wnrEftsify !!!* 
 
 siSRA^y 
 th 
 
 of 
 
 Lrr\(» to 
 
 4 
 
 J 
 
 <P 
 
 0 
 
 /7 
 
 S' 
 
 _S 
 
 [Reprinted from the Journal of the American Chemical Society. Vol. XXVII. 
 No. 3. March, 1905.] 
 
 ON THE PRESENCE OF COTTON-SEED OIL IN LARDS 
 FROiT HOGS FED UPON COTTON=SEED iTEAL. 
 
 By A. D. Emmett and H. S. Grindley. 
 
 Received December 21, 1904. 
 
 Having had occasion to make for the Department of Animal 
 Husbandry, of the Experiment Station of the University of 
 Illinois, quantitative determinations of cotton-seed oil in lards 
 rendered from the fat of hogs fed largely upon cotton-seed meal, 
 a more detailed study of the samples was undertaken in order to 
 add, if possible, some information as to whether cotton-seed oil 
 actually existed in the body fat of the animals. 
 
 That many of the vegetable oils enter into the make-up of the 
 body fat in some manner or other, there can be but little doubt. 
 Lebedeff , 1 and Henriques and Hansen 2 in feeding linseed cake, 
 Monk 3 in feeding rape oil and the fatty acids from mutton tallow, 
 Lebedeff 4 and Rosenfelt 5 in feeding mutton fat; and Shutt 6 in 
 feeding maize, have shown that the resulting fat differs mate- 
 
 1 Ztschr. physiol. Chem. ,‘ 6 , 149. 
 
 2 Thier. Chem. Ber., 29, 68. 
 
 3 Ibid., 14, 411. 
 
 4 Ztschr. Physiol. Chem., 6 , 149. 
 
 5 Thier. Chem. Ber., 25, 44. 
 
 6 Canada: Cen. Expt. Sta. Bull., 38. 
 
 ^1.5by 1 
 
264 
 
 A. D. EMMETT AND H. S. GRINDLEY. 
 
 rially from the normal fat body in appearance and properties — 
 melting-point and iodine number. 
 
 On the other hand, Radziejewsky 1 with rape seed oil, Subbotin 2 
 with soap, Tebedeff 3 with tributrine, Henriques and Hansen 4 
 with linseed oil on cows, Bomer, 5 Harrington and Adriance, 6 
 Virchow, 7 Van Engelen, 8 Solstein, 9 Langfiirth, 10 and Fulmer 11 with 
 cotton-seed meal, were unable to obtain affirmative results in 
 their feeding experiments, although in almost every instance 
 there were some abnormal changes detected in the fat. In the 
 cases of cotton-seed meal some of the investigators found, by the 
 Halphen and Bechi colorimetric tests, as high as 15 to 30 per 
 cent, of the oil. 
 
 Thus, admitting that the presence of vegetable oils influences 
 the fat of animals to some extent, the question arises as to how 
 these oils exist in the body fat. Are they assimilated unchanged 
 through the lacteals and thus find their way into the circulatory 
 system, being carried on to the fat cells? Are they first broken 
 up into simpler bodies and then absorbed by the lacteals, or are 
 they subjected to the intricate processes of metabolism — saponi- 
 fied by the bile, etc. — and converted into animal fat? 
 
 The fact that Monk, in feeding rape oil to dogs, not only found 
 the olein content of the body fat much above the normal, but 
 that he was able to separate eruic acid, a characteristic ingre- 
 dient, tends to show that in one case, at least, the vegetable oil 
 was evidently absorbed unchanged. On the other hand, since 
 Henriques and Hansen found, in their work on the feeding of 
 linseed oil to pigs and dogs, that the same ration may affect the 
 fat of one animal and not that of another; and since Lebedeff 
 in feeding the same animal on different fats — linseed oil and 
 tributrine — was able to detect a change in the former case but 
 none in the latter, it tends to indicate that the nature of the 
 assimilation of the rations varies under certain conditions — with 
 the animal and kind of animal and with the fat and kind of fat. 
 
 1 Virchow’s Archiv., 56, 211; 43, 268. 
 
 2 Ztschr. f Biol., 6, 73. 
 
 3 Ztschr. Physiol. Chem., 6, 149. 
 
 4 Thier. Chem. Ber ., 39, 68. 
 
 6 Ztschr. Unt. Nahr. u. Genus., 1899, 2. 
 
 6 Texas Expt. Sta. Bull., 29. 
 
 7 Ztschr. Unt. Nahr. u. Genus. (1899), 8, 46-48. 
 
 8 Chem. Centrbl., 1899, 2. 
 
 9 Zeit. offentl. Chem., 8, 140. 
 
 10 Ztschr. angew. Chem., 14, 685. 
 
 11 This Journal, 24, 1148; 26, 837. 
 
COTTON-SEED OIL IN LARDS. 
 
 265 
 
 To answer this question of assimilation in the case of cotton- 
 seed meal-fed hogs, there seemed but one way as suggested by 
 Hehner 1 in 1880, when European markets were being flooded with 
 adulterated American lards, namely, to actually establish the 
 presence of a vegetable oil. At that time, the solution resolved 
 itself, in the case of cotton-seed oil, into the following statement : 
 If a lard gave the Bechi-Maumene test together with a high iodine 
 value and a low melting-point, it should be condemned. But in 
 cases of doubt, Salkowski’s 2 method for the separation of vegetable 
 cholesterin should be applied. Eater, many other colorimetric 
 tests were submitted, chief among them being the nitric acid and 
 the Halphen tests. Of the former, Eewkowitsch 3 stated, previous 
 to the appearance of the latter, that he preferred it to all others. 
 Halphen’s test, however, has come into prominence and is con- 
 sidered to-day to be the most convenient and reliable method, 
 both from the qualitative and the quantitative standpoint, for 
 estimating cotton-seed oil, and by many it has been given the 
 preference to the once official Bechi test. 
 
 In dealing with lards from hogs fed upon cotton-seed meal, 
 the factors are to some extent different from those in the case of 
 rendered lards mixed with cotton-seed oil. The analytical num- 
 bers — iodine absorption value, and melting-point — are not trust- 
 worthy, although indicative. Bomer 4 has shown that the varia- 
 tions in the iodine number make this test impracticable. Albert 5 
 and likewise Henriques and Hansen 6 have found that the compo- 
 sition of the fat from different parts of the same animal is subject 
 ' to great differences in melting-point and iodine number. In the 
 case of Albert, these variations were greater, on the average, 
 than could be ascribed to any feeding ration. Hence, in this 
 short study, we have directed our attention chiefly to other tests 
 for determining qualitatively the presence of a vegetable oil. 
 
 One of the supposed essential differences between animal and 
 vegetable oils is in their monatomic alcohols called cholesterins. 
 The former, animal cholesterin, crystallizes from hot alcohol in 
 rhombic plates, and the latter, vegetable cholesterin and desig- 
 
 1 Analyst , 13, 165. 
 
 2 Ztschr. a?ial. Chem., 26, 557. 
 
 3 “Oils, Fats and Waxes,” Benedickt and Lewkowitsch. 
 
 4 Ztschr . Unter. Nahr. u. Genus., 1898, 532. 
 
 5 Landw. Jahrb., 28, 961. 
 
 6 Bied. Centr. Blatt. Agr. Chem., 30, 182. 
 
266 
 
 A. D. EMMETT AND H. S. GRINDLEY. 
 
 nated by Hesse 1 as phytosterin, crystallizes in needles arranged 
 in stars, fans or bundles. According to Bomer, 2 in a mixture of 
 the two cholesterins, the phytosterin either takes on the form 
 of elongated plates which may or may not have angles at the 
 ends, or if it is in much the greater proportion, of needles which 
 lie upon the rhombic plates of the animal cholesterin. Van 
 Kettel 3 states that another distinguishing feature of some veg- 
 etable oils, and especially cotton-seed oil, is that they give the 
 pentosan reaction. In this connection Rudzinski 4 has recently 
 shown that the pentosans occur in company with the carbohy- 
 drates in the plant cells and are closely connected with cellulose, 
 a prominent constituent of cotton-seed meal. Slovtzov 5 also 
 states that a part at least of pentosans is absorbed by the animal 
 body. In 1891 Welman 6 proposed his phosphomolybdic test for 
 vegetable oils, suggesting that the reaction might be due to gluco- 
 sides. In 1901 7 he defended the attacks upon his test and added 
 that the final blue coloration which would also result if amines 
 were present, was of little value, as the characteristic feature 
 was the green color produced when the, nitric acid was added. 
 Seiler and Verda 8 have lately stated that Welman’s reaction, 
 though not a positive test, gives good indications of the presence 
 of vegetable oils in admixtures of animal fats or mineral oils. 
 
 Again, as suggested above, Bomer worked upon the separation 
 or detection of a vegetable oil in lards from cotton-seed meal-fed 
 hogs. His conclusions were that inasmuch as he could not obtain 
 crystals of phytosterin, the vegetable oil was not retained in the 
 animal fat. He also stated that since a trace of the fatty acids' 
 lowers the melting-point of the crystals, the differences in their 
 form should be relied upon rather than that of the melting-point. 
 Virchow, who also worked upon this question, obtained results 
 agreeing with Bomer as to the separation of phytosterin and the 
 presence of a vegetable oil in the body fat. Solstein in his work 
 was likewise unsuccessful in isolating the vegetable cholesterin and 
 also in obtaining positive results with the Welman reaction. He 
 
 1 Ann . Chem. (niebig), 192, 175. 
 
 2 Mitt. d. Kgl. Techn. Versuck., 18, 255. 
 
 8 Monat. Scieniifique , 1900. 
 
 4 Ztschr. physiol. Chem 40, 317. 
 
 5 Ztschr. Unter. d. Nahr. u. Genuss., 1904, p. 2. 
 
 6 Pharm. Ztg., 36, 798. 
 
 7 Ztschr. f. offentl. Chem.. 6 , 143 ; 4, 852. 
 
 8 Chem. Ztg.. 27. 
 
COTTON-SEED OIL IN LARDS. 
 
 267 
 
 stated that a lard, although it might give the Halphen test, should 
 not be condemned unless it gave Welman’s reaction and crystals 
 of phytosterin. 
 
 It is noteworthy, at this point, that where the Halphen, Bechi, 
 and nitric acid colorimetric tests hkve been applied to lards from 
 cotton-seed meal-fed hogs, these reactions have been obtained 
 almost without fail. The majority of investigators agree that 
 each of these characteristic reactions for cotton-seed oil is de- 
 pendent upon a specific constituent in the oil for its ultimate 
 color. If this is the case, then the animal fat has apparently 
 assimilated these three bodies in their unaltered condition, either 
 through the medium of the unchanged cotton-seed oil or its 
 simpler decomposition products, for, should they have been sub- 
 jected to the regular processes of metabolism, it is difficult to con- 
 ceive that they could have been transmitted to the fat cells 
 without one of them having lost its characteristic property. 
 
 In our work, samples of leaf and ham fat were obtained from 
 each of four hogs which had been fed largely upon cotton-seed 
 meal. The fats were rendered in the laboratory at a low tempera- 
 ture, not exceeding 120° C. A sample of leaf fat, known to be 
 free from cotton-seed oil, was also rendered at the same time 
 and under the same conditions. Quantitative determinations of 
 cotton-seed oil were then made on each of the eight samples of 
 lard by the Halphen and Bechi-Hehner tests. The amount of 
 cotton-seed oil was found to range from 5 to 15 per cent. The 
 Bechi-Hehner method gave somewhat higher results throughout. 
 
 The first four samples of the above lards were then taken for 
 the more detailed study as to the presence of the vegetable oil. 
 
 In applying the nitric acid tests to these lards, the character- 
 istic brown color appeared at once and was very pronounced. A 
 blank test upon the pure lard gave no color, even after standing 
 twenty-four hours. 
 
 As suggested by Van Kettel, Tollen’s 1 phloroglucinol hydro- 
 chloride reaction for pentosans was tried. With each of the sam- 
 ples a distinct reddish pink ring was noticeable at the juncture 
 of the two layers. A parallel determination upon the pure sample 
 gave no such results, showing that pentosans were present in the 
 lards in question. 
 
 Welman’s phosphomolybdic test was carefully tried. The 
 
 1 Jour.f Landw., 40, 13. 
 
268 
 
 A. D. EMMETT AND H. S. GRINDLEY. 
 
 characteristic green and blue colorations were obtained repeatedly 
 when applied to the samples in question, but in no case could they 
 be produced with the pure lard. 
 
 Having been successful in these characteristic tests, attention 
 was then directed to the separation of the vegetable cholesterin, 
 phytosterin, believing, with Salkowski, Hehner, Bonier, Solstien 
 and others, that should this unsaponifiable substance be found, 
 the evidence would be still stronger in favor of the presence of a 
 vegetable oil in the lards. 
 
 Three attempts were made to separate the phytosterin. First, 
 Von Raumer’s 1 method was found to yield a substance which 
 agreed fairly well with the description of phytosterin, but the 
 quantity obtained was so small that no positive results could be 
 procured. Forster and Riechelman’s 2 method was then applied. 
 The yield was somewhat larger than in the previous case and the 
 same characteristic features concerning the crystalline formation 
 were noticeable. In this instance, a parallel determination was 
 made upon a sample of cotton-seed oil. There was a similarity 
 in their crystalline structure, but we were still dissatisfied with 
 the results. In a somewhat recent article by Ritter, 3 he has shown 
 that his method for the separation of cholesterin from fats will 
 give a larger yield than any of the others. We were led to try 
 this procedure in our case. Briefly stated, it is as follows : Fifty 
 grams of the fat are heated in a porcelain dish upon a water- 
 bath with ioo cc. of alcohol. This solution is saponified by the 
 addition of a hot mixture of 8 grams of sodium hydroxide 
 and 160 cc. of strong alcohol. It is then evaporated and 
 75 grams of sodium chloride are added and the whole is 
 dissolved in hot water. The solution is brought to dryness with 
 continued stirring and the residue pulverized, and left over sul- 
 phuric acid for some time. The dried mass is then extracted 
 with ether for at least nine hours, using a Soxhlet apparatus. 
 The extract is freed from ether at a low temperature and the 
 residue of cholesterin is finally taken up with hot alcohol and 
 allowed to cool slowly, when crystals will separate. 
 
 Two hundred grams of the lard in question and 50 grams of 
 pure cotton-seed oil were carefully treated as just stated. As a 
 result, we obtained a much larger yield than in either of the other 
 
 1 J, Soc. Chem. Ind 1898, p. 774. 
 
 2 Analyst , 1897, p. 131. 
 
 3 Ztschr. physiol. Chem., 34, 461. 
 
COTTON-SEED OIL IN LARDS. 
 
 269 
 
 two methods, although the amount obtained in this case was 
 still small. The crystals from the evaporation of the ether were 
 needles. After treatment with hot alcohol and allowing the crys- 
 tallization to continue slowly, needles again separated. In the case 
 of the cotton-seed oil, they were arranged in stars, and in the case 
 of the lard, in leaf-like forms. Upon recrystallization, the stars 
 retained their same general shape, but the leaf-like forms changed 
 into bundles of needles. These bundles crossed each other at 
 different angles, forming clusters. A second recrystallization of 
 these bundles gave crystals of a distinct fan shape with a star 
 of needles as a base and having the fascicular portion built up 
 from this nucleus. By recrystallizing these again, they passed into 
 the previous form — bundles of needles. Salkowski’s chloroform- 
 sulphuric acid test for cholesterin was applied to this small quan- 
 tity of crystals. The test proved the presence of a cholesterin 
 and the crystalline form indicated that it was probably phytosterin. 
 
 However, as a further confirmation of the presence of phyto- 
 sterin, Bomer ascertains the melting-point of the acetic esters of 
 these crystals. He states 1 that this method is a very sensitive one 
 for detecting the kind of cholesterin. If the melting-point of the 
 crystals is 116 0 C. or higher, it shows phytosterin to be present. 
 This method has been used by Tolman 2 in attempting to identify 
 phytosterin in lard from cotton-seed fed hogs, in which he ob- 
 tained negative results. Juckenack and Pasternack 3 agree with 
 Bomer in regard to the acetate test for phytosterol, but 
 they do not make any statement as to its delicacy. On the 
 other hand, Siegfeld 4 states that while the test is accurate, it is 
 not sensitive in the cases of the fats with which they worked — 
 margarin and palmitin — to less than 10 per cent. Bomer claims 
 that cotton-seed and sesame oils, which are frequently present 
 to considerable extent in margarin, can be detected by this test 
 when present to the extent of only 1 to 2 per cent. It is thus 
 apparent that there is still a difference of opinion as to the delicacy 
 of this test. In our work, as a result of repeated experiments, we 
 always obtained crystals which closely resembled those of phyto- 
 sterin but as yet we have not been able to prove conclusively the 
 presence of this substance by the melting-point of its acetate. 
 
 1 Ztschr. Unter. Nahr. u. Genus., 4, 1901 ; 5 , 1902 . 
 
 2 Science , 1904 , 763 . 
 
 3 Ztschr. Unter. Nahr. u. Genus., 7 , 1904 . 
 
 4 Ibid.., 7 , 1904 . 
 
270 
 
 COTTON-SEED OIL IN LARDS. 
 
 This brief study, which was originally undertaken in order to con- 
 firm the Halphen and Bechi tests for cotton-seed oil in lards from 
 hogs fed upon cotton-seed meal, has been found to present so 
 many subjects of interest upon which as yet very little has been 
 published that we are now making still further investigations upon 
 the problems involved. 
 
 conclusions. 
 
 These several samples of lard rendered from the fat of cotton- 
 seed meal-fed hogs, were found to give the following tests for 
 vegetable oils: (I) Welman’s reaction; (II) Tollen’s pentosan re- 
 action; (III) Salkowski’s cholesterin reaction ; and (IV) crystals re- 
 sembling phytosterin. The samples also gave the following tests for 
 cotton-seed oil : (I) the Bechi ; (II) the nitric acid ; and (III) the 
 
 Halphen reactions. From this it is evident that two statements 
 can be made, first that the lards contain a vegetable oil, and 
 second, if we agree with the most recent authorities, they contain 
 three distinct constituents of cotton-seed oil. Hence, it seems 
 safe to say that a part at least, of the oil existing in cotton-seed 
 meal, is absorbed, in the case of hogs fed upon this ration, by 
 the animal body and transmitted in its unaltered condition 
 to the fat cells.