most of their samp-les were obtained in the Pacific coastal waters of United States and Alaska. The work here reported was undertaken with a view to investigating the varia-tions in vitamin A potency and chemical characteristics of the l i v e r o i l s from different species of salmon caught in B. C. waters. A preliminary test for vitamin D in one sample of salmon l iver o i l was also made. SAMPLING In selecting samples for an investigation of this type several factors were considered. The three main variables are variation between species, geographical variation, and seasonal variation. Of these, the f i r s t two are most important. Seasonal variation i s less significant, since the commercial fishing season for salmon i s relatively short. A more comprehensive sampling program was undertaken in the case of salmon l ivers than for salmon body o i l s . Livers were obtained from fish landed in northern, central and southern B. 0. , several species being represented from each of these loca l i t i e s . The approximate location and range of the distr icts in which the f ish were caught are shown in figure I. Samples were taken at different times through-- 17 -out the fishing season to give some indication of the seasonal variation. When the l ivers had been removed from the f i sh , they were frozen and shipped to Prince Rupert, where they were stored at - 2 0 ° C. un t i l the o i l was extracted. A small aliquot of the minced l ivers was dried, and ex-tracted i n the Sargent extraction apparatus with ethyl ether for 48 hours to determine the percentage of o i l . Details of the samples and data regarding the percentage l iver in the f ish and o i l in the l ivers are given i n table IV. Table IV Description and Analyses of Salmon Liver Samples. Sample _ no.. Species Distr ict of . catching • Date 1andfid Per cent l i v -ffTS in f i sh . j Per cent o i l in l ivers 1 Spring Skeena r ivei • May 2/32 \ jj,y—aJ-LcX-L ^  ft 1 ft y 4.7 2 Red spring Skeena River ' July 5/53 4.6 5 white spring Skeena River July 5/53 3.9 [ 4. Pink salmon Skeena River Sept 10/52 5 Sockeye Skeena River July 19/52 1.5 5.0 | 6 Chum Butedale July 25/52 2.0 i 5.7 7 Pink salmon Butedale July 25/32 1.8 2.5 8 Coho Butedale July 25/32 1.7 X <Z O . \j 9 Sockeye Butedale ' July 25/52 1.8 5.4 10 Chum Butedale Aug 27/32 5.7 11 Pink salmon Butedale Aug 27/52 5.5 12 Coho Butedale Aug 27/52 5.7 13 Spring Vancouver Apr. 26/32 14 Spring Vancouver August 1952 5.5 15 Chum Vancouver August 1952 5.5 16 Pink salmon j Vancouver August 1952 4.5 17 ! Coho . | Vancouver August 1932 5.9 18 ! Sockeye I Vancouver August 19341 5.0 PREPARATION OF OILS The l i ver o i l s were extracted with petroleum ether, and the sol-ution dehydrated with anhydrous sodium sulphate. The solvent was subsequent-l y removed by d i s t i l l a t i on in vacuo over a hot water bath. The o i l samples were placed in small glass v ia l s , under inert gas. When in use they were o o stored in a refrigerator at 8 C , and at other times in a cold room at -20 C. PHYSICAL JUMI; CHMJOAL ChAiiACTEiilS'i'ICs Several of the oi l s were solid at room temperature and for this - 18 -reason i t was considered advisable to determine the melting points. Consid-erable d i f f icu l ty , however, was experienced in carrying out the measurements as the melting points were not sharp, the o i l softening gradually over a range of several degrees. In order to get soms information regarding the state of the samples at room temperature, the bottles were placed in a ther-o mostat at 15 C. and their condition noted when equilibrium had been reached. Those remaining solid at this temperature were then transferred to a thermo-o stat at 20 C , and their state at equilibrium again observed. The iodine values of the oi l s were determined by the Wijs method and the unsaponifiable matter, by the ethyl ether extraction of the saponified o i l . The procedure for determination of unsaponifiable matter was as already described for salmon body oi l s except that samples of only 1.0 to 2.0 gm. were used, since the amount of o i l available was small. Proportionally less alcohol and potassium hydroxide solution were used for saponification, and 75 to 100 cc. of ethyl ether for extraction. Data for the physical and chemical character-i s t i c s are given i n table V. (See page 18a for Table V) Discussion nf Physical and nhfttni on.1 Qharantft-H st.i f.s. Some interesting facts have been disclosed by the data shown i n table V. In the f i r s t place the iodine values are a l l very high, ranging from 182.0 to 248.C. Secondly, i t i s remarkable that the highest degree of unsaturation i s shown by the o i l s of highest melting point, Butedale pink salmon l iver o i l s , (Samples #8 and #12) both of which were solid at 2 0 ° C. The degree of unsaturation does not appear to have any relation to the amount of unsaponifiable matter present. If this high unsaturation i s not due to constituents of the unsaponifiable fraction, there must be some very highly unsaturated fatty acids present in some of these o i l s , notably in pink sal-mon o i l . The composition of these highly unsaturated o i l s i s receiving fur-ther study in these laboratories. - 18a -T A B L E V . Physical and Chemical Constants of Salmon Liver Oi l s . Sample No. Species i Distr ict of j % Unsap-\ catching j onifiable i j matter Iodine value State at 1 5 ° C. 1 |Spring Skeena River 2 Red spring Skeena River 5 :White spring Skeena River 4 j Pink salmon Skeena River 5 Sockeye Skeena River 6 Chum Butedale 7 :Pink salmon Butedale 8 j Coho Butedale 9 :Sockeye Butedale 10 I Chum Butedale 11 i Pink salmon Butedale 12 ! Coho Butedale 13 Spring Vancouver 14 , Spring Vancouver 15 : Chum Vancouver 16 Pink salmon Vancouver 17 Coho Vancouver 18 [ Sockeye Vancouver 5.3 7.0 6.0 7.3 5.4 203.9 232.4 200.8 194.0 219.1 248.0 196.2 182.0 198. 2 220.2 209.2 182.0 Liquid Liquid Liquid Solid Solid Liquid Solid Solid Solid Liquid Liquid Liquid Solid Liquid Liquid Liquid State at 20 C. Liquid Liquid Liquid Semi-solid Solid Liquid Serai-solid Solid Solid Liquid Liquid Liquid Liquid Liquid Liquid Liquid - 19 -VITAMJi A iJiiilKRialNATIONS As has already been pointed out, the colorimetric test of Carr and Price (9) i s very useful when a large number of similar o i l s are to be com-pared for their potency in vitamin A, Both Schmidt-Nielsen (5), and Ahmad and Drummond (6) obtained satisfactory results in applying the colorimetric test to salmon l i v e r o i l s . In order to establish the constancy of the conversion factor re-lating the colorimetric and biological test s for vitamin A for the oi l s under examination, typical samples were assayed biological ly, and the values thus obtained related to the blue colour developed by the antimony trichloride test. Bioassays for Vitamin A. The samples chosen, Vancouver chum (sample 15), Vancouver sockeye (sample 18), and Vancouver spring salmon (Sample 14) were fed in graded doses to rats which had ceased to grow on a vitamin A-free diet . The growth re-sponse over an 8-week period was compared to that produced during the same period in similar rats when they were fed daily 2 mg. of the cod l iver o i l chosen as a standard, containing 500 rat units of vitamin A per gram. The diets and general technique were as outlined in the section on the biological determination of vitamin A in salmon body o i l s . From preliminary colorimetric tests i t appeared that salmon l iver o i l s were f a i r ly potent in vitamin A. For feeding, the l i ver o i l s and the standard o i l were dissolved in Wesson o i l in such a way that 20, 40 and 60 mg. of the solutions contained the required dosages. These solutions were fed to the rats direct ly from specially calibrated pipettes, similar to those used for feeding salmon body o i l s . The data thus obtained are set forth in table VI from which i t can be seen that 0.7 mg. of chum l iver o i l ; 0.15 mg. of sockeye l iver o i l ; and 0.15 mg. of spring salmon l iver o i l produce a growth rate which closely ap-proximates that produced by 2 mg. of the standard cod l iver o i l which contained - 20 -1 rat unit of vitamin A. The vitamin A potencies of these o i l s expressed in terms of the above unit are therefore chum 1428, sockeye 6667 and spring 6667 units per gram. TAbLE VI. Bio-assay for vitamin A in salroon 1 i V P T n i l p. Sample No. Species Feeding level No. of . rats l i No. rats sur-viving test ;Average gain : per week 1 t in cm - 1 15 Chum 0.40 mg. 5 5 v- 1-"—o»i*.f _ 1.8 0.60 mg. 5 5 5.2 I i 0.70 mg. 5 5 ; 8 . 5 18 Sockeye 0.05 mg. 5 4 -1.6 0.10 mg. . 5 5 4.9 0.15 mg. 5 5 6.7 14 Spring 0.05 mg. 5 • 5 0.5 0.10 mg. 5 5 3.0 0.15 mg. 5 5 6.9 Positive Control C.L.O. 2 mg. 7 7 6.7 Negative Control ! 7 0 -HTest Period = 8 weeks, ftolorlmetriP. Assays for Vitamin A A l l of the samples collected for the investigation were examined by means of the colorimetric test for vitamin A. The technique was that of Drummond and Hilditch (13) (Method B, Page 51). The blue colours v/ere meas-ured i n a Rosenheim-Schuster tintometer. Determinations were made with var-ious concentrations of o i l so that values on both sides of 5 blue units were obtained. These data were plotted and the amount of o i l producing 5 blue units was determined by interpolation. From this the number of blue units equivalent to 1 mg. of o i l when measured at this color intensity was calculated. The number of blue units produced by an o i l i n the colorimetric test i s not a very convenient means of expressing the vitamin A potency. The data are easier to comprehend when expressed in units of vitamin A per gram. It was therefore desirable to compute a conversion factor for the oi l s which had been assayed biologically , : - 21 -•' by relating the values which had been obtained by the two methods of testing and using this factor to calculate the vitamin A potency of the other oi l s from their blue values. Sample 15 produced 4.2 blue units per milligram of o i l , while, by the biological assay, i t was found to contain 1428 rat units of vitamin A per gram. Each blue unit per Mg. i s therefore equivalent to 4.2 = 340 rat units of vitamin A per gram. Compared in the same way the other two samples (14 and 18) were found to have conversion factors of 345 and 353 rat units of vitamin A per gram of o i l for each blue unit developed when the blue value i s measured at the level of 5 blue units/ The average conversion factor can therefore be taken as 340, which value was used to calculate the vitamin A potencies of the other oi ls from their blue values. These data are shown in table VII: TABLE VII Vitamin A potencies of salmon l iver o i l s as determined by the colorimetric Test. Sample No. Species Distr ict of .. natch incr Blue units p<=>T- mgT o-n 1 Spring Skeena River 40.0 2 Red spring Skeena River 60.3 3 White spring Skeena River 55.2 4 Pink salmon Skeena River 7.8 i Sockeye Skeena River 28.5 6 ! Chum Butedale • 2.5 7 Pink salmon Butedale 4.0 8 ! Coho i Butedale 12.5 9 ! Sockeye Butedale 25.0 10 1 Chum Butedale 3.8 11 ; Pink salmon Butedale 4.2 12 coho Butedale 10.0 13 Spring Vancouver , 20.0 14 Spring Vancouver 20.0 15 : Chum Vancouver 4.2 16 Pink salmon Vancouver ! 5.2 17 1 Coho Vancouver 17.0 18 ! Sockeye Vancouver 19.3 Vitamin A units per UQ . grar 13,600 20,502 18,768 2,652 9,690 850 1,360 4,250 8,500 1,292 1,428 3,400 6,800 6,667 1,428 1,768 5,780 6,667 Dlanuanion of Vitamin A V A I U P S , The data show clearly that salmon l iver o i l i s a r ich source of vitamin A. The richest samples are those prepared from the l ivers of spring salmon caught in the Skeena River d i s t r i c t . The Vancouver springs and sock-eye samples were next in potency while the coho and pink samples were least potent. The least potent sample contained almost twice as much and the most potent sample over 40 times as much vitamin A per gram as did the cod l iver o i l used as a standard of comparison. ' The dist inct parallelism between the vitamin A potency of the l iver o i l and the red colour of the body o i l i n the different species i s very str iking. As already stated, the pigmentation of the body o i l in chum, pinks, coho and sockeye i s known to increase in the order given. Data for the average colour of pink and sockeye body oi l s had a l -ready been secured in this investigation. Several cans each of coho and chum salmon from different loca l i t ies were therefore obtained and the o i l pressed out. The average colours of these o i l s , together with average co l -ours for pink and sockeye body oi ls calculated from the data already obtained, are given in table VIII. TABLE VIII O i l Colours Red Yellow (Lovibond Units) Chum salmon body 2.1 4.S pink salmon body 4.8 24.4 Coho salmon body 7.2 29.0 Sockeye salmon body 14.S 27.7 From the data in table VII i t w i l l be seen that the vitamin A potency of the l iver o i l s from each d i s t r i c t , with the exception of spring salmon, increases in the same order as the red colour given above. Since spring salmon vary in bod;/ pigmentation from colourless (white) to deep red, they are not considered in this connection. Euler, Euler and Hellstrora (14), ^oore (15) and others have shown that the pigment carotene can replace vitamin A i n the diet . Olcott and Mc-Cann (16) showed the existence of an enzyme, which they called carotinase, in the l ivers of rats , by means of which carotene could be converted to - 23 -vitamin A in v i t ro . If carotene i s one of the pigments in salmon body o i l , i t would be quite plausible to postulate an interchange of body pigment and vitamin A in the l iver of the f ish during l i f e . From these relationships i t was evident that an investigation of the pigments present in salmon body oils would be of value. A preliminary study was carried out, the results of which are reported following the next section. VlTAi^iiM D ASSAY A preliminary test for vitamin D was made on one of the samples of salmon l i v e r o i l (spring salmon l iver o i l sample #1) to establish the approx-imate potency. The technique used was as already described in the section on the vitamin D assays of salmon body o i l s . The data are given in table IX. TArtT.H! Ty  Dose 0.5 mg. 1.0 mg. 2.0 mg. + c - c # Rats 2 2 2 1 2 No. Healing Line Test Resuits 1* 2 2 2+ 1 1 3+ 4+ The number of rats used i s only sufficient to indicate approximately the potency of the o i l in vitamin D. However, while two rats receiving 1.0 mg. spring salmon l iver o i l dai ly showed no healing, two receiving 2.0 mg. dai ly showed 1+ and 2+ healing. Taking 2.0 mg. as the minimum daily dosage, the o i l would thus contain approximately 500 units of vitamin D per gram. This i s considerably higher than medicinal cod l iver o i l s , which contain from 75 to 100 units of vitamin D per gram. THE PIGMENTS OF THE RED MUSCLE OF THE SOCKEYE SALMON The f i r s t object of investigating the red muscle pigments of the salmon was to find out whether carotene was present in an appreciable quan-t i t y . It was also planned to separate the pigments, i f more than one were - 24 -present, to show whether they were carotlnoids, and, i f so, to classify them. Time and cost of material precluded a complete investigation of the pigments of a l l species, so sockeye was selected as the most suitable, since i t had the deepest pigmentation in the red muscle. When this work was commenced, only two investigations of the pigments in salmon muscle had been reported, those of Krukenberg and Wagner (17), and Newbigan (18). In view of the meagre knowledge of the chemistry of o i l -soluble pigments at that time, a re-investigation was obviously necessary. Quite recently Euler, Hellstrom and Malmberg (19) published the results of an investigation of the pigments in the flesh of the salmon. They isolated a new acidic carotinoid, which they called "salmon acid" . In addition they found some xanthophyll and a l i t t l e carotene present. Their spectroscopic identif ication of these pigments was, however, unsatisfactory, and they re-l i ed mainly on the phase separation between methyl alcohol and petroleum ether for their identif ication. EXTRACTION AilD PUBIIIOATIUN OF PIGfcMTS Several preliminary experiments were carried out to find the best method of extracting the pigments, and of separating them from extraneous material. Extraction with hot 95% alcohol wa3 found to be the most satis-factory method. The alcohol was purified by d i s t i l l i n g i t over sodium hydroxide and s i lver nitrate, u n t i l a 50% solution of potassium hydroxide in i t had only a straw yellow colour. One hundred grams of sockeye salmon red muscle were f inely minced and boiled with several successive portions of purified 95% alcohol. The extract was f i l tered and then refluxed with 20 gm. potassium hydroxide for • | hour to saponify any o i l . After cooling, the alcoholic solution of the soaps was diluted with water and extracted with ethyl ether. Both the ether and soap layers had an orange colour. The soap layer was drawn off and the ether layer washed with d i s t i l l e d water to remove the soaps, a lka l i and a l -cohol. The f i r s t washing had a dist inct orange colour. At the water-ether interface a red flocculent precipitate, similar to the "salmon acid" des-cribed by Euler and his associates (19), separated. This precipitate was drawn off and treated separately. EAAlalJiATIUN OF PIGiAEHT FRACTIONS Examination nf Pigropnts in Ethpn- Lnyft-r. The general methods for separation and identification of carotinoid pigments were applied to the pigments contained in the ether layer, using f i r s t phase separation by immiscible solvents (Willstatter's method), as des-cribed by Palmer (20). The other solution was dehydrated with sodium sul-phate, and the solvent removed in vacuo. The residue was taken up i n ethyl ether - petroleum ether (1:1). This solution was extracted with 70$ methyl alcohol to remove any fucoxanthine, which has six hydroxyl groups in i t s molecule. When the 70$ methyl alcohol solution had been washed with petrol-eum ether - ethyl ether (5:1) to remove the xanthophyil, which is a carotinoid containing two hydroxyl groups, i t was practical ly colourless. The small amount of pigment remaining was taken up in ethyl ether. This solution was concentrated to a small volume and tested with concentrated hydrochloric add. No blue colour developed, indicating the absence of fucoxanthine. The ether mixture was removed from the solution containing the bulk of the pigments by d i s t i l l a t i o n , the residue was dissolved in petroleum ether, and this solution was extracted successively with 85$, 90$ and 92$ methyl alcohol, to separate xanthophyil. The alcohol took up most of the pigment, which would indicate that i t was mainly xanthophyil. However, when this solution was shaken with carbon disulphide the pigment was quantitative-l y taken up by that solvent with a deep red colour, which i s characteristic of the hydrocarbon carotinoids, carotene and lycopin. A small portion of the - 26 -carbon disulphide solution of pigments was allowed to evaporate to dryness; the pigment formed tetragonal crystals - which are shown i n Plate X(a). These crystals do not resemble those of any of the carotinoids, the photomicrographs of which were available. The fact that, in phase distribution between petroleum ether and 85% methyl alcohol, the pigment went into the alcohol layer, whereas on sub-sequently shaking the alcoholic solution with carbon disulphide, i t went into the carbon disulphide layer, indicated that the pigment possibly i s not a carotinoid. Usually carotin and lycopin, the hydrocarbon carotinoids, are separated quantitatively in petroleum ether or carbon disulphide, while the xanthophylls and other hydroxyl-containing members, go into the alcohol layer. The petroleum ether solution after extraction with 92% alcohol, had only a faint yellow colour. It was evaporated to a small volume and the residue tested for carotene by the method of Deleano and Dick (21), which consists in adding a few drops of trichloracetic acidjio a solution of the pigment in petroleum ether which, with carotene develops a blue colour; which upon the addition of water i s decolorized, the carotene separating in the petroleum ether layer with a yellow colour. No carotene was found to be present by this method. The carbon disulphide solution which contained the bulk of the pigments was evaporated almost to dryness, the residue taken up i n petroleum ether and part tested for carotene as above, but this test was also negative. The colours which different carotinoids display in some solvents are characteristic properties and give useful information regarding their identity, lycopin in carbon disulphide possesses a deep red colour, acquir-ing a bluish tinge upon di lut ion, which is a very characteristic test; i t s solution in petroleum ether i s yellow. Xanthophyll gives an orange solution in carbon disulphide, easily distinguishable from colours produced by carotene and lycopin and furthermore produces a green colour with formic acid while - 27 -neither carotene nor lycopin w i l l dissolve in this solvent. Solutions of carotene (carotene supplied by Mead Johnson & Co.), and lycopin (isolated in the laboratory from tomatoes), were prepared for comparison with the salmon pigment. The colours of solutions of the salmon pigment i n carbon disulphide and i n petroleum ether corresponded very closely to those of lycopin; a dilute solution in carbon disulphide showed the characteristic bluish tinge. However, after standing In stoppered test tubes exposed to daylight at room temperature for one week, the lycopin solution had become almost colourless, while the salmon pigment was apparently unaffected. Another characteristic property of carotinoid pigments i s their ad-sorption behaviour. A convenient means of studying this phenomenon i s the chromatographic analysis of Tswett, as,described by Palmer ( 2 0 ) . C hromato-graphic analyses of the salmon pigments and also of carotene, lycopin, and the crude petroleum ether extract of tomatoes for comparison, were carried out. The latter contained carotene, lycopin, xanthophyil c>; , and xanthophyil B; i t s chromatographic analysis by adsorption with calcium carbonate has been described by Coward ( 2 2 ) . The pigment from salmon did not behave l ike any of the others in i t s chromatographic analysis with calcium carbonate. The pigment was adsorbed from several solvents, the best results being obtained with benzene. From this solvent it.was adsorbed and moved slowly ahead, as the column of ad-separating -into, two bands, sorbent was washed with pure solvent,/ The colour of the bands of salmon pig-ment when adsorbed on calcium carbonate from benzene was rose red colour, while lycopin was yellow and carotene was orange. The salmon pigment did not correspond in i t s adsorption behaviour to any of the pigments in the tomato extract. After trying a number of solvents, i t was found that the two - 28 -salmon pigments could, be separated from the calcium carbonate on which they were adsorbed by means of chloroform. They both gave rose pink solutions. Both the ether soluble salmon pigments were adsorbed on alumina, from petroleum ether and carbon disulphide, or benzene. KlA^liiaTIUH UF PiteCIPIlATEiJ ItEu PlGivuiriT The red pigment was easily dissolved i n acetic acid, giving a deep red solution. Part of this solution was neutralized with potassium hydroxide whereupon the pigment reprecipitated, not in a crystalline condition, but as an amorphous substance (A), as found by microscopic examination. Part was shaken with petroleum ether, which took up some of the pigment. The petroleum ether solution, which contained a considerable amount of acetic acid, was divided into 2 parts. One part was allowed to evaporate, the other was neutralized by shaking with potassium hydroxide solution. Although no crystals were obtained from evaporation of the acidic solution, that which had been neutralized gave dist inct crystals (B), which, however, did not appear to be a pigment, since they were not deeply coloured. Microphotographs of the amorphous pigment (A) are shown in plate X(b), while the crystals (B) obtained from the neutralized solution are shown in plate X(c). Euler found his "salmon acid" to be readily soluble in pyridine and benzol, as well as acetic acid. The pigment here described was very sol-uble in acetic acid or pyridine, but only sl ightly soluble in benzol. Fbcntni nation of thR Pigments i n tVm Snap Solution. The pigment in the soap solution could not be completely removed by extraction of the soaps with ethyl or petroleum ether. While there was the possibi l i ty that i t was a part of the ether-soluble pigment which could not be washed out, i t was also possible that i t was a separate pigment. Upon acidification with acetic or hydrochloric acids, the pigment dissolved in the fatty acids. (To face page 28) Plate X (a) Plate X (b) Ether-soluble Pigment Amorphous Pigment "Salmon Acid" Plate X (c) Crystalline "Salmon Acid" - 29 -Pi ,q mission of Pigmp.nt fit.nrii P.H . From the above studies i t i s evident that there are at least three pigments in the red muscle of the sockeye salmon. These are an acidic pigment similar to Euler's "salmon acid" , and two other pigments which are very closely related to each other. The latter two could be separated only by chromatographic analysis; although they do not appear to be identical with any of the known carotinoids, they showed a number of carotinoid properties. The colour of solutions of these pigments in carbon disulphide and petroleum ether resembled those of lycopin. They differed from that pigment in s tabi l -i t y and i n adsorption behaviour. The possible presence of a fourth pigment, acidic in nature, i s suggested. It has been shown that i f carotin i s present, the amount i s too small to play an appreciable part i n the pigmentation of the red muscle. Since the vitamin A potency of the l iver o i l s paralleled the gross pigment-ation of the flesh, i t i s apparent that the pigment which plays the most prominent part in the suggested conversion of body pigment to vitamin A i s not carotene. GMiMlAL SUMJUihl. An examination of six samples of salmon body o i l has been made with a view to showing the variations in vitamin A potency, vitamin D potency and chemical characteristics of the o i l s , due to species and geographical distribution of loca l i t ies of catching. In general i t was found that salmon body oi ls contained very l i t t l e vitamin A, but were comparable in vitamin D potency to cod l iver o i l s . Several physical and chemical characteristics of the salmon body oi ls were determined and a biochemical interpretation for the variations in degree of unsaturation was made. A general survey of the geo-graphic and seasonal variation, and the variation between species in the - 30 -vitamin A potency of salmon l iver o i l s from fish caught in Bri t i sh Columbia water has been made, and from data obtained, a conversion factor relating the vitamin A potency of salmon l iver o i l s as determined, by biological assay to the value found by the colorimetric method has been calculated. The v i t -amin A potency varied from a minimum of 850 units per gram, i n one sample of chum salmon l iver o i l , to a maximum of 20,500 units per gram in a sample of spring salmon l iver o i l . From the data obtained in these investigations a remarkable para l le l -ism between the vitamin A potency of the l i ver o i l s and the average depth of colour of the body oi l s in four species of salmon was noted. In an attempt to throw further l ight on this relationship a study was made of the pigments in the red muscle of the sockeye salmon. ACKMOWLLlGiifaMTS I wish to thank Dr. H. N. Brocklesby of the Pacific Fisheries Experimental Station, Prince Rupert, and Dr. D. B. Finn, late Director of that Station, for their helpful advice, and encouragement, and Mr. J . B. Finn for the care of the experimental animals during the latter part of the investigation. Dr. N. M. Carter, the present director, has also very kindly read the manuscript and offered suggestions regarding the presentation. - 31 -BIBLIOGRAPHY 1. Bailey and Johnston. Determination of hexabromide and iodine values of salmon o i l as a means of identifying species. Ind.Eng.Chem. 10, 999, (1918). 2. Brocklesby, H. N. Hydrolysis of the body o i l of the salmon. Contr.Can. . B io l .F i sh . VII, No. 40 (Series C, Industrial No. 12). (1933). 3. Brocklesby and Denstedt. The industrial chemistry of f ish o i l s with particular reference to those of Bri t i sh Columbia. Biol .Bd.Can.Bull . XXXVII (1933). 4. Holmes and Pigott. Studies of the vitamin potency of cod l i v e r o i l s . XVII. The vitamin potency of salmon body o i l . Boston Med. Surg. J . 193, 725 (1925). 5. Schmidt-Nielsen, S. a.nd S. A difference between the vitamin content of the flesh o i l and the l iver o i l of the salmon. K. Morske Vidensk. Selsk. Farh. 1, 189(1928). 6. Ahmad, B. & J . C. Drummond, The relative vitamin A value of the body and l i v e r o i l s of certain f i sh . Biochem. J . 24, 27 (1930). 7. Truesdail, R. W. & L . C. Boynton. Vitamin A content of the body oi ls of Pacif ic coast salmon. Thesis, Univ. of Washington (1927). Published Ind. Eng. Chem. 23, 1136 (1951). 8. Tol le , C. D. and E. M. Nelson. Salmon o i l and canned salmon as sources of vitamins A and D. -Ind. Eng. Chem. 23, 1066 (1951). 9. Carr, F. H . , and E . A. Price. Colour reactions attributed to vitamin A. Biochem. J . 20, 497 (1926). 10. Norris, E . R., and L . S. Danielson. Comparison of biological and color-imetric assays for vitamin A as applied to f ish o i l s . J . B i o l . Chem. 83, 469 (1929). 11. American Drug Manufacturers Assoc. Vitamin Assay Committee. Annual Re-port 1931. J . Amer. Pharm. Assoc. 20, 588 (1951). 12. Lee, C. F . , and C. D. Tolle , Salmon l i v e r and salmon egg o i l s , vitamin content, and chemical and physical properties. Ind. Eng. Chem. 26, 446 (1934). 15. Drummond, J . C , and L . P. Hi ld i t ch . The relative values of cod l iver o i l s from various sources. H.M.Stationery Office, London (1950). 14. Euler, B .v . s H.v.Euler and H.Hellstrom. A vitaminwirkung der Lipochromei• Biochem. Z . , 203, 370 (1928). 15. Moore, T. Vitamin A and Carotene I. The association of vitamin A act ivity with carotene in the carrot root. Biochem. J . 23, 803 (1929). 16. Olcott, H. S., and D. C. McCann. The transformation of carotene into vitamin A in v i t ro . J.Biol.Chem. 94, 185 (1931). - 32 -17. Krukenberg, C.F.W., and H.Wagner. Ueber Besonderheiten des chemisches Baues contractiler gewebe. Z .B io l . 21, 24 (1885). 18. Newbigan, M.I. PignientB in the muscle and ovary of the salmon and their exchange. Report of investigations on the l i f e history of the salmon by D. Noel Paton. Fishery Board of Scotland, Article XV. (1898). 19. Euler, H. v . , H. Hellstrom and M. Malmberg. Salmensaure, ein Carotinoid des Lachses. Svensk Kemisk Tidskrift , 45, 151 (1935). 20. Palmer, L . S. Carotinoids and related pigments. Chem.Cat.Co., Hew York, (1922). 21. Delano, A . T . , and J . Dick. Studies in carotene. I. Mew methods for i t s preparation, demonstration and determination. Biochem. Z. 259, 110-33, (1933). 22. Coward, K. H. Some observations on the extraction and estimation of lipochromes from animal and plant tissues. Biochem. J . XVIII, 1114 (1924). Pacific Fisheries Experimental Station, Prince Rupert, B. C. March, 1936. BIOLOGICAL BOARD OF CANADA PROGRESS REPORTS O F PACIFIC BIOLOGICAL STATION N A N A I M O , B . C . AND FISHERIES EXPERIMENTAL STATION P R I N C E R U P E R T , B . C . N o . 13 R O S E , C O W A N &. L A T T A L I M I T E D PRINCE RUPERT, B .C. , JULY, 1932 These progress reports are issued from time to time to acquaint the fishing industry with some aspects of investigations which are being under-taken by the Biological Board of Canada through its Pacific Coast Stations. CONTENTS THE SALTING OF HERRING. THE FOOD OF THE KAMLOOPS TROUT. THE VITAMIN D POTENCY OF OILS FROM B.C. CANNED SALMON. DO FISH REACT TO NOISE ? THE SALTING OF HERRING, by Neal M. Carter, Pacific Bioligical Station, Nanaimo, B.C. The preservation of fish by means of salt or brine is a process which dates back to prehistoric times and because of the simplicity and almost unfailing efficacy of the method little thought was given to the principles underlying the preservative action. The commercial competition of modern times has necessitated more careful control of salting processes and the small margin of profit which prevails in the fish salting industry has led the packers to request the aid of science in an effort to ascertain how a minimum expenditure of time, materials and. labour may be consistently related to a satisfactory product. Particularly is this true in the case of the herring salting indus-try of British Columbia, which supplies a large proportion of the Orient's requirements. Salted "round" herring sell at a very low price and any hastening of the pickling process which will satisfactorily allow a greater production per unit of equipment should tend to increase the small mar-gin between profit and loss. Within the past twelve years a considerable amount of laboratory research'has been undertaken in order to determine the efficiency of both dry salting and brining as applied to various species of fishes which are usually cleaned and split, but little information is available for the pro-cess as applied to whole, "round" herring. For this reason, a preliminary investigation of the salting of such herring was undertaken at this Station. Before presenting the results obtained, a brief description of the agencies which contribute toward the spoilage of freshly caught fish and the effect of salt in hindering or preventing their action is necessary. The vital processes of a living fish continually combat the tendency of natural agencies to bring about dissolution of the living tissues; a dead fish, on the other hand, is merely a collection of chemical compounds which immediately commence to undergo decomposition according to de-finite, well-established physical and chemical laws. This decomposition is known to the fish industry as spoilage and is a result of three processes normally taking place in the folowing order : (1) Autolysis, (2) Putre-faction (bacterial decomposition), (3) Oxidation. Autolysis ("self digestion") is a process which commences immed-iately after the death of the fish and is due to the action of chemical ferments known as enzymes, which are always present in animal tissues and are particularly abundant in the blood and viscera. During life, these enzymes perform many necessary, carefully controlled functions such as digestion, etc., but after death their unrestrained action leads to the break-ing down of tissues, often accompanied by the production of objection-able odors, and the fish is described as "sour." Partial liquefaction of the flesh might eventually occur were it not for the fact that the more rapid and destructive process of putrefaction usually masks the advanced stages of autolysis. Putrefaction results from infection by the bacteria which are everywhere present on the equipment used in handling fish from the time 3 of capture until they reach the consumer. It is not the bacteria them-selves hut the chemical enzymes produced as a by-product of their activity which cause this putrefaction. Bacterial growth and activity, already rapid at 45° F., increase greatly with rising temperature, producing poison-ous compounds which render the flesh quite unfit for human consump-tion. Oxidation is a process which takes place too slowly to be observed before bacterial decomposition sets in; its effects are noticeable as a ran-cidity of. the fat of salted or lightly smoked fish which have been long ex-posed to the atmosphere, particularly in the presence of sunlight. Enzymes, whether autolytic or putrefactive in action, act best in dilute solutions. Therefore, any treatment Avhich will lessen the water content of animal tissue will retard or altogether prevent such spoilage, depending upon the completeness of the dehydration (removal of water). The function of the salt and brine in the process of salting is to remove the water from the cells which compose the tissues of the fish; they have little if any disinfecting power and merely suspend the activity of existing and bacterially-produced enzymes. Almost any non-poisonous chemical substance fairly soluble in water might be used in place of salt for this purpose, salt being used only because of its cheapness and.the fact that we are accustomed to its taste. Any existing bacteria are not necessarily killed during the salting process, since many forms can assume a resting (spore) state in which they are very resistent to drying. Oxidation is not prevented by salting. On the contrary, it is greatly accelerated by certain impurities such as iron and copper compounds which may exist in the salt or become introduced during the process. The statement was made above that retardation of spoilage depends upon the completeness of the dehydration of the fish. The effect of brines of various strengths in bringing about this retardation may be briefly summarized as follows. Percentage Degrees as tti/iKnirPn C\Y\ RETARDATION O F : of salt in briiie salinomcter Autolysis Bacterial decomp. Oxidation 4 10 20 25 16 40 80 100 slight considerable great very great nil very slight considerable very great nil nil nil nil The mechanism of the action of salt in effecting its preservative action on fish is attributed to two physical phenomena known as osmosis and diffusion. Osmosis.—When two solutions of different concentration, or one solution and water are separated by any partition (such as parchment, skin of an egg, or even glass with fine cracks in it) which allows the pas-sage of water but hinders the passage of the dissolved substance, it is a well-recognized law that water will pass into the more concentrated solu-tion and equalize the concentration on both sides of the partition. This process is called osmosis. The innumerable cells making up the tissues I of the fish contain a dilute salt solution of protein material. The cell wall represents the partition which allows the passage of water. When salt is applied to the exterior of the fish, it dissolves in the moisture clinging to the exterior parts and thus forms the concentrated solution on the other side of the cell-wall partition. Osmosis at once commences and water is withdrawn from the cells of the fish to dilute the surrounding concentrated salt solution (brine or pickle). The rapidity of this action depends upon the difference in concentration between the brine and the solution remain-ing in the cells. Diffusion.—When two different solutions, or one solution and water alone are placed in liquid contact (no partition), the two will very grad-ually mix with another. Tins process is called diffusion. As previously slated, the cell walls of the fish tend to hinder the passage of a dissolved substance; the brine nevertheless gradually diffuses into the tissues sim-ultaneously with the flow of water from the cells due to osmosis. The rate of diffusion is considerably influenced by certain impurities in the salt used in making up the brine; calcium and magnesium salts in parti-cular retard diffusion and should never be present in the dry salt in amounts exceeding one per cent. Osmosis usually takes place far more rapidly than diffusion and it will be seen that in order to. obtain the most rapid and complete with-drawal of water, it is essential to keep the fish in full contact with brine which is as concentrated as possible, that is, saturated with salt. Osmosis ceases when, as a result of the brine being diluted by the water taken from the fish, the concentration of the brine becomes equal to the concentration of the solution left in the cells. It must be emphasised that a fish which has been allowed to lie in an understrength brine during the early stages of pickling can not be brought up to standard quality by completing the period of pickling in stronger or saturated brine. The weak brine does not cause sufficient osmosis and the simultaneous diffusion fills the tissues with a salt solution which is too weak to preserve the fish, but sufficiently strong to greatly hinder the completion of osmosis and diffusion by the stronger brine applied later. When a salted fish is soaked in fresh water, osmosis and diffusion take place in reverse directions; water enters the fish and the salt "soaks out." It is obvious that the exposure of packed, fully cured fish to rain or even excessively moist air will bring about the same effect, with a deleter-ious result on the keeping quality. The salting of round herring presents two difficulties not encoun-tered in the salting of cleaned and split fish. (1) Because of the retention of the viscera and blood, the fish must be very fresh and the salting must be performed as rapidly and completely as possible in order to check the action of the enzymes which are particularly abundant in the viscera and blood. (2) The whole fish does not give the brine access to the inner flesh as readily as when the fish is split open. In an effort to follow the course of combined osmosis and diffusion under these conditions, the following preliminary experiments were per-formed on a small sample of moderately lean herring caught near Nan-aimo in early February. Two dozen fish were scaled, marked for identification, wiped dry with a towel, weighed and divided into six groups of four each within a few hours of capture. Each lot of four fish was immediately placed in.a brine bath of definite salt concentration which had been made up from weighed quantities of dried, commercial salt and tap water. The following concentrations were chosen: (1) Salt content 10 per cent, by weight; salinometer reading, 40°. (2) Salt content 15 per cent, by weight; salinometer reading, 60°. (3) Salt content 20 per cent, by weight; salinometer reading, 80°. (4) Salt content 22.5 per cent, by weight; salinometer reading, 90°. (5) Sat'd (26.5%) sol'n in contact with excess solid salt; 102°. (6) Salt content increasing from 21 to 25 per cent, during six days The fish were left in the brines for eleven days at a temperature varying between 45° F. and 50°F. The hydrometer reading of the first four brines was kept constant by frequent stirring of the bath and fortifi-cation with saturated salt solution when necessary throughout the entire period. The fifth brine needed occasional stirring only; the concentration of the sixth brine was at first quickly, and later more gradually, increased. After fifteen hours of pickling and at the end of each following 24-liour period, the fish were taken out, wiped and weighed in an exactly comparable manner. The loss (or gain) of weight at the end of each period was computed, as a percentage of the original weight of the fresh fish, and after averaging the closely agreeing values for the four fish in each type of brine, the following chart was constructed. Fig. 1. DAYS OF PICKLING. -Chart showing percentage loss (or gain) of original fresh weight of herring pickled in brine of different concentrations. The chart shows very clearly the greatly increased efficiency of strong brines in effecting the removal of water from the herring. This effi-ciency apparently increases very rapidly as the salt concentration ap-proaches that of saturation, although the crossing" of the curves for the 80° and 90° brines was unexpected and requires further investigation. From the foregoing discussion of osmosis and diffusion, it will be realized that these two effects combine in contributing to a change of weight in the fish. The rapid loss of water through osmosis is partially balanced by the slow inward diffusion of the brine, resulting in a sudden decrease in weight during the first thirty—six hours followed by a gradual increase which had not ceased at the end of the legal pickling period (six days prior to December first, five days thereafter). Subsequent weighings showed that the weaker brines required three weeks for complete diffusion, whereas no further change took place after eleven days in the case of the saturated pickle. At the end of a month the fish in the three weakest brines had decomposed and those in the 90° brine had a somewhat ob-jectionable odour. This experiment does not allow a separate estimation of the two combined effects of osmosis and diffusion, but a proposed series of deter-minations of the moisture and salt content of herring pickled in possibly 9Q,° 94°, 98° and 102° brines for varying periods of time will allow the plotting of curves showing each effect separately. At first sight it would oppear from the shape of the curves in the above chart that the diffusion of salt into the fish was less in those which were pickled in the saturated brine than in those which, pickled in a weaker brine, gained weight rapidly during the latter days of the experiment. This is not the case, as was shown by a salt analysis made on two of the lots of fish, those in the 90° and saturated brines. The salt content was 25 and 33 per cent, respectively, based on dry weight of residue after heads, tails, fins and viscera had been removed. It is essential that any brine left in the pickling tank from the pre-vious salting should give a salinometer reading of at least 95° before the next batch of fresh herring and salt are introduced. The new brine formed from the fresh fish and salt must be sufficient in quantity to cover all the fish by the end of the first twenty-four hours; any fish remaining above the surface of the brine, even though covered by a layer of solid salt, will not cure properly since the original small quantity of brine produced from the moisture on the skin drains down into the underlying layers of fish as fast as it is formed. Daily tramping down of the surface layer of fish with its adhering salt, and a thin sprinkling of additional salt at the surface is an advisable procedure for ensuring uniformity of pickling throughout the tank. 7 THE FOOD OF THE KAMLOOPS TROUT, by C. McC. Mottley and Jean C. Mottley, Pacific Biological Station, Nanaimo, B.C. It is well known that the feeding habits of trout is one of the most important factors affecting angling. In Progress Report No. 11 an account ,was given of the general effect of food supply on the growth of the Kam-loops trout. The present report, on the other hand, proposes to show the feeding habits of these fish based upon an examination of the stomach contents. The food organisms of trout may be classified according to habitat as follows : (1) on or near the surface, (2) free-swimming in the water, or (3) associated with the bottom. Terrestrial insects, such as ants, and adults of insects having aquatic larval stages, such as dragon flies and caddis flies, comprise the surface food. Fish, emerging insects, water fleas, copepods and immature forms of certain insects compose the free-swimming organisms. Leeches, fresh-water shrimps, snails and immature insects make up the food associated with the bottom. It is difficult to classify certain forms on this basis, as for example, aquatic beetles and bugs, because they live, near the shoreline where the three habitats merge. However, the identification of the organisms in a stomach affords a pic-ture of the habitat in which a fish has been feeding. Observations on the food are thus useful in determining the best type of tackle to employ in angling. The Food of Trout at Paul Lake. The stomach contents of two hundred and fifty Kamloops trovit caught at Paul Lake during May, June, July, August and September, 1931, were examined. The different kinds of organisms were identified, their volumes measured and individuals counted. The accompanying table, shows the average number of the different organisms per stomach and the greatest number found in any one stomach. .Water fleas, fresh-water shrimps, immature dragon flies, caddis flies and two-winged flies are the most important in number. TABLE I. . Greatest num-Averaee b e r f o u n d i n Systematic Grouping Common Names number per a n y o n e s t o m . stomach Hirudinea Leeches 2 4 Cladocera Water Fleas 1,036 2,800 Copepoda Copepods 4 12 Amphipoda Fresh-water Shrimps 167 1,873 Aquatic Insecta Aquatic insects 57 586 Plecoptera Stone flies 2 2 Ephemerida May flies 12 57 Odonata Dragon flies 11 115 Anisoptera True dragon flies 3 16 Zygoptera Blue dragon or Damsel flies 13 108 Hemiptera Bugs 1 2 Coleoptera Beetles 7 59 Trichoptera Caddis or Sedge flies 17 186 Diptera Two-winged flies 100 585 Terrestrial Insecta Terrestrial- insects 20 220 Hymenoptera Flying ants 33 220 Coleoptera Beetles 3 11 Miscellaneous Terrestrial insects 3 20 Mollusca Snails 13 75 8 — Fresh-water Shr imps . — A q u a t i c Insects. Dragon Flies. Caddis Flies. Two-winaed Flies Miscellaneous. — Terrestrial Insects. Ants beetles. FKscaUaTieoas-— Snails. — Water Fleas. — Leeches. — Copepoda. Fig. 1—Diagram illustrating importance of various food items in stomachs of Kamloops trout. / S T R E A M T Y P E . F R Y . Aquatic Insects. Copepoda. Water Fleas. 1 0 0 % L A K E . T Y P E F R Y . A.I. Terrestr ia l Insects. YEARLING T R O U T Fresh-water .Shrimps. Agnatic Insects. T e r r e s t r i a l Insects. L.WF. A D U L T T R O U T . Fresh-water Shr imps 4 0 % Aquatic Insects. T.I. S. 1SS7. 59% 32 7o Fig. 2.—Diagram illustrating percentage volume of each kind of food in each of four groups of individuals of Kamloops trout. L.—Leeches; W.F.—Water Fleas; A.I.—Aquatic Insects;-T.I.—Terrestrial Insects; S.—Snails. June. August. Sept Katj. June. ^uty- August. Sept Figure 3—Diagrams showing relations of fly fishing and trolling to character of food consumed. A—Percentages of occurrences of insects compared with fresh water shrimps in food in year 1931; B—Percentages of eighty record fish caught by fly fishing and twenty-seven record fish caught by trolling for past ten years, assigned to the various months. The length of the rectangles in figure 1 represents the percentage of stomachs in which the organism occurred; the height represents the percentage frequency of occurrence of each type. This figure then, gives a graphic picture of the importance of occurrence of the organisms found in the diet of adult trout at Paul Lake. Figure 2 shows graphically the percentage volume comprised by each kind of food for four groups of fish. The first bar shows that the stream type fry appear to take aquatic insects exclusively. These fry were taken from Paul Creek in September. Their average length was 1.4 inches. Two-winged flies formed ninety per cent, of the volume, with im-mature stone flies and beetles making up the remaining ten per cent. The second bar indicates that the lake type fry live on a more varied diet. Free-swimming forms, such as copepods and water fleas which are part of the plankton of the lake, made up sixty-eight per cent, of the volume. The remainder consists of aquatic and terrestrial insects. The lake type fry were taken from Paul Lake in September. Their average length was 3.2 inches, more than twice the average length of the stream type fry, and the average volume of the stomach contents was twenty-seven times that of the stream type. The lake apparently offers more food for the fry and this condition appears to promote more rapid growth. The variety of food may also be a factor in the increased growth rate. The third bar 10 MisceUa-neoua. Figure 4—Percentage of eighty record fish of past ten years taken on each of the listed flies. shows the importance of fresh-water shrimps in the diet of yearling trout al Paul Lake. Aquatic and terrestrial insects constitute the remainder of the diet. The yearlings are eight to ten inches long by the end of the summer, at which stage they are readily caught by anglers. The fourth bar shows that the percentage volume of shrimps increases in the diet of older trout. During the winter months shrimps are the staple food item and therefore appear to be very important for the maintenance and growth of the fish the year.round. Because emerging and mature aquatic insects and terrestrial insects attract fish to the surface, where they may be taken by fly-fishing, special note should be taken of their percentage volumes. The diet of the adults is even more varied than that of the yearlings. It is an interesting fact that up to the present time no small trout have been found in the stomachs of fish caught at Paul Lake. Kam-loops trout is the only species of fish occurring in the lake. The Relation Between Food and Fishing. It is well known that sport fishing depends on locating a feeding fish and presenting to it a natural or artificial lure which the fish may take as food. The types of food present in different seasons determine the fish-ing. Fly fishing at Paul Lake predominates in the spring and fall while trolling is the more effective method during July and August. Figure 3A shows the percentage of occurrences of aquatic and terrestrial insects com-pared with fresh-water shrimps during the fishing season. Figure 3B shows the percentages of eighty record fish caught by fly fishing and twenty-seven record fish caught by trolling for the last ten years at Paul Lake. It can be readily seen from the curves that fly fishing varies direct-ly with the occurrence of insects and that trolling becomes effective as in-sects decrease and the fish feed largely upon shrimps. Figure 4 gives the percentages of the eighty record fish taken by the listed flies. 11 The majority of these flies are caddis representations. The Black Gnat was used chiefly in May and September. It imitates insects such as flying ants, which have fallen on the surface of the water. We are much indebted to Mr. J. A. Scott for permission to use the data regarding the record fish. The Food of the Trout at Penask Lake A sample of twenty-five trout was obtained from Penask Lake. The tabulation below gives a comparison of the percentage volumes' of the main food items with those of the adult fish from Paul Lake. Fresh-water Aquatic Terrestrial Average Average Shrimps Insects Insects Volumes Length Paul Lake 59% 32% 5% 5.8cc. 16 in.-Penask Lake 42% 42% 14% 0.7cc. 11 in. The fish from Penask Lake consume a proportionately smaller volume of shrimps and a greater volume of insects than the fish from Paul Lake. This may be a factor affecting the growth and feeding habits of the fish and is probably the reason why fly fishing is so successful at Penask Lake. The difference in the average volume of the stomach contents may indicate a comparative scarcity of food at Penask Lake and thus account for the difference in the average length of the fish. Adult Trout Tend to Become Piscivorous. In Paul Lake the food possibilities of trout are limited owing to the absence of other species of fish on which they might feed. In bodies of water where other species of fish do occur it is found that fish comprise a considerable proportion of the food of adult trout. Four Kamloops trout, each weighing over three pounds, were caught at Little River in May, 1931. Ninety spring salmon fry were found in one stomach and thirty-four in another. The other two stomachs each contained between thirty and forty sockeye fry. These stomachs also had aquatic and terrestrial insects in them. The stomach contents of two hundred Kam-loops trout caught at Kootenay Lake during June and July, 1929, were examined. Eighteen of these fish, ranging in length from 13 to 28 inches, had been feeding on Kokanees and one had eaten a young trout. Fishery Guardian Forfar reports that in 1929-30-31 he made observations on the food of two hundred Rainbow trout, of the Upper Fraser River watershed, weighing from 4 to 15% pounds. Kokanees were found in the stomachs of all these fish and a young trout occurred in one stomach. While it has been shown that Kamloops trout turn to a fish diet as they grow larger, few cases of their eating the young of their own kind have been found. In "barren" lakes such as Paul, Knouff and Hyas no evidence of cannibalism has been shown. Due to the fact that the older fish seek greater depths during the summer, they are not apt to be asso-ciated with the fry and a certain amount of safety is provided for the young. 12 THE VITAMIN D POTENCY OF OILS FROM B.C. CANNED SALMON by B. E. Bailey Fisheries Experimental Station, Prince Rupert. The plans for a preliminary investigation of the nutritive value of B.C. canned salmon were outlined in Progress Report No. 11. It was slated there that an important part of the investigation was to consist in a study of the vitamin potencies of the oils. The purpose of the vitamin work was (1) to establish the general potencies of the oils, and (2) to show if there were any large variations in the potencies of the B.C. pack. The determinations of vitamin D have now been completed and the results are briefly summarized in this paper. Canned sockeye and pink salmon, each from three localities, were chosen for this preliminary study. The samples obtained did not re-present the entire season's production in the various districts, since, when the work was started (Sept. 1931), the canning season was practically over and samples had to be chosen at random. One sample case was taken from the pack of each of several canneries, and composite samples of oil made up from these. In view of the possible sampling error, there-fore, it would be inadvisable to draw definite conclusions regarding the difference in vitamin D content of canned salmon from those districts in which the pack was sampled. Vitamin D Value of Canned Salmon \ • Kind of Salmon Locality Caught ' Vitamin D. Potency (Relative to Cod Liver Oil 100) Pink Butedale 67 Pink Johnstone Straits &8 Pink Fraser River 88 Sockeye Skeena River 67 Sockeye River's Inlet 88 Sockeye Fraser River 88 As shown in the table, the average vitamin D content of the salmon oils examined in this investigation was high. Compared with a medicinal cod liver oil, the vitamin D content of which was arbitrarily called 100, the salmon oils showed an average potency of about 80. These findings show the importance of canned salmon as a source of this vitamin. The variation in the potencies of the several samples is not of any great significance. Although final conclusions cannot be drawn from ex-periments on so few samples it would appear the oils from pink and sock-eye salmon are uniformly high in vitamin D h-respective of the locality in which they are packed. The oil content of the canned sockeye salmon is somewhat greater than that of the pink. Thus, although the oils are of approximately equal potency, the total amount of vitamin D per can is greater in the sockeye than in the pink. 13 DO FISH REACT TO NOISE? by V. H. K. Moorhouse, University of Manitoba, Winnipeg. It is generally believed that noises of any kind (but especially those made by m§or boats and outboard engines) should beavoidecwhen fish-ing Because of certain complaints in this connection, which were re cefved at the Pacific Biological Station, a preliminary investigation was initiated In the experiments described herein, there was used an aquarium approximately 4 x 3 x 3 feet in size with glass sides and provided with a constant supply of well-aerated water. The noises were produced by an electrically-driven tapper, a bell, a buzzer and a motor horn, respectively inside a rather large rectangular can at one end of the aquarium. The tapper made the gentlest and the motor horn the loudest, roughest and most abrupt sound. That they were all definitely audible to the human ear below the surface of the water was shown by the use of a submerged stethoscope. ' At the outset it was found that rock cod, flounders, sculpins and dog-fish gave no consistent sign of being affected. Some of these fish seemingly became restless but never definitely avoided the can where the noise was most intense. On the other hand, small perch, of which forty to fifty were placed in the aquarium, at first avoided the noise, group-ing distinctly at the opposite end of the tank. Arery soon, however, even these perch became accustomed to the new environment. In spite of this, it would be entirely erroneous to conclude that these fish are un-affected because, as demonstrated later, when the proper conditions are present, they respond to noise quickly and strongly. Since close observation had shown that the perch rise actively to the surface of the water for food, it was decided1 to combine the noises with feeding, in an effort to determine whether they would eventually l-ise in response to the noise only. Accordingly, when the apparatus was started, a small amount of food was spread on the surface from a tip bucket hanging over the aquar-ium, toward the opposite end to the origin of the noise. Care was taken to arrange to carry out all manipulations from a screened position about twelve feet distant, so that the nearness of the observer would not com-plicate matters. The fish were kept under close observation at all times and photographs were made of the reactions shown. . It was found that it was only necessary to repeat this experiment eight to ten times, when the perch began to group at the surface immed-iately on starting the noise; The reaction increased in strength until after .a considerable number of repetitions, the sound alone brought most of the fish rapidly to the surface. Photographs of this conditioned or "trained" response are identical with those of an actual feeding response. Such experiments indicate the mistake of. concluding that fish are 14 insensible to. noise at all times because they happen to show no sign of reaction under certain conditions. When these are altered by making the noise a signal for food, a very strong response to sound is readily demon-strated. Sound vibrations in water act much more strongly upon some species of fish than upon others, as was apparent from the preliminary experiments in this series. The fact that fish become rapidly accustomed lo-noise makes it improbable that their behaviour will be affected except in a temporary fashion. This work does not support the popular belief that noises are very repellant, at least to the species used in the investiga-tion. It demonstrates that the nervous system of the perch is quite cap-able of building up a conditioned reflex to sound. NOTE. Through an oversight acknowledgment to the owners of the copy-right photograph of Princess Louisa Inlet, reproduced in Progress Report No. 1 2 was omitted. Permission to use the photograph was given by the Gowan Sutton Co. Ltd., whose courtesy is hereby gratefully acknowledged. 15 NEWS ITEMS During the current year recaptures are expected from the following Pacific salmon marking experiments : (1) 180,000 pink salmon fry from McClinton Creek, Massetl Inlet, marked by the removal of the adipose fin, (2) 123,550 sockeye salmon fingerlings from Eagle River, a I Tafl, marked by the removal of the left pelvic and the adipose fins, (3) 100,000 sockeye salmon yearlings from Cullus Lake, marked by the removal of both pelvic fins. Returns are expected to yield exceedingly important data in respect to efficiency of transplantation, age at maturity, percentage of return from the sea, return to parent stream, etc. All engaged in the fishing industry are asked to co-operate by watching closely for marked fish and by re-turning the scars and the information requested on the posters which have been widely distributed. The booklet, "The Trout and Other Game Fishes of British Colum-bia" by Professor J. R. Dymond, is now available. It consists of 50 printed pages with 7 excellent coloured and 2 black and white illustrations. The booklet is published by the Department of Fisheries, Ottawa, and a charge of $1.00 per copy is made because of the cost of reproducing the coloured plates. Any person desiring a copy should write to the Department of Fisheries, Ottawa, and enclose a dollar. Mr. D. B. Finn has resumed the Directorship of the Fisheries Experimental Station (Pacific) having returned from The Low Tempera-ture Laboratories in Cambridge, England, where he undertook a study of Freezing and Its Relation to Physical and Chemical Changes in Animal Tissues. The work, which has a direct bearing on "drip" and the proper temperature for cold storage, is being applied to fish. Professor R. A. Wardle, of the University of Manitoba, is continuing his fish parasite studies at the Pacific Biological Station', Nanaimo. In a recent publication Professor Wardle points out that the fish of the Cana-dian Pacific Coast are singularly free from tapeworm infection. Dr. J. L. Hart has commenced a search for the eggs and larvae of the pilchard off the west coast of Vancouver Island. He reports that the young fish now occurring in large schools in the sounds^ are yearling fish, that is, they are just entering their second year. 16 The Nutritive Value of Marine Products VI. The Vitamin A Potency of Salmon Liver Oil B Y B . E . BAILEY Fisheries Experimental Station (Pacific) Contributions to Canadian Biology and Fisheries, Vol. VIII, No. 21 (Seriei C, Industrial, No. 14). TORONTO 1 9 3 4 No. 2 1 (SERIES C, INDUSTRIAL, NO. 14) T H E N U T R I T I V E V A L U E O F M A R I N E P R O D U C T S VI . T H E V I T A M I N A P O T E N C Y O F S A L M O N L I V E R O I L BY B. E. BAILEY Fisheries Experimental Station (Pacific) (Received far publication October 18, 1933) A B S T R A C T The vitamin A content of liver oils of five species of salmon in the genus Oncorhynchus has been determined. The richest samples contained 40 times, and the poorest twice as much vitamin A per gram as a sample of cod liver oil stated to contain 500 A . D . M . A . units per gram. Potencies of samples were in descending order: Skeena spring, Vancouver spring and sockeye, coho, pink and chum salmon. O i l content of the livers averaged 5 per cent and percentage of liver in the fish 2 per cent. The melting point, tmsaponifiable matter and iodine value of the oils have been determined. The Nutritive Value of Marine Products VI. The Vitamin A Potency of Salmon Liver Oil BY B. E. BAILEY Fisheries Experimental Station (Pacific) During recent years considerable attention has been directed to the study of fish liver oils other than cod liver oil, as sources of vitamin A . Numerous liver oils have been examined and some have been found considerably richer in vitamin A than cod liver oil. No comprehensive study of the vitamin A potency of salmon liver oil and no work whatever on British Columbia salmon liver oil has yet been reported. Schmidt-Nielsen (1929) and Ahmad and Drummond (1930) have both reported finding relatively large amounts of vitamin A in samples of salmon-liver oil. The work here reported was undertaken with a view to investi-gating the variations in vitamin A potency and chemical characteristics of the liver oils from different species of salmon-caught in British Columbia waters. The Pacific salmon (genus Oiicorhyncliiis) is caught in the coastal waters and streams of the Pacific coast from California to the Bering sea. There are five species belonging to this genus, the spring salmon, Oncorhynchus tscha-wytscha; the sockeye, Oncorhynchus vcrka; the coho, Oncorhynchus kisntcli; the humpback salmon, Oncorhynchus gorbuscha; and the chum, or dog salmon, Oncorhynchus beta. Although one representative of the genus Salmo, the steel-head, Sahno gairdncri, is also caught in Pacific coast waters, the genus Onco-rhynchus is by far the more important. In British Columbia the salmon fishery is by far the most important from the standpoint of both catch and value. The landings for the years 1930, 1931 and 1932 were respectively 2,296,231 cwt. (104,155 metric'tons), 1,287,041 cwt. (58,379 metric tons), and 1,166,671 cwt. (52,919 metric tons). Since very little of the offal from this fishery is now utilized, the value of a study of salmon liver oil as a source of vitamin A is evident. S A M P L I N G In selecting samples for an investigation of this type several factors must be considered. The three main variables are species, geographical distribution, and season of catching. 'Of these, the first two are most important. Seasonal varia-tion is less significant, since the commercial fishing season for salmon Is relatively short. For this investigation livers were obtained from fish landed in northern, central and southern British Columbia, several species being represented from each of these districts. Samples were taken at different times throughout the fishing season to give some indication of the seasonal variation. When the livers had been removed from the fish they were frozen and shipped to Prince Rupert, where they were stored at — 2 0 ° C. until the oil was 267 4 extracted. Details of the samples and data regarding the percentage liver in the fish, and oil in the livers are given in table I. T A B L E I Sample no. Species District of catching Date landed Per cent, livers in fish Per cent, oil in livers (by analysis) 1 2 3 Sockeye Sockeye Sockeye Skeena river Butedale Vancouver July 19/32 July 25/32 August 1932 1.52 1.78 - 4.9 3.4 5.0 4 5 6 Coho Coho Coho Butedale Butedale Vancouver July 25/32 August 27/32 August 1932 1.75 3.3 3.7 3.9 7 8 • > 9 10 11 Springs Redsprings Whitesprings Springs Springs Skeena river Skeena river Skeena river Vancouver Vancouver May 2/32 July 5/33 • July 5/33 August 1932 April 26/32 4.7 4.6 3.9 5.5 • 12 ' 13 14 15 Pinks Pinks Pinks Pinks Skeena river Butedale Butedale Vancouver September 10/32 July 25/32 August 27/32 August 1932 1.85 2.5 3.3 4.3 16 17 18 Chum Chum Chum Butedale Butedale Vancouver July 25/32 August 27/32 August 1932 2.00 3.7 3.7 3.5' 19 Steelhead Skeena river September 10/32 • P R E P A R A T I O N O F O I L S The liver oils were prepared by extracting the cooked livers with petroleum ether, and dehydrating the solution with anhydrous sodium sulphate. The solvent was subsequently removed by distillation in vacuo over a hot water bath. The oil samples were placed in small glass vials, under inert gas. When in use they were stored in a refrigerator at 5° C. and at other times in cold room at —20° C. P H Y S I C A L A N D C H E M I C A L C H A R A C T E R I S T I C S Several of the oils were solid at room temperature and for this reason it was considered advisable to determine the melting points. Some difficulty, however, was experienced in carrying out the measurements as the melting points were not sharp, the oil softening gradually over a range of several degrees. In order to get some information regarding the state of the samples at room temperature, the bottles were held at a constant temperature of 15° C. and their condition noted when equilibrium had been reached. Those remaining solid at this tem-perature were then held at 20° C , and their state of equilibrium again observed. 268 These data, together with percentage unsaponifiable matter (ethyl ether extract) and iodine value (Wijs) are given in table II. T A B L E I I Sample Species District of Unsaponifiable Iodine State at State at no. catching matter value 15° C. 20° C. , 1 Sockeye Skeena river 2 Sockeye Butedale 194.0 Solid Semi-solid 3 Sockeye Vancouver 5.4 182.0 Liquid . Liquid 4 Coho Butedale 200.8 Liquid Liquid 5 Coho Butedale 196.2 Liquid Liquid 6 Coho Vancouver 7.3 209.2 Liquid Liquid 7 Springs Skeena river Liquid Liquid 8 Redsprings Skeena river Liquid Liquid 9 Whitesprings Skeena river Liquid Liquid 10 Springs Vancouver 5.3 182.0 Liquid Liquid 11 Springs Vancouver Liquid Liquid 12 Pinks Skeena river 13 Pinks Butedale 232.4 Solid Solid 14 Pinks Butedale 248.0 Solid Solid 15 Pinks Vancouver 6.0 220.2 Liquid Liquid 1G Chum Butedale 203.9 Solid Semi-solid 17 Chum Butedale 219.1 Solid Solid 18 Chum Vancouver 7.0 198.2 Solid Solid 19 Steelhead Skeena river V I T A M I N A S S A Y S In comparing the vitamin A potency of a large number of oils, the colori-metric test of Carr and Price (1926) is very convenient providing the blue values of the various samples bear a constant relationship to the vitamin A potency. Both Schmidt-Nielsen (1929) and Ahmad and Drummond (1930) obtained satisfactory results when the colorimetric test was applied to salmon liver oil. The biological method is at present the standard test for vitamin A . Before the colorimetric test can be used as a measure of the vitamin A potency of an oil parallel tests must be made on several samples by both methods. From the data thus obtained a conversion factor can be calculated, relating the vitamin A potency to the colorimetric blue value of an oil. In order to establish the constancy of the conversion factor for the oils under examination typical samples were assayed biologically and the values thus obtained related to the blue colour developed by the antimony trichloride test. The samples chosen, chum (sample 18), sockeye (sample 3) and spring salmon (sample 10), were fed in graded doses to rats which had ceased to grow on a vitamin A free diet. The growth response over an 8-week period was 2 6 9 compared to that produced during the same period in similar rats when they were fed 2 mg. daily of a cod liver oil chosen as a standard, which contained 500 A . D . M . A . units of vitamin A per gm. The diets and general technique of the test were those outlined by the Vitamin Assay Committee of the American Drug Manu-facturers' Association with the exception that the rats were placed on test between the 35th and 50th days after starting the vitamin A-free diet instead of between the 30th and 45th day. The liver oils together with the standardized oil were dissolved in Wesson oil in such a way that 20 mg. of the solutions contained the required dose. These were fed to the rats directly by means of a specially calibrated pipette. The data thus obtained are set forth in table III , from which it can be seen that 0.7 mg. of chum liver oil, 0.15 mg. of sockeye liver oil and 0.15 mg. of spring liver oil produce a growth rate which closely approximates that produced by 2 mg. of standard cod liver oil which contained 1 A . D . M . A . unit of vitamin A . The vitamin A potency of these oils expressed in the above unit is therefore chum 1428, sockeye 6667 and spring 6667 units per gm. The technique of the colorimetric test of these and the remainder of the samples was method B of Drummond and Hilditch (1930). The blue colours were measured in a Rosenheim-Schuster tintometer. Determinations were made with various concentrations of oil so that values on both sides of 5 blue units were obtained. These data were plotted and the amount of oil producing 5 blue units was determined by intrapolation. From this the number of blue units equivalent to 1 mg. of oil when measured at this level was calculated. Sample 18 produced 4.2 blue units per mg. of oil while by the biological assay it was found to contain 1428 A . D . M . A . units of vitamin A per gm. Each blue unit is therefore equivalent to 340 units of vitamin A per gm. Compared. T A B L E I I I • " District N o - R a t s A v - g a i n Sample Species . o { Level No. Rats s u r v j v j n g per week no. catchin- m g - test period for 8 week 3 Sockeye Vancouver 0.05 0.10 0.15 5 5 5 4 5 5 - 1 . 6 4.9 6.7 10 Springs Vancouver 0.05 0.10 0.15 5 5 5 5 5 5 0.6 3.0 6.9 18 Chum Vancouver 0.40 0.60 0.70 5 5 5 5 5 5 1.8 5.2 8.5 Positive control C .L .O. 2.00 7 7 6.7 Negative control 7 0 270 7 in the same way, the other two samples (3 and 10) were found to have con-version factors of 345 and 333 units of vitamin A per gm. of oil for each blue unit developed when the blue value is measured at the level of 5 blue units. The average conversion factor can therefore be taken as 340, which value was used to calculate the vitamin A potencies of the other samples tested. These values are shown in table I V . T A B L E I V Sample no. Species District of catching B.U. per mg. oil Vitamin A units per gm. 1 Sockeye Skeena river 28.5 9,690 2 Sockeye Butedale 25.0 8,500 3 Sockeye Vancouver 19.3 6,667 4 Coho Butedale 12.5 4,250 5 Coho Butedale 10.0 3,400 6 Coho Vancouver 17.0 5,780 7 Springs Skeena river 40.0 13,600 8 Redsprings Skeena river 60.3 20,502 9 Whitesprings Skeena river 55.2 18,768 10 Springs Vancouver 20.0 6,667 11 Springs Vancouver 20.0 6,800 12 Pinks Skeena river 7.8 2,652 13 Pinks Butedale 4.0 1,360 14 Pinks Butedale 4.2 1,428 15 Pinks Vancouver 5.2 1,768 16 Chum Butedale 2.5 850 17 Chum Butedale 3.8 1,292 18 Chum Vancouver 4.2 1,428 19 Steelhead Skeena river 20.0 6,800 D I S C U S S I O N The data presented show clearly that salmon liver oil is a rich source of vitamin A . The richest samples were those prepared from the livers of spring salmon caught in the Skeena river area. The Vancouver springs and the sockeye samples were next in potency while the coho and pink samples were least potent. The least potent samples examined contained twice as much and the most potent over 40 times as much vitamin A per gm. as did the cod liver oil used for a standard of comparison. According to the definition laid down by the Vitamin Assay Committee of the American Drug Manufacturers' Association (1931) a unit of vitamin A is that amount which will produce an increase of weight of at least 15 gms. in 35 days in rats that have ceased to grow on a vitamin A-free diet. This (1932) 271 has recently been amended to an increase of at least 12 gms. in 28 days. Obviously this definition, depending as it does upon the response of the albino rat, must be limited by the variation in this response which is found between different colonies. It is well known that similar diets, when fed to different colonies of rats, will produce different growth rates depending on the strain of rat used and the conditions of environment. Thus there will be a variation in the true value of the American Drug Manufacturers' Association unit depending on the,place of assay. This difficulty could have been avoided had the vitamin A unit been defined by the growth response produced by a certain quantity of growth-promoting substance such as the special carotene prepared under the direction of the Health Organization of the League of Nations, but unfortunately this material was not available during the course of the work. In this work the test period used was 8 weeks, instead of the 5-week period specified in the 1931 A . D . M . A . technique, since it had been found in these laboratories that, during the earlier part of the test period, individual variations in growth rate of animals on the same level were rather large. After 8 weeks the degree of uniformity in response was considerably greater, so the test was continued for that length of time. The term "Uni t of Vitamin A " is defined here as being that amount of growth-promoting substance which, when administered daily, produces an average growth rate of between 6 and 7 gms. per week in the albino rats of this colony over a period of 8 weeks, after the animals have ceased to grow on a vitamin A -free diet. The terms of this definition were chosen because they were satisfied by the response of positive control rats in this test when they were fed 1 unit doses of cod liver oil that had been assayed by other laboratories, using the A . D . M . A . technique, particularly those of the large pharmaceutical houses. The results of this assay therefore bear a direct relation to their terms of evaluation. The amount of oil in all livers was very low, and did not compare favourably with the oil content of livers from other fish. Halibut livers, for example, have been found in these laboratories to contain from 15 to 25 per cent oil, while the maximum for salmon livers was approximately 5 per cent. The distinct parallelism between the vitamin A potency of the liver oils and the red colour of the body oil in the different species, with the exception of spring salmon, is very striking. While the liver oils of the salmon are all dark coloured, the body oils vary greatly in pigmentation. That of the sockeye is bright red; of the coho reddish, though somewhat paler than the sockeye; and of the pinks, or pink salmon as they are frequently called, orange pink. The body oil of the chum has only a slight yellowish pigmentation. That of the spring salmon, on the other hand, varies from white to bright red. A s already stated, the pigmentation of the body oil in chum, pinks, coho and sockeye increases in the order given. The colours of average samples of body oils from these four species of salmon were found to be:— 272 9 O i l Colours—red yel low ( L o v i b o n d units) C h u m salmon body 2.1 4.6 P i n k salmon body 4.8 24.4 Coho salmon body 7.2 29.0 Sockeye salmon body 14.3 27.7 F r o m the data in table I I I it w i l l be seen that the v i tamin A potency of the l iver oils f rom each district, wi th the exception of spr ing salmon, increases in the same order as the red colour given above. Since spring salmon vary in body pigmentation from colourless (white) to deep red, they are not considered in this connection. V e r y little is yet k n o w n regarding the nature and function of the body pigments in salmon. Newbigan (1898) studied the interchange of pigments between muscle and ovaries of salmon dur ing the spawning cycle. N o work , however, has been reported regarding the actual metabolism of the pigments. T h e above data suggest that there may be a relationship between the amount of body pigment present, and the v i tamin A potency of the l iver. E u l e r , E u l e r and He l l s t rom (1928) , M o o r e (1929) and others have shown that the pigment carotene can' replace vi tamin A in the diet. Olcott and M c C a n n (1931) showed the existence of an enzyme, which they called carotinase, in the livers of rats, by means of which carotene could be converted to vi tamin A in vitro. If carotene was one of the pigments in salmon body o i l , it would be quite plausible to postulate an interchange of body pigment and vi tamin A in the l iver of the fish dur ing l i fe . Pre l iminary tests carried out in these laboratories have indicated, however, that the body oils f rom sockeye salmon appear to contain little or no carotene. It has also been found that deeply pigmented salmon body oils have very little v i tamin A activity (Ba i ley , M S . ) , which indicates that caro-tene is not present in appreciable quantities. A further study of the body oi l pigments is now in progress. A l t h o u g h this paper is pr imar i ly concerned wi th the v i tamin A potency of the oils, some interesting facts have been disclosed by the data of the chemical analyses. I n the first place, the iodine values of a l l the oils are rather high, ranging f rom 182.0 to 248.0. Secondly, it is remarkable that the highest degree of unsaturation is shown by the oils of highest melt ing point, Butedale p ink salmon l iver oils (samples N o . 13 and N o . 14), both of which were solid at 2 0 ° C . T h e degree of unsaturation does not appear to have any relation to the amount of unsaponifiable matter present. A C K N O W L E D G E M E N T The author wishes to thank M r . H . R . Beard , Chie f Chemist of the Canadian F i s h i n g Co . L t d . , Vancouver , B . C . , for collection of livers, and his assistant, M r . J. R . Townsend, for collection of l ivers and data regarding the percentage of livers f rom Butedale. T o the Director , M r . D . B . F i n n , and to M r . H . N . Brocklesby of the Fisheries Exper imenta l Station (Pac i f i c ) , he expresses appre-273 1 0 ciation for the advice and assistance given dur ing the research, and to M r . J . B . F i n n for the care of the experimental animals. R E F E R E N C E S A H M A D , B . , AND J. G . DRUMMOND. T h e relative v i t amin A value of body and l iver oils of certain fish. Biochem. J. 24, 27. 1930. CARR, F . IT. AND E . A . PRICE. C o l o u r reaction attributed to v i tamin A . Biochem. J. 20, 497. ' 1926. DRUMMOND, J . C , AND T . P . HILDITCIT. T h e relative values of cod-l iver oils f rom various sources. IT. M . Stationery Office, L o n d o n . 1930. EULER, C . V., IT. v. EULER, AND H . ITELLSTROM. V i t a m i n A action of l ipochrome. Biochem. Z. 203, 370. H A A S , P . , AND T . G . H I L L . Chemistry of plant products. V o l . 1, 4th ed., L o n g -mans, Green & Co . 1928. HEILBRON, I. M . R . A . MORTON AND E . T . WEBSTER. Structure of v i tamin A . Biochem. J. 26, 1194. ' 1933. MOORE, T . Carot in and v i tamin A . Lancet 1, 499. 1929. NEWBIGAN, M . I. Pigments of the muscle and ovary of the salmon and their exchanges. In D . N o e l Pa ton , Report of Investigations on the l i fe history of the salmon. Fish. Bd. Scot. A r t i c l e X V . 1898. OLCOTT, IT. S. AND D . C . M C C A N N . Carotenase. Trans format ion of carotene into v i tamin A in vitro. J. Biol. Chem. 94, 185. 1931. SCHMIDT-NIELSEN, S., AND S. SCHMIDT-NIELSEN. A difference between the v i tamin content of the flesh oi l and the l iver oi l of the salmon. K. norske Vidensk. Sclsk. Fork. 1, 189. 1929.. V I T A M I N ASSAY COMMITTEE OF T H E A M E R I C A N D R U G MANUFACTURERS' A s s o c . A n n u a l report, 1931. / . Amer. Phnrm. Ass. 20, 588. 1931.' 274 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0061972/manifest

Comment

Related Items