I 11 " ' ■ ■ llliil lH MBSm SSBSsSSjShKSK ■:••■■'.-.■■■ ifiiiiiiiiii w m ■Hi ■ ■■■■'*■■>■■ •■■*.-'', { ■'■;'. ■ '^- •■;. ■ : ■ :•;:■:::■•-■■':' ■ • BWIi^WiiMtiiiiV'ttfflrW nfflllKHIIIIIWWBffBff 8§£1 'V- " . • ;■;;;-■: .-. V .-'■ ;. li 3Sg&*gg£ THE PROTEINS OF THE WHEAT KERNEL BY THOMAS B. OSBORNE WASHINGTON, D. C. : Published by the Carnegie Institution of Washington 1907 CARNEGIE INSTITUTION OF WASHINGTON Publication no. 84 ''i^ 3 PRESS OF JUDD & DETWEILER, INC. WASHINGTON, D. C. CONTENTS. Page. Introduction 5 Review of the Literature 6 Experimental Part. 17 The aqueous extract of wheat flour iS The albumin of wheat flour — leucosin 20 The aqueous extract of the wheat embryo 22 Hydrolysis of leucosin 30 The proteins of wheat flour soluble in sodium-chloride solution 37 The globulin of wheat flour 37 The proteins of the wheat embryo soluble in sodium-chloride solution 40 The globulin of the wheat embryo 40 The proteose of wheat flour 45 The proteose of the wheat embryo 46 The proportion of the various protein substances of the wheat embryo 48 Digestion of the phosphorus-containing protein preparations with pepsin- hydrochloric acid 49 Protein soluble in dilute alcohol — gliadin 53 Direct extraction with dilute alcohol 53 Extraction with dilute alcohol after extracting the flour with 10 per cent sodium-chloride solution 58 Extraction of gluten with dilute alcohol 59 Extraction of " shorts " with dilute alcohol 60 Protein extracted by alcohol from whole-wheat flour 62 The proportion of glutaminic acid yielded by various fractions of the alcohol- soluble protein 64 Hydrolysis of gliadin 70 Protein insoluble in water, saline solutions, and alcohol — glutenin 82 Protein extracted by dilute alkaline solutions after extracting the flour with 10 per cent sodium-chloride brine and then with dilute alcohol. ... 82 Protein extracted by dilute alkaline solutions from gluten after extracting the latter with dilute alcohol 84 Protein extracted by dilute alkaline solutions after complete extraction with dilute alcohol of gluten from whole-wheat flour 88 Hydrolysis of glutenin 91 The amount of the various proteins contained in the kernel of wheat 100 The formation of gluten 103 Summary 108 Gliadin 109 Glutenin 112 Leucosin 113 The globulin 115 The proteose 116 The gluten 116 The nutritive value of the wheat proteins 117 3 THE PROTEINS OF THE WHEAT KERNEL. INTRODUCTION. Of the protein substances used as food none is of more importance than those contained in the seeds of wheat. Although these bodies attracted the attention of investigators more than one hundred years ago and have since then been many times the subject of study, the published statements respecting them are so conflicting and uncertain that it has heretofore been impossible to know what the truth regarding them actually is. With the purpose of clearing up the existing confusion and determining the real value of the evidence offered, as well as extending, as far as possible, our knowledge of these important substances, the writer some years ago under- took an investigation of this seed which has recently been concluded by work done under grants from the Carnegie Institution of Washington. As the results of these investigations have been published from time to time in a number of different papers, appearing in four different journals, it has been thought desirable to bring all this work together in one paper. In so doing the details have been reproduced in full, for the nature of the evidence is such that its value largely consists in concordant results of many experi- ments, repeated under different conditions, since it is not yet possible to establish the chemical individuality of different protein substances by demon- strating their possession of definite physical properties, as may be done with the simpler organic compounds. The experience of the writer in his endeavors to understand and repeat the work of many of his predecessors has made him feel the importance of these details to future workers along the same lines and is his excuse for giving with so much minuteness the results of his own work, which to those not familiar with the difficulties of the subject must appear to a large extent unnecessary. In order to make this work available to those who wish sim- ply to know the results, a comprehensive summary of this paper is given at the end of this publication, and the details of the many operations and experi- ments need be read only by those who wish to become familiar with these. The account of this work is preceded by a review of the literature of the subject, from which an idea of the unsatisfactory state of our previous 5 6 THE PROTEINS OF THE WHEAT KERNEL. knowledge can be obtained. This review is interesting, also, as it shows the slow development of the study of vegetable proteins and how the several investigators have been influenced by the knowledge of the animal proteins prevailing at the time the work was done. In carrying out his investiga- tions of these proteins the writer has received the assistance of Messrs. Voorhees, Campbell, Harris, and Clapp, for which he wishes here to make acknowledgment ; but especially is he indebted to Prof. S. W. Johnson, under whose direction and with whose advice and encouragement this work was first undertaken in the laboratory of the Connecticut Agricultural Experiment Station, where it has since been continued. REVIEW OF THE LITERATURE. The fact that gluten can be obtained from wheat flour by washing with wTater appears to have been first published by Beccari.1 That alcohol extracts a protein substance from wheat flour was first stated by Einhof,2 who considered this to be the same as the gluten. Taddei3 found that gluten consists of two substances, one of wdiich is soluble in alcohol, which he named " gliadin," the other insoluble in alcohol, which he named " zymom." De Saussure4 obtained from wheat gluten about 72 per cent of plant- albumin in the insoluble form, about 20 per cent of plant-gelatin, or, as he proposed to call it, "glutin," and about 1 per cent of mucin, which latter substance he considered to be similar to the mucin described by Berzelius. Berzelius5 thought that the alcoholic extract contained another protein substance, which he called "mucin," and that the part of the gluten which was insoluble in alcohol was so similar to albumin that he called this 11 plant-albumin." Boussingault6 agreed with Einhof th.it the part of the gluten that was soluble in alcohol was the same as the entire gluten protein. 1 Beccari. Reference to this publication has, for many years, appeared in the literature as Comon. Bonon. I. 1, p. 122. It should be De Bononiensi Scientiarum et Artium Insti- tuto atque Academia Commentarii, 1745, 11, part 1, p. 122. In this paper Beccari refers to the fact that in 1728 he had orally communicated to the Academy the previously un- published fact that wheat flour can be separated into two parts, one of vegetable, the other of animal character. The substance of this communication was published in the above-cited paper in which the separation of gluten from wheat flour, by washing with water, is described. 2 Einhof, Journal der Chemie von Gehlen, 1S05, v, p. 131. 3 Taddei, Annals of Philosophy, 1820, xv, p. 390. 4De Saussure, Schweiger's Journal, 1S33, lxix, p. 188. 5 Berzelius, Lehrbuch der Chimie, Auflage 3, 1837, vi, p. 453. e Boussingault, Annales de Chimie et de Physique, 1838, l found, Cu 21.65 P- ct. The alcoholic solution of the copper salt of the /-proline was evaporated to dryness under reduced pressure. The dried residue weighed 8.24 grams, equivalent to 6.5 grams of /-proline. For identification a small portion was freed from copper with hydrogen sulphide, and the free proline converted into the phenylhydantoin, accord- ing to the directions of Emil Fischer.1 The hydantoin melted sharply at 1430 and gave the following analysis : Carbon and hydrogen : 0.2334 gm. substance gave 0.5676 gm. C02 and 0.1204 gm. H20. Nitrogen : 0.1373 gram substance gave NH3 = 1.78 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for C12H1202N2, C 66.60, H 5.61, N 12.99 p. ct. ; found, C 66.32, H 5.73, N 12.96 p. ct. J Fraction IV. Temperature of bath up to 2000. ( Pressure, 0.8 mm. Weight, 49.39 grams. From fraction IV the ester of phenylalanine was removed in the usual manner by shaking out with ether, and after freeing from ether the residual ester was saponified by dissolving in concentrated hydrochloric acid and evaporating on the water-bath. Of phenylalanine hydrochloride 12.12 grams were obtained, which is equivalent to 9.93 grams of phenylalanine. For identification the hydrochloride was recrystallized from strong hydro- chloric acid and converted into the free acid by evaporation with excess of ammonia. When once recrystallized from water, it melted at 2630 t0 265°. Carbon and hydrogen : 0.2962 gram substance, dried at no°, gave 0.7089 gram C02 and o. 180 1 gram H20. Nitrogen : 0.1924 gram substance gave NH3 = 1.65 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for C9Hn02N, C 65.39, H 6-73. N 8-5° P- ct- \ found, C 65.27, H 6.76, N 8.57 p. ct. ^iscber, E., Zeitschrift fur pbysiologische Cbetnie, 1901, xxxnr, p. 251. 3 34 THE PROTEINS OF THE WHEAT KERNEL. The aqueous layer was saponified by warming with an excess of barium hydroxide on the water-bath for 7 hours. After standing for some time the crystals of racemic barium aspartate were filtered out and decomposed with an equivalent quantity of sulphuric acid. The filtrate from the barium sulphate gave, on concentration, 5.06 grams of aspartic acid, which, when recrystallized from water, was analyzed. Carbon and hydrogen : 0.24S3 gram substance, dried at no0, gave 0.3278 gram CO, and 0.1 2 15 gram H20. Nitrogen : 0.2769 gram substance gave NH3 = 2.93 cc. HC1 (1 cc. HCl = o.oi gram N). Calculated for C4H.04N, C 36.06, H 5.31, N 10.55 P- ct. ; found, C 36.00, H 5.44, N 10.59 P- ct. The filtrate from barium aspartate was freed from barium quantitatively with sulphuric acid and the filtrate from the barium sulphate concentrated to small volume and saturated with hydrochloric acid gas. After long standing on ice, 2.15 grams of glutamiuic acid hydrochloride separated. The filtrate from this was evaporated under reduced pressure, the residue taken up in water, and the chlorine removed with silver sulphate. After removing the sulphuric acid with an equivalent quantity of barium hydroxide, the solution was boiled with an excess of copper hydroxide. The filtered solution, on standing, separated tyrosine-like needles of copper laevo- aspartate which weighed 7.41 grams, equivalent to 3.57 grams of aspartic acid. Nitrogen : 0.1853 gram substance, air-dried, gave NH3 = o.93 cc. HC1 (1 cc. HC1 = 0.01 gram N). Copper : 0.1677 gram substance, air-dried, gave 0.0491 gram CuO. Calculated for C4H504N, Cu ■ $% H20, N 5.09, Cu 23.06 p. ct.; found, N5.02, Cu 23.37 p. ct. From the copper salt the free acid was regenerated and analyzed. Carbon and hydrogen: 0.2041 gram substance gave 0.2715 gram C02 and 0.1054 gram H20. Calculated for C4H704N, C 36.06, H 5.31 p. ct.; found, C 36.29, H 5.74 p. ct. Specific rotation. — Dissolved in 20 per cent hydrochloric acid, (a) ^ = +23.8° Fischer & Dorpinghaus1 found / n 2°° 1 O («)-5-= + 22.0° An effort to isolate serine in the filtrate from the copper aspartate failed. Fischer & Dorpinghaus, Zeitschrift fur physiologische Chemie, 1902, xsxvi, p. 462. EXPERIMENTAL. 25 The Residue after Distillation. The residue remaining after distillation of the esters weighed 57 grams. This was dissolved in boiling alcohol, and after cooling 1.98 grams of needle crystals were filtered out. The filtrate was evaporated to a sirup under reduced pressure, saponified by heating with an excess of barium hydroxide, and, after removing the barium, evaporated to small volume under reduced pressure. The solution was then saturated with hydrochloric acid, and, after standing on ice for a long time, yielded 7.39 grams of glutaminic acid hydrochloride. The free acid prepared by evaporating with an exactly equivalent quantity of potassium hydroxide, when recrystallized from water, melted at 202 ° to 2030 with effervescence. Carbon and hydrogen : 0.3646 gram substance, dried at no°, gave 0.5432 gram C02 and 0.2019 gram H20. Nitrogen: 0.3696 gram substance gave NH3=3.55cc. HC1 (1 cc. HCl = o.oi gram N). Calculated for C3H;,04N, C 40.82, H 6.12, N 9.52 p. ct.; found, C 40.63, H 6.15, N 9.60 p. ct. The total glutaminic acid obtained from leucosin was 17.5 grams, or 6.73 per cent. This result is higher than that recently recorded in this labora- tory, namely, 5.72.1 This protein is one from which the glutaminic acid hydrochloride can be directly obtained only with great difficulty. In the former paper attention was directed to this fact, and the statement made that it is possible that the result given was too low. Tyrosine. Forty grams of leucosin, equal to 34.96 grams dried at no°, were boiled for 12 hours with a mixture of 120 grams of sulphuric acid and 240 grams of water. After removing the sulphuric acid with an equivalent quantity of barium hydroxide, the solution was evaporated with an excess of barium carbonate in order to remove ammonia. After removing the barium, the solution was concentrated to a small volume on the water-bath and allowed to stand for some time. The substance which separated was washed with cold water, dissolved in ammonia, the solution treated with bone-black, and evaporated. On cooling, 1.0360 grams tyrosine separated in colorless needles. The filtration from this, on further concentration, yielded 0.13 gram more tyrosine, making a total of 1. 1660 grams, or 3.33 per cent. This was recrys- talized and analyzed. Carbon and hydrogen : 0.4573 gram substance, dried at no0, gave 0.9994 gram C02 and 0.2813 gram H20. Calculated for C9Hu03N, C 59.62, H 6.13 p. ct. ; found, C 59.60, H 6. 11 p. ct. 1 Osborne & Gilbert, American Journal of Physiology, 1906, xv, p. 333. 3^ THE) PROTEINS OF THE WHEAT KERNEL. Arginine, Histidine, and Lysine. The filtrate from the tyrosine, by Kossel's method, yielded a solution in which the nitrogen found corresponded to 0.99 gram of histidine, or 2.83 per cent. The identity of the histidine could not be established owing to its very small amount. The solution of the arginine contained nitrogen equal to 0.6720 gram, which is equal to 2.08 grams of arginine, or 5.94 per cent. A part was con- verted into the nitrate and this latter into the copper salt by boiling with an excess of copper hydroxide and the copper salt recrystallized from water. Water: 0.19S4 gram substance, air-dried, lost 0.019 1 gram H20 at no0. Copper : 0.1766 gram substance, dried at no0, gave 0.0262 gram CuO. Calculated for C12H2804N8Cu (N03)2 ' 3 H20, H20 9.15 p. ct.; found, H20 9.62 p. ct. Calculated for C12H2804N8Cu (N03)2, Cu 11.85 P- ct.; found, Cu 11.83 P- ct. The filtrate from the copper salt was freed from copper and the solution evaporated with an excess of sulphuric acid under reduced pressure. The sulphuric acid was removed with an excess of barium hydroxide, and the barium with carbonic acid. The filtrate from the barium carbonate was evap- orated to dryness and the arginine converted into the picrolonate, according to the directions of Steudel. This melted at 2260 to 2270; Steudel gives 2250. Nitrogen: 0.0516 gram substance, dried at no0, gave 11. 8 cc. moist N3 at 250 and 765 mm. Calculated for C6Hu02N4 • C10H8O5N4, N 25.62 p. ct. ; found, N 25.71 p. ct. The filtrate from the silver precipitate which contained the arginine and histidine was freed from silver and barium, and the lysine precipitated with phosphotungstic acid and then converted into the picrate, of which 2.47 grams, equal to 0.9616 gram of lysine, or 2.75 per cent, was obtained. This was recrystallized from water and analyzed. Nitrogen: 0.2288 gram substance gave 38.6 cc. moist N2 at 25. 50 and 759 mm. Calculated for C6Hu02N2 ■ C6H307N3, N 1S.70 p. ct. ; found, N 18.77 p. ct. The results of this hydrolysis are given in table 5. Table 5. — Leucosin. p. ct. Glycocoll 0.94 Alanine 4.45 Aminovalerianic acid 0.18 Leucine n-34 a-proline 3. 18 Phenylalanine 3-83 Aspartic acid 3.35 Glutaminic acid 6.73 P. ct. Tyrosine 3.34 Lysine 2 75 Histidine 2.83 Arginine 5.94 Ammonia 1.41 Tryptophane present 5032 EXPERIMENTAL. 37 PROTEINS OF WHEAT FLOUR SOLUBLE IN SODIUM-CHLORIDE SOLUTION. The Globulin of Wheat Flour. Beside the proteins soluble in water, 10 per cent sodium-chloride brine extracts from ground wheat kernels a globulin which is present in the seed in small quantity. Ten kilograms of ' ' straight flour ' ' were extracted with 34 liters of 10 per cent sodium-chloride solution by suspending the flour in the liquid, stirring frequently, and then allowing the whole to stand at rest over night. The extract, separated from the flour and filtered as clear as possible, had a very slight acid reaction, was of a pink color, very viscid consistence, and formed about one-half of the solution added to the flour. This was saturated with ammonium sulphate, and the resulting precipitate filtered off and dissolved as far as possible in 4 liters of 10 per cent sodium- chloride solution. The exceedingly viscid solution was filtered with diffi- culty, placed in a dialyzer, and left in a stream of running water until the chlorides were removed. As the salts dialyzed out the globulin gradually separated in minute particles, the larger of these being evidently spheroidal in form. This precipitate weighed 5.8 grams. This protein, dissolved in a 10 per cent sodium-chloride solution, when heated slowly, gave a very slight turbidity at 870, which increased slightly up to 990. On boiling, some coagulum developed, and on adding acid to the solution filtered from this coagulum a very considerable precipitate formed. Dilution of the solution of the globulin in 10 per cent sodium-chloride brine precipitated the protein. Saturation with sodium chloride gave no precipitate. Saturation with magnesium sulphate, and also with ammonium sulphate, completely precipitated the globulin. When dried at no°, this preparation, 23, gave, when analyzed, the results which are shown in the table on page 38. Another preparation, 24, was made in the same way as the preceding, ex- cept that the precipitation with ammonium sulphate was omitted, the filtered extract being placed at once in dialyzers. Like the preceding solution, this was at first very viscid, but after the removal of the chloride the viscid property was entirely lost. This viscidity can hardly be due to the globulin, for solutions of the precipitated globulin showed no trace of it. The aqueous extract of the flour had no such property, and it is difficult to say to what this was due unless to the presence of carbohydrate. After complete removal of the chlorides the solution was filtered from the precipitate, the latter dissolved in 10 per cent sodium-chloride solution, and the insoluble matter filtered off. The residue so removed consisted chiefly of an insoluble form of the globulin. This was dissolved in o. 2 per cent potassium-hydroxide water, the solution filtered clear, and precipitated by neutralization with 38 THE PROTEINS OF THE WHEAT KERNEL. 0.2 per cent hydrochloric acid. The precipitate weighed 2 grams. The solution of the globulin was dialyzed till free from chlorides and the sepa- rated globulin found to weigh 5 grams. The yield in this extraction was therefore 7 grams, or nearly the same as in the preceding. The composi- tion of preparation 24 is given by the analysis shown in the following table : Preparations 23 and 24. Preparation 23. Preparation 24. I. II. Average. Ash- free. I. II. Average. Ash- free. Hydrogen Oxygen P.ct. 50.87 6.74 18.13 0.97 P.ct. 50.79 6.67 18.25 P.ct. 50.83 6.71 18.19 o.97 P.ct. 51-07 6-75 18.27 0.97 22.94 P. Ct. 5o.77 7-03 18.43 0.71 P.ct. 50-63 6.S4 18.32 P. ct. 50.70 6-93 18.38 0.71 P.ct. 51 01 6.97 18.48 0.71 22.83 Ash 0.48 0.62 100.00 100.00 The globulin contained in the " shorts" from the spring wheat was next extracted by treating 2000 grams with 10 per cent sodium-chloride solution for 3 hours, with frequent stirring, and then squeezing out the extract in a screw-press. After the suspended starch had settled, the extract was decanted and saturated with ammonium sulphate. The precipitate was dis- solved in 10 per cent sodium-chloride brine, and the solution filtered from the insoluble matter. The clear extract was then dialyzed till free from chlorides, and the precipitated globulin found to consist of well-formed spheroids and masses of confluent spheroids. The globulin was strongly colored, since much coloring matter was extracted from the shorts, render- ing the extract a deep red-brown in color. The precipitate was then filtered off, again dissolved in the salt solution and placed in a dialyzer. When free from chlorides, the precipitated globulin was filtered off and found to be still much colored. After pa'rtly washing with water it began to dissolve,1 and was therefore next washed with dilute alcohol and finally with absolute alcohol and with ether. Dried over sulphuric acid, it weighed 2.22 grams, and had the composition shown under the heading " Preparation 25 " in the table on page 39. JThis solubility of a globulin is due to the formation of acid compounds which are soluble in water, but insoluble in a dilute saline solution. Cf. Osborne, Journal Amer- ican Chemical Society, 1902, xxiv, p. 39. EXPERIMENTAL. 39 The filtrate and washings obtained from this preparation after the second precipitation by dialysis were precipitated by adding a few drops of sodium- chloride solution. The precipitate produced, which was nearly free from coloring matter, weighed 1.25 gram. When analyzed it gave the results shown under the heading " Preparation 26" in the accompanying table. Preparations 25, 26, and 2j. Preparation 25. Preparation 26. Preparation 27. I. Ash- free. I. Ash- free. I. II. Average. Hydrogen . . . Nitrogen .... Oxveren P.ct. 50.79 6.80 18.19 0.66 P. ct. 51.00 6.83 18.26 0.66 23-25 P. ct. P. ct. P.ct. P. ct. P.ct. 18.56 18.64 18.15 18.43 18.29 Ash 0.40 0.47 IOO.00 The solution filtered from this preparation was next heated to boiling and the coagulum obtained, which weighed 1 gram, when analyzed was found to contain the same amount of nitrogen as the globulin. It was considered to be a coagulated globulin which had been held in solution in water by acid. Its nitrogen content is shown under the heading " Preparation 27." The total quantity of globulin obtained from the shorts was in this case 4.47 grams, being nearly twice as much as was similarly obtained from a like quantity of the flour.1 Table 6. — Summary of analyses of wheat globulin. 23- 24. 25- 26. 27. Average. Hydrogen .... P. ct. GI-07 6-75 18.27 } 23.91 P.ct. 51.01 6.97 18.48 f 0.71 I 22.83 P. ct. 51.00 6.83 18.26 0.66 23-25 P.ct. P.ct. P.ct. 51-03 6.85 13.39 0.69 23.04 18.64 18.29 100.00 100.00 100.00 100.00 1 Teller (Bull. 42, Ark. Agr. Exp. Sta. , 1896) found a much larger quantity of leucosin and globulin nitrogen in the samples of bran which he examined than he found in the flours. 4-0 THE PROTEINS OE THE WHEAT KERNEL. THE PROTBINS OF THE WHEAT EMBRYO SOEUBEE IN SODIUM- CHLORIDE SOLUTION. The Globulin of the Wheat Embryo. Wheat-germ meal treated with 10 per cent sodium-chloride brine forms a dense jelly-like mass, from which it is nearly impossible to separate the solution. With 3 per cent brine a manageable extract can be made by using from six to ten times as much solvent as meal. Thus ioo grams of the meal treated with 600 cc. of 3 per cent sodium chloride yielded in 15 hours 400 cc. of clear filtrate. As has just been shown, the aqueous extract on dialysis, in consequence of a change which affects leucosin, deposits a large amount of protein, chiefly in the coagulated form. In order to obtain preparations of the protein substance soluble in salt solutions, but insoluble in water, which should be free from this coagulable albumin, 2000 grams of germ meal were treated with 20 liters of 3 per cent sodium-chloride solution heated to 700, whereby the leucosin was coagulated and the salt-soluble globulin brought into solution. The extract, neutral to litmus, was filtered clear, at once saturated with ammonium sulphate, and the proteins thus precipitated collected on a filter, dissolved in water, and the clear solution dialyzed in running water. Protein matter separated on dialysis in spheroids, which, like many other plant-globulins, united to a plastic mass on the bottom of the dia- lyzer. This precipitate was dissolved in sodium-chloride solution and, after filtering absolutely clear, dialyzed for 48 hours, the large precipitate which separated allowed to settle, and the solution, which was nearly free from protein, decanted. A portion of the precipitate was washed first with water, which rendered it opaque and dense, then with dilute and finally absolute alcohol, and dried over sulphuric acid. This weighed 5.22 grams, and is preparation 28. The rest of the precipitate was completely dissolved in 125 cc. of 10 per cent sodium-chloride solution. To this, water was added until its volume was 425 cc, thus making a sodium-chloride solution of nearly 3 per cent. From this diluted solution a gummy deposit separated, from which the fluid was soon completely decanted. The latter was further diluted with 325 cc. of water and the precipitate which resulted allowed to settle to a viscid trans- parent deposit. From this precipitate the solution was again decanted and dialyzed for 48 hours, but not more than a trace of globulin was deposited. The two precipitates produced by dilution formed preparations 29 and 30, weighing respectively 11.4 grams and 8.15 grams. A part of each of these preparations was set aside for analysis, and the rest, dissolved together in 10 per cent sodium-chloride solution, allowed to stand over night at 40. The EXPERIMENTAL. 41 solution was then decanted from a slight sediment, filtered clear, and heated to 8o°, in order to coagulate any leucosin which might be present, and after 2 hours filtered from a very small coagulum which had gradually formed. This filtrate was dialyzed in water for 4 days, and the globulin which sepa- rated gave preparation 31. The solution filtered from the first dialysis precipitates, which yielded preparations 28, 29, and 30, was further dialyzed ; a little globulin, which separated, was filtered out and the filtrate dialyzed iii alcohol for 4 days. A precipitate was produced which, when washed with absolute alcohol and dried, weighed 25 grams. This substance con- sisted of protein which is described on page 47. Another series of fractional precipitations of this globulin-like protein was made by extracting 4 kilograms of the oil-free germ meal with 27 liters of 3 per cent sodium-chloride solution heated to 67 ° at the time it was applied to the meal. The mixture was thoroughly stirred and thrown on filters. A clear filtrate of about 12 liters was finally obtained, which was saturated with ammonium sulphate. The precipitate produced was dissolved in water and its solution dialyzed for 48 hours ; where- upon a large quantity of spheroids separated, which on settling united to a coherent mass. This precipitate was washed by decantation with water, dissolved in brine, and its solution made faintly alkaline to litmus by cautiously adding decinormal potassium-hydroxide solution. In order to separate phosphoric acid, a little calcium-chloride solution was then added to this very slightly alkaline liquid, and the latter, though apparently free from any precipitate of calcium phosphate, was filtered, whereby a little suspended matter was removed. The solution was made exactly neutral to litmus by adding 56 cc. decinormal hydrochloric acid and dialyzed for 18 hours. A gummy precipitate (A) adhering to the bottom of the dialyzer then separated, from which the solution (B) was decanted almost completely. The precipitate (A) was dissolved in about 200 cc. of 5 per cent sodium- chloride solution and the liquid was poured into 800 cc. of water. The resulting flocculent precipitate settled rapidly to a coherent deposit, from which the solution was decanted. The deposit was repeatedly washed by decantation with water, which caused it to lose its gummy character and become opaque, white, and granular. It weighed 15.5 grams and was preparation 32. The solution marked B was further dialyzed for 48 hours, when a second precipitate formed, which, like 32, completely dissolved in sodium-chloride solution to a solution perfectly neutral to litmus. This precipitate was washed by decantation with water, but the finer part settled so slowly that it was necessary to decant it together with the water. The sediment, after exhausting with absolute alcohol and drying, weighed 23.5 grams, and formed preparation 33. On long standing the decanted washings deposited the finely divided matter, which was then collected on a 42 THE TROTEINS OF THE WIII-AT KERNEL. filter, dissolved in sodium-chloride solution, and precipitated by water, giving 15.4 grams of preparation 34. To determine the quantity of globulin contained in the oil-free germ meal, 200 grams of the meal were treated with 2000 cc. of 3 per cent sodium- chloride solution heated to 65 ° and the extract filtered perfectly clear. Of this, 1000 cc. were dialyzed until free from chlorides, when the precipitate of spheroids was filtered out. This preparation, 35, formed 5.05 per cent of the oil- free meal. To obtain a quantity of this globulin for digestion with pepsin, a quantity of germ meal was extracted with 3 per cent sodium-chloride solution heated to 700; the extract was filtered clear and saturated with ammonium sul- phate. The precipitate produced was dissolved in water and the resulting gummy and somewhat turbid solution filtered clear. The filtrate was dia- lyzed until the solution gave no turbidity on pouring into distilled water. The protein, which had then separated in spheroids, weighed 27.3 grams, preparation 36. A part of the extract from which 36 had been prepared was mixed with an equal volume of decinormal potassium hydroxide solution — about twice the quantity necessary to neutralize the extract to phenolphthalein. The solu- tion was then dialyzed in distilled water frequently renewed, and in this way a considerable quantity of phosphorus was separated in the alkaline dialysate. When all, or nearly all, which it was possible to separate in this way had been removed, the solution in the dialyzer was neutralized with hydrochloric acid until it no longer reacted alkaline to litmus. This caused a turbidity. The acid was then further added until an acid reaction with litmus was obtained, producing a precipitate from which, after settling, the solution was decanted. The precipitate was then dissolved in sodium-chloride solution, its solution filtered clear and dialyzed, whereby a substance was precipitated in spheroids, which formed preparation 37, weighing 3.0 grams. These preparations had the composition shown in table 7. Table 7. — Composition of preparations extracted by sodium-chloride solution from the wheat embryo. Carbon Hydrogen. . . Nitrogen Sulphur Phosphorus . Ash 2Os in ash. 28. P.ct. 18.14 0.49 1.15 2.29 1.66 29. P.ct. IS.2I 0.56 1.03 1.86 1-34 30- P.ct. 48.77 6.44 18.12 0.51 i-35 2.25 1.68 3i- P.ct. 50-03 7.04 18.39 0.60 0.76 I.V> 0.84 32. P. ct. 50.23 6.89 IS.23 0.53 0.56 1.22 O.80 33- P.ct. 48.17 6.54 IS.06 0.55 1.41 3-85 2.00 34- 35- P.ct. 49-39 6.78 17.95 0.4S 1. 17 2.60 1.82 P.ct. 48.75 6.52 18.16 0.63 1. 41 2.66 2.00 *6. P.ct. 49-79 6.76 18.01 0.61 1. 11 1. 11 0.68 37- P.ct. 48.67 6.56 17-97 0.61 1-55 2-94 2.30 EXPERIMENTAL. 43 These analyses, when calculated free from nucleic acid and ash, as was done for the albumin preparations, in the manner described on page 27, gave .the results set out in table 8. Table 8. — Composition of the globulin contained in the preparations extracted from the wheat embryo by sodium-chloride solution. 28. 29. 3°- 3i- 32. 33- 34- 35- 36. 37- P.ct. P. ct. P.ct. 51-37 6.83 1S.62 0.60 22.58 P.ct. 51-58 7-31 18.70 0.66 21-75 P. ct. 51.40 7.08 18.45 0.57 22.50 IOO.CO P. ct. 5I-56 7.07 18.85 0.67 2185 P.ct. 51-86 7.19 18.41 o.55 21.99 P.ct. 51.40 6.94 18.71 o.75 22.20 100.00 P.ct. 51-98 7.12 18.37 0.70 21.83 P. ct. 51-70 7-05 18.53 0.75 21.97 Nitrogen . . . Sulphur Oxygen.. . . 18.59 0.57 18.59 0.63 roo.oo IOO.CO 100.00 100.00 100.00 100.00 These figures plainly indicate that these globulin preparations are mixtures of nucleates of protein substance of the same ultimate composition, and con- tain from 5 to 15 per cent of nucleic acid. This protein has very nearly the same composition as the globulin occurring in the kernel of wheat, rye, barley, and maize. In the entire kernel so little of this globulin is present that it is difficult to prepare it pure therefrom. For this reason the analyses given in table 9 do not agree as closely as they might otherwise be expected to. From the whole seed this globulin is obtained entirely free from phos- phorus, which is probably due to the much greater proportion of protein matter to nucleic acid in the entire seed compared with that existing in the embryo. Table 9. — Composition of the globulin contained in various cereals. Wheat embryo. Wheat kernel. Rye1 kernel. Maize2 kernel. Barley3 kernel. Carbon P. ct. 51-57 7.07 18.60 0.65 22.11 P. Ct. 5I-03 6.S5 18.39 0.65 23.08 P. Ct. 51-19 6-74 18.19 } 23.88 P.ct. 51-99 6.81 18.02 f 0.66 I 22.52 P.ct. 5088 6.65 18.10 } 24.37 Oxygen 100.00 100.00 100.00 100.00 100.00 1 Osborne, Journal American Chemical Society, 1895, xvn, p. 429. 2 Chittenden & Osborne, American Chemical Journal, 1891, xin, pp. 327, 385, and 1892, xiv, p. 20. 3 Osborne, Journal American Chemical Society, 1895, xvn, p. 539. 44 THE PROTEINS OF THE WHEAT KERNEL. Having determined the composition of this globulin-like protein and also that of the albumin, it became clear that several preparations obtained from the aqueous extract were mixtures of these two substances, thus showing the globulin to be present to some extent in the aqueous extract. As noted on page 24, when 2000 cc. of an aqueous extract of about 650 grams of the meal were dialyzed in running water for 4 days a dense tur- bidity was formed, which could not be removed by filtration. This, however, on adding a little hydrochloric acid, was converted into a precipitate, which was readily dissolved by adding sodium chloride sufficient to make a 3 per cent solution, and was precipitated from this solution by dialysis. Preparation 38, weighing 9 grams, was thus obtained, which, dried at 1 io°, had a composition which corresponds pretty nearly with that of a mixture of 60 per cent of the globulin with 40 per cent of leucosin, except that the amount of sulphur found was somewhat greater than that calculated. The analysis is shown in the following table : Preparations 38 and jg. Preparation 38. Preparation 39. I. II. Arer- age. Corrected for ash and nu- cleic acid. Calculated for 60 per cent globu- lin and 40 per cent leucosin. I. Corrected for ash and nu- cleic acid. Calculated for 40 per cent globu- lin aud 60 per cent leucosin. Phosphorus .... P. ct. 48.30 6.49 17.40 0.83 1. 91 P.ct. 47.92 6.41 17.24 0.85 P.ct. 48.11 6.45 17.32 0.84 1. 91 P. ct. 5I-70 7.07 17-74 1.08 P.ct. 51-95 7.07 17-74 0.91 P.ct. 49-49 6.81 16.87 0-93 0.89 4.00 2.01 P.ct. 5I-80 7.14 17.32 1. 14 P. ct. 52.13 7-03 17.30 I.05 22.51 22.53 22.6o 22.49 Ash 3-95 2-95 P205 in ash 100.00 100.00 IOO.OO IOO.OO After heating another portion of the same aqueous extract to 65 ° for some time and filtering off the coagulum, the filtrate was dialyzed for 5 days into alcohol and the precipitate thereby produced filtered out and exhausted with water. The residue of protein matter coagulated by alcohol, weighing 6.7 grams and marked preparation 39, was then dried at no° and analyzed with the results shown in the preceding table. This analysis corresponds quite nearly with that of a mixture of 40 per cent of the globulin with 60 per cent of leucosin. EXPERIMENTAL. 45 No earlier analyses of the wheat globulin are on record. Weyl1 was the first to call attention to the presence of globulin in wheat, and says that " besides vegetable-vitellin I detected in the 10 per cent sodium-chloride extract of the pulverized seeds of wheat, peas, oats, white mustard, and sweet almonds a second globulin substance. ' ' This he calls vegetable- myosin and gives its coagulation-point at 55 ° to 6o°. It is probable that Weyl in some way mistook for a globulin the albumin already described. Later, Weyl & Bischoff 2 state : On investigating the proteins of wheat, one of us found chiefly an albuminous sub- stance which, on account of its resemblance to myosin, was named vegetable-myosin. This vegetable-myosin must be the mother-substance of the gluten, since in wheat meal, together with it, other nitrogenous substances exist, at the most, only in very small amount. On what experimental evidence this statement rests the writer has been unable to discover, and in view of his experience he is at a loss to under- stand it. Martin 3 considers wheat flour to contain a large amount of globulin of the myosin type, coagulating between 55 ° and 6o°, precipitated by saturation with sodium chloride and ammonium sulphate. Here again the small quantity of albumin contained in the flour has evidently been mistaken for a large quantity of vegetable-myosin. This perhaps is not surprising, as the precipitates obtained by saturating sodium-chloride extracts with ammonium sulphate appear very bulky, and in the absence of an actual determination of the weight of these precipitates misleading conclusions might easily be reached. The only globulin found by the writer in extracts of wheat meal, either winter or spring wheat, is the one just described, which in properties and composition closely resembles those globulins which have been found in other seeds. THE PROTEOSE OF WHEAT FLOUR. As already stated in describing the reaction of the extract freed from globulin by dialysis, there was found in it one or more proteoses, besides the albumin just described. These were almost wholly precipitated by saturation with sodium chloride or by adding 20 per cent of this salt to the solution, together with a little acetic acid. If the albumin is completely removed by heat and the filtered solution then concentrated, a coagulum gradually develops. This substance must be derived from the proteose-like protein, as this forms nearly if not quite all the protein substance remaining in solution before concentration. If 1 Weyl, Zeitschrift fur physiologische Chemie, 1877, 1, p. 72. 2 Weyl & Bischoff, Berichte der deutschen chemischen Gesellschaft, 1880, xiii, p. 367. 3 Martin, British Medical Journal, 18S6, 11, p. 104. 46 the; proteins of the wheat kernel. the coagulum is removed b)' concentration and long-continued heating and subsequent filtration, wholly uncoagulable proteose-like substances are found in solution. The amount of proteose is extremely small, and no preparations were made for analysis. The coagula obtained by concen- trating the solution filtered from preparations 3 and 4, respectively, were analyzed with the results shown in the accompanying table under the head " Preparation 40." Another coagulum similarly obtained from the solution filtered from prep- aration 4 was also analyzed and the following figures obtained : Preparations 40 and 41. Preparation 40. Preparation 41. I. II. Average. Ash-free. I. II. Average. Ash-free. Carbon Hydrogen . . . P.ct. 51.35 P.ct. 51.60 P. ct. 51.48 P.ct. 51.62 P.ct. 51-73 6.82 17.29 I.... P. ct. 17.26 P. ct. 51-73 6. 82 17.28 P.ct. 51-86 6.82 17.32 24.00 Nitrogen .... Sulphur 16.79 16.58 16.69 16.73 Oxygen J 0.25 0.25 Ash 0.27 100.00 the proteose op the wheat embryo. In making the preparations from the embryo already described consider- able quantities of crude proteose were obtained from both the aqueous and sodium-chloride extracts. After the leucosiu and the globulin had been separated as completely as possible, the solutions containing the proteoses were dialyzed in alcohol, and the precipitates produced were washed and dried over sulphuric acid. A mixture weighing 15.4 grams was made by uniting several such prep- arations that had been obtained from aqueous extracts from which most of the other proteins had been separated, without heat, by saturating with sodium chloride and dialysis in alcohol. The mixture contained much matter made insoluble in water by the final treatment with alcohol. This was filtered out, washed thoroughly with water and with alcohol, and when dried weighed 4.18 grams, and was marked preparation 42. The filtrate from this was saturated with ammonium sulphate, the precipitate redissolved? and again precipitated in the same way. The solution of the second pre- cipitate was dialyzed in cold distilled water until free from sulphate, and then for several days in alcohol ; the precipitate thus produced was dissolved in water, a little insoluble matter filtered out, and its clear solution saturated EXPERIMENTAL. 47 with sodium chloride, which produced a small precipitate. This was filtered out, dissolved, and its solution dialyzed in water. The salt-saturated filtrate was likewise dialyzed, and when both solutions were free from chlorine the dialyzers were transferred to alcohol and the proteose thereby precipitated. The proteose separating on saturation with salt gave o. 6 gram of prepara- tion 43 ; that from the salt-saturated solution 0.97 gram of preparation 44. This small yield of proteose indicates that the greater part had diffused through the parchment paper during the long dialysis to which the solutions had been subjected. Another crude product was obtained by dialyzing an aqueous extract in alcohol after separating the leucosin which had been coagulated by heat. This, weighing 35 grams, was dissolved in water and the insoluble matter filtered out, washed and dried, giving preparation 45, weighing 7.26 grams. The filtered solution was saturated with ammonium sulphate, the pre- cipitate dissolved in water, and the clear solution dialyzed in distilled water until free from sulphates, and then in alcohol. The substance thus sep- arated was again dissolved in water and its solution saturated with salt ; the precipitate thus produced was dissolved in water, and its solution, as well as the salt-saturated filtrate, were dialyzed in water. When free from chlorine, these solutions were dialyzed in alcohol and yielded, respectively, prepara- tions 46, weighing 4 grams, and 47, weighing 1.84 grams. Another preparation of crude proteose was obtained by extracting the meal as described on page 41 with 3 per cent sodium-chloride solution heated to 700, dialyzing the extract in water, coagulating the leucosin by heat, and precipitating the proteose by dialysis in alcohol. A mixture of such prep- arations, weighing 31.6 grams, was treated with water, the insoluble matter filtered out, washed, and dried, giving 5.16 grams of preparation 48. The filtered solution was saturated with ammonium sulphate, the precipi- tate dissolved in water, the solution dialyzed in distilled water till free from sulphate, and then in alcohol. The separated proteose was redissolved in water and its solution saturated with sodium chloride. The precipitate which resulted was filtered out, dissolved in water, and its solution, as well as the salt-saturated filtrate, were dialyzed in water till free from chlorine, and finally in alcohol. The products thus obtained formed, respectively, preparation 49, weigh- ing 0.75 gram, and 50, weighing 1.35 grams. One other proteose prepara- tion was made from the aqueous extract described on page 26, from which the phosphorus was largely separated by making it slightly alkaline and adding calcium chloride. After heating the extract to boiling and filtering out the coagulum, the filtrate was dialyzed in alcohol, the resulting pre- cipitate dehydrated with absolute alcohol, dried over sulphuric acid, redis- solved in water, and precipitated by saturating with ammonium sulphate. 48 THE PROTEINS OF THE WHEAT KERNEL. The gumm)' precipitate, having the general appearance and properties of similar precipitates of the proteoses obtained by the action of pepsin, was dissolved in water, dialyzed free from sulphates, and then precipitated by dialysis in alcohol, giving 2 grams of preparation 51. These preparations were dried at no0 and analyzed with the results shown in table 10. Tabi^e 10.- —Composition of alcohol coagula and of proteose prepared from the wheat embryo. Residues of other pro- teins coagulated by alcohol. Proteose precipitated by sodium chloride. Proteose soluble in saturated NaCl solution. 42. 45- 48. 43- 46. 49- 44- 47- 5°- 5i- Carbon Hydrogen Nitrogen Ash P. ct. 52.36 6.98 16.01 I.85 22.80 P.ct. 49-44 6.85 16.00 4.08 23-63 P. ct. 5i 93 6.87 16.30 1.30 23.60 100.00 o.95 P.ct. 16.79 P.ct. 49-94 6.80 17.08 1.24 24.94 P.ct. 16.26 P.ct. 48.46 6.70 16.91 } 27.93 P.ct. 48.70 6-73 16.76 27.81 P.ct. 48.44 6.71 16.16 28.6c P. ct. 48.99 6.85 16.89 f 1. 10 I 26.17 IOO.00 O.Sl 100.00 14-13 100.00 0. t,o\ 0.77 IOO.OO i-i3 100.00 1. 00 100.00 0.74 100.00 1.27 From these analyses it is seen that the matter insoluble in water, forming preparations 42, 45, and 48, consists of coagulated protein apparently mostly derived from leucosin. The high proportion of sulphur in 42 and 45 is due to calcium sulphate precipitated by alcohol from the aqueous extract. The remaining preparations have the low percentage of carbon characteristic of proteoses made by pepsin digestion. THE PROPORTIONS OF THE VARIOUS PROTEIN SUBSTANCES OF THE WHEAT EMBRYO. Twenty grams of fresh germ meal, from which the ether-soluble constit- uents had not been separated, were treated with 500 cc. of water, and after shaking for some time the extract was filtered clear. Two portions of 100 cc. each were treated with a few drops of very dilute hydrochloric acid and heated in a boiling water-bath. The coagulum which separated was col- lected on a filter and its nitrogen determined. To the filtrate from one coagulum tannin was added, and nitrogen was determined both in the pre- cipitate and in the filtrate. Another lot of 20 grams was treated in the same way and nitrogen determined in the heat-coagulum formed in each of two portions of 100 cc. The amount of nitrogen corresponding to 1 gram of germ meal was found in the four coagula to be 0.0163 gram, 0.0156 gram, 0.0159 gram, and 0.0162 gram, in the tannin precipitate 0.0062 gram, and in the solution filtered from the latter 0.0062 gram. EXPERIMENTAL. 49 Twenty grams of germ meal were extracted with 500 cc. of 3 per cent sodium-chloride solution heated to 700, whereby the leucosin was coagu- lated and the globulin and proteose dissolved. Of the clear filtered extract 100 cc. yielded with tannin a precipitate containing 0.0166 gram nitrogen per gram of meal extracted. Two portions of the meal, each of 1 gram, were exhausted with 3 per cent sodium-chloride solution heated to yo° and nitrogen determined in the residues. The 0.0331 gram and 0.0309 gram of nitrogen found in the resi- dues were from the leucosin and insoluble nitrogenous bodies, so that the nitrogen belonging to the latter equaled 0.0171 gram and 0.0149 gram. From the average of these figures we find the following amounts of the different forms of nitrogen in 1 gram of the wheat-germ meal : N, grains. Insoluble in water and salt solution 0160 Insoluble in water, but soluble in salt solution (globulin nitrogen) . . .0100 Soluble in water and coagulable by heat (albumin nitrogen) 0160 Soluble in water, uncoagulable by heat, precipitable by tannin (pro- teose nitrogen) 0050 Not precipitable by tannin (non-protein nitrogen) 0060 Total 0530 Found by direct nitrogen determination 0531 It has been shown that the coagulated leucosin preparations contain about 10 per cent of nucleic acid, the globulin about 15 per cent, while those of the proteose contain none. Deducting these quantities from the nitrogen given above, it is found that 9.5 percent of the embryo is leucosin, 4.84 per cent globulin, and 3.03 per cent proteose. The bodies which are represented by the insoluble nitrogen could not be separated from the embryo. The residue, after extraction with hot-salt solution, contained 0.0076 gram of phosphorus. In view of the large proportion of nucleic acid found in the extracts of the embryo, it is not improbable that this phosphorus mostly belongs to nucleic acid, and that the insoluble nitrogen largely belongs to compounds of protein with relatively much nucleic acid. DIGKSTION OF THE PHOSPHORUS-CONTAINING PROTEIN PREPARATIONS WITH PEPSIN-HYDROCHEORIC ACID. IvEucosin Nucleate. Ten grams of the coagulated albumin, preparation S, were suspended in 400 cc. of water and dissolved by adding 100 cc. of decinormal potassium- hydroxide solution. To the nearly clear solution which resulted an equal volume of 0.4 per cent hydrochloric acid was added, together with some pepsin, and the mixture digested at 37 °. In a short time the solution 4 50 THE PROTEINS OF THE WHEAT KERNEL. became perfectly clear, but later deposited a large coherent precipitate, which gradually contracted, but at the same time retained the form of the lower part of the beaker. From this the clear solution was decanted, the precip- itate thoroughly washed by decantatiou, suspended in water and dissolved by adding 28 cc. of decinormal potassium-hydroxide solution, an amount of alkali just sufficient to dissolve all the substance, and at the same time make the solution neutral to litmus. When to this solution decinormal acid was gradually added, no precipitate appeared until nearly one-half the quantity of acid required for complete neutralization had been added, but with 28 cc. the solution was neutralized and the nuclein completely precipitated, the addition of 2 cc. more acid giving no turbidity in the filtered solution. This precipitate formed preparation 52, weighing 1.54 grams. To precipitate this substance a quantity of acid was added exceeding that of the alkali employed for solution by just 2 cc. The filtrate from the pre- cipitate, however, required not 2 cc. of alkali, but 8.5 cc. for neutralization to phenolphthalein, showing 6.5 cc. of alkali to have been neutralized by the acid of the nuclein originally dissolved. The neutralized filtrate left on evaporation 0.3975 gram of substance, the aqueous solution of which was precipitated by hydrochloric or nitric acid, but not by ammonium molybdate solution until after boiling with acid for some little time, when yellow phos- phomolybdate was precipitated. These facts indicate the presence in this filtrate of a nucleic acid. More nuclein was made from the same preparation, 7, by suspending 30 grams in 0.2 per cent hydrochloric acid containing pepsin, which, even at 200, caused within 2 hours complete solution of the coagulated protein. The solution was digested at 37 ° for 48 hours, during which time much nuclein separated, having the appearance and properties of the preparation just described. After decanting the clear solution and thoroughly washing the residual nuclein, the latter was suspended in water and dissolved in 72 cc. decinor- mal potassium-hydroxide solution. The solution thus obtained was made neutral to litmus by adding 11 cc. of decinormal hydrochloric acid, but no precipitate appeared till 1.5 cc more of acid were added. To the solution 72 cc. decinormal hydrochloric acid were added, giving a precipitate, prepa- ration 53, which weighed 3.4 grams. The filtrate from this precipitate, as in the former case, was strongly acid, requiring 12 cc. of decinormal potas- sium hydroxide to neutralize it to phenolphthalein. Two other preparations of nuclein were made from 8.493 grams of 13 and 9.804 grams of 16, both being substances precipitated from the aqueous extract by saturating with sodium chloride. Each portion was suspended in about 300 cc. of 0.2 per cent hydrochloric acid, containing 0.1 gram of pepsin, and, with frequent EXPERIMENTAL. 51 stirring, digested at 400 for 24 hours. Throughout the digestion a large part of the substance remained undissolved. An equal volume of 0.2 per cent hydrochloric acid, containing o. 1 gram of pepsin, was again added to each, and the digestion continued for 24 hours longer. The insoluble matter which remained was not coherent like the two former nuclein products, but consisted of a white, very finely divided substance, which was easily filtered out and washed. From 13, 4.04 grams of preparation 54 were obtained, and from 16, 4.16 grams of 55. Globulin Nucleate. Fifteen grams of a mixture of nearly equal parts of the globulin prepa- rations 32 and 33 were next suspended in 0.2 per cent hydrochloric acid containing 0.2 gram of pepsin, which, within a short time, almost completely dissolved the protein matter. From this solution, on further digestion, the nuclein separated, forming a coherent deposit. After 72 hours' digestion the clear solution was decanted, the deposit dissolved in a little ammonia, and its solution filtered perfectly clear from a very slight gelatinous residue. The solution was then treated with acetic acid added in excess of the amount necessary to neutralize it to litmus. Since, even on standing, the precipitate so produced separated imperfectly, an equal volume of alcohol was added. The substance, which then separated well, gave 2.38 grams of preparation 56, or about 16 per cent of the original substance. The filtrate from the acetic acid precipitate, on adding hydrochloric acid, gave a further slight precipitate, which had properties characteristic of nucleic acid. Still another preparation of nuclein was made from the globulin by sus- pending 10 grams of 36 in water and adding 50 cc. of decinormal potassium- hydroxide solution. This solution was neutralized and an equal volume of 0.4 per cent hydrochloric acid at once added, producing a turbid solution, which, however, contained no visible particles. To this pepsin was added and the mixture digested for 40 hours, during which time a coherent deposit of nuclein formed on the bottom of the beaker. From this the clear solution was decanted. The deposit was then thoroughly washed with water and dissolved in 43 cc. of decinormal potassium-hydroxide solution. To this clear solution 43 cc. of decinormal hydrochloric acid were added, causing a gummy precipitate, which could not be filtered until 15 cc. more acid had been added, when the precipitate rapidly settled as a coherent deposit, from which the solution was soon decanted. This solution required for neutrali- zation to litmus 16 cc. of decinormal potassium-hydroxide solution, and to phenolphthalein 18 cc. The precipitate when washed and dried gave 2.2 grams of preparation 57. 52 THE TROTEINS OF THE WHEAT KERNEL. These six preparations were all dried at i io° and analyzed with the results shown in table n. Table ii. — Composition of nuclein from the proteins of the wheat embryo. Carbon Hydrogen.. . Nitrogen.. . . Sulphur Phosphorus . Ash P205 in ash. . Bases in ash. 52. 53- 54- 55- 56. 57- P. ct. P.ct. P.ct. P. ct. P. ct. P.ct. 44.87 44-35 42.68 43-35 39-42 41.92 5.82 5-77 5-45 5-47 5-03 5-25 16.04 16.64 16.12 16.01 16.05 17.00 0.97 1.03 0.65 0.85 o-53 0.46 4.58 5-07 5-32 4.88 5-27 5.63 0.60 0.78 1.72 1.72 17.42 1. 17 0.29 o.55 1.24 0.94 10.56 0.69 0.31 0.23 0.48 0.78 6.86 0.48 These analyses of nuclein were calculated free from the bases of the ash and from nucleic acid, in the way previously described, with the results given in table 12. Table 12. — Composition of protein matter contained in the nuclein. Carbon . . . Hydrogen Nitrogen. . Sulphur.. . Oxygen. . . 52. 53- 54- 55- 56. 57- P.ct. P. ct. P.ct. P.ct. P. ct. P. ct. 53.65 54-77 5l-8o 52.36 52 30 51.64 7.23 7.46 6.S5 6.73 6.91 6.60 16.68 17-56 16.31 16.31 19.31 18.93 1.98 2-37 1. 61 1.89 1-53 125 20.46 17.84 23-43 22.71 19-95 21.58 100.00 100.00 100.00 100.00 100.00 100.00 The composition of the protein matter in 54 and 55 is very nearly that of leucosin, except as regards sulphur, the amount of which is decidedly greater. On the other hand, 52 and 53, which also were derived from preparations whose protein matter was leucosin, differ in composition very decidedly from that substance. This is probably because on pepsin digestion the substance of preparations 54 and 55 remained throughout undissolved, whereas 52 and 53 separated on pepsin digestion from nearly clear solutions, and therefore doubtless their protein matter had been to some degree altered by the pepsin before separating as an insoluble compound with nucleic acid. The two nucleins, 56 and 57, from the globulin, which also had separated from solution, show similar differences in composition when compared with the unaltered globulin, carbon and nitrogen being higher and sulphur very much higher than in the globulin. EXPERIMENTAL. 00 PROTEIN SOLUBLE IN DILUTE ALCOHOL — GLIADIN. As already stated, wheat flour yields to dilute alcohol a considerable amount of protein matter. Treatment of the residue remaining after ex- tracting the flour with ioper cent sodium-chloride brine likewise removes a large amount of protein, as does also extraction of the gluten obtained by washing the dough with water. Extracts were made with alcohol under all these conditions, and the protein extracted subjected to repeated fractional precipitations. Direct Extraction with Dilute Alcohol. 5000 grams of the straight flour were extracted with 10 liters of alcohol, 0.90 sp. gr., and allowed to soak over night. The next morning the mixture was stirred, and, after settling, the clear solution poured off. Three liters more of alcohol, 0.90 sp. gr., were then added, and after standing some time the supernatant liquid was decanted and the residue squeezed nearly dry. The solution so obtained was designated "extract 1." The residue was again treated with 4 liters of 0.90 sp. gr. alcohol and pressed nearly dry. This formed extract 2. The same process twice repeated gave two extracts which, when united, formed extract 3. Each of these three extracts, after filtering perfectly clear, was separately concentrated to one-third its volume, and, after cooling, decanted from the very glutinous, viscid mass which had separated. 1 and 3 yielded much more substance than 2. On stirring with a glass rod the precipitated mass formed a very thick, viscid liquid. This substance was in each case dissolved in a small amount of hot alcohol of 0.90 sp. gr. , in which it was very soluble, and the solution was allowed to cool over night. Most of the substance separated on cooling and the liquid was decanted from it. The solution decanted from the second and third extracts was treated with a quantity of distilled water and a little sodium chloride added. This threw down a small precipitate, which on standing collected on the bottom of the vessel as a clear semi-fluid mass. This was treated with water, absolute alcohol, and ether, and yielded 7.27 grams of prepara- tion 58 from extract 2, and 10.7 grams of the preparation 59 from extract 3. These had the following compositions : Preparations 58 and 50. Preparation 58. Preparation 59. I. Ash-free. I. Ash-free. Nitrogen P. Ct. 17-05 0.76 P. ct. 17.18 P. ct. 17-15 0.65 P. ct. 17.26 Ash 54 the; proteins op the; wheat kernel. On examination both were found to contain some fat which could not be wholly removed, as the substance had dried in a dense, horny form. The residues which had separated from the solutions just described were next washed by thoroughly intermixing with distilled water. The water was found to dissolve some of the protein, which was subsequently precipitated by the addition of a little sodium-chloride solution. After standing over night this precipitate settled to the bottom of the vessel in a transparent layer, from which the solution could be completely decanted. After treating this substance with absolute alcohol, it formed a voluminous white, porous mass, which was digested for some time with ether. From extract i preparation 60 was obtained, weighing 12.4 grams ; from extract 2, united with that from 3, preparation 61, weighing 8.6 grams. These preparations were found on analysis to have the following compo- sitions : Preparations 60 and 61. Preparation 60. Preparation 61. I. II. Average. Ash-free. I. Ash-free. Carbon Hydrogen . . Nitrogen . . . Sulphur .... P. ct. 52.40 6.77 I7.52 1.05 P.ct. 52.58 6.78 17.73 I. II P. Ct. 52.49 6. 78 17-63 1.08 P.ct. 52.52 6.78 17.64 1.08 21.98 P.ct. 52.69 6.77 17.74 1.26 0.15 P. ct. 52-17 6.78 17.77 1.26 21.42 100.00 Ash 0.06 100.00 The residues remaining after washing with distilled water were then digested with alcohol of 0.820 sp. gr. , which dissolved much of the substance. After standing some time, the strong alcoholic solutions were decanted from the residues and found to consist of milky liquids. The addition of a few drops of 10 per cent sodium-chloride solution immediately produced in each a very large, curdy precipitate, the liquids from which they separated being left perfectly clear. From the solution from the first extract, which was about one liter in volume, 32.26 grams of substance were obtained after dehydration with absolute alcohol and digestion with ether. This was marked preparation 62. From the second extract was similarly obtained preparation 63, weighing 5.34 grams, and from the third extract preparation 64, weighing 17.43 grams. The filtrates from 62 and 63 were found to be almost wholly free from protein, but that from 64 left on evaporation a residue which, when freed from fat, weighed 7.53 grams, preparation 65. EXPERIMENTAL. These preparations were found to have the following compositions : Preparations 62 and 63. 55 Preparation 62. Preparation 63. I. Ash-free. I. II. Average. Ash-free. Hydrogen . . Nitrogen . . . Sulphur , , , Oxygen .... P. ct. 52.59 6.70 17.64 1.22 P.ct. 52.67 6.71 17.66 1.22 21.74 P. ct. 52.28 6.87 17.90 1. 21 P.ct. 52.54 6.79 P. Ct. 52.41 6.83 17.90 1. 21 P. Ct. 52.55 6.85 17.94 1. 21 21.45 Ash O.I5 O.27 100.00 100.00 Preparations 64 and 65. Preparation 64. Preparation 65. I. II. Average. Ash-free. I. II. Average. Ash-free. Carbon Hydrogen . . Nitrogen . . . Sulphur .... Oxygen .... P. ct. 52.52 6.72 17.60 1.23 P. ct. 52.82 6.79 P.ct. 52.67 6.76 17.60 1.23 P. ct. 52.74 6.77 17.62 1.23 21.64 P. ct. 52.13 6-97 17-35 i-35 P. ct. 52.42 6.85 17.19 1.41 P.ct. 52.28 6.91 17.27 1.38 P. ct. 52.39 6-93 I7.3I 1.38 21.99 Ash 0.14 0.22 100.00 100.00 The residues which remained after treatment with alcohol of 0.820 sp. gr. were then dehydrated with absolute alcohol and digested with ether. From extract 1 preparation 66 was obtained, weighing 63 grams ; from extract 2 preparation 67, weighing 2.1 grams, and from extract 3 preparation 68, weighing 41.2 grams. Preparation 66, which constituted the principal fraction of the protein extracted, was further treated in the following manner : 20 grams were dissolved in 250 cc. of 0.90 sp. gr. alcohol and found to yield a clear solu- tion, which was then poured into 800 cc. of absolute alcohol, whereby a considerable precipitate was at once separated, leaving the solution milky. This substance was dehydrated with absolute alcohol and digested with ether, yielding preparation 69. The filtrate was then treated with a few drops of 10 per cent sodium-chloride solution, causing a heavy precipitate, which on stirring rapidly agglutinated and adhered as a mass to the stirring- 56 Tin: PROTEINS OE THE WHEAT Ki'.RXEE. rod. This was removed, treated in the usual manner, and marked prepara- tion 70. The mother- liquor from which this separated, after standing over night, deposited a further small amount of protein, which, after treatment with absolute alcohol and ether, gave preparation 71. These six substances were analyzed and found to have the following compositions : Preparations 66 to 7/. Preparation 66. Preparation 67. I. Ash-free. I. II. Average. Ash-free. Carbon Hydrogen . . . Nitrogen .... Sulphur Oxygen P. Ct. 52.81 6.81 17.66 1. 11 P. ct. 52.S4 6.S1 17.67 1. 11 21-57 P. ct. Pet. P. ct. P. ct. 15-42 15-28 15-35 15-42 Ash 0.06 0.42 0.42 100.00 Preparation 68. Preparation 69. I. II. Average. Ash-free. I. Ash-free. Carbon Hydrogen. . . Nitrogen .... Sulphur Oxvsren P. ct. 52.90 6.74 17.24 1.05 P. ct. 52.99 6.82 17.36 P. ct. 52.95 6.78 17.30 1.05 Pet. 53-02 6.79 17-32 1.05 21. S2 P. ct. P. ct. 17.67 17.69 Ash 0.15 0. 10 100.00 Preparation 70. Preparation 71. I. II. Average. Ash-free. I. Ash-free. Carbon Hydrogen . . . Nitrogen . . Sulphur Ash P. Ct. 52.15 6-93 I7-52 \ P. ct. 52.35 6.87 17.84 P. Ct. 52.25 6.90 17.68 P. ct. 52.33 6.91 17.70 23.06 P. ct. 52.2S 7.12 17.79 0.19 Pet. 52.33 7-13 17.82 22.67 J 0.15 1 I CO. 00 100.00 EXPERIMENTAL. 57 If the preceding analyses are brought together as in table 13, the effect of the various fractional solutions and precipitations may be seen at a glance. Tabi^e 13. — Gliadin extracted by direct treatment of the flour with alcohol. From 0.9 sp. gr. al- coholic solution. From water wash- ings. From 0.82 sp. gr. alcoholic solution. 53. 59- 60. 61. 62. 63. 64. P. ct. P. ct. P. ct. 52.52 6.78 17.64 I.08 21.98 IOO.OO 12.40 P. ct. 52.77 6.78 17.77 1.26 21.42 P. ct. 52.67 6.70 17.66 1.22 21-75 P. ct. 52.55 6.S5 17.94 I. 21 21-45 P. ct. 52.74 6.77 17.62 1.23 21.64 Nitrogen 17. iS 17.26 Weight of sub- stance in grams. 100.00 8.60 IOO.OO 32.26 IOO.OO 5-34 100.00 1743 7.27 10.70 From nitrate from 64. Residue after extraction with 0.820 sp. gr. alcohol. Fractional reprecipitations of preparation 66. 65. 66. 67. 68. 69. 70. 7i- Carbon P. ct. 52.39 5-93 I7.3I 1.38 21.99 P. ct. 52.82 6.8l 17.67 I. II 21.57 Pet. 15.42 Pet. 53-o2 6.79 17-32 1.05 21.82 P. ct. 17.69 P. Ct. 52.33 6.91 17.70 1 23.06 P. ct. 52.38 7.13 17.82 22.67 Hydrogen Oxveen Weight of sub- stance in grams. IOO.OO 7-53 IOO.OO 63.O 2.IO 100.00 41.20 100.00 100.00 It is evident that 58 and 59 contain less nitrogen than the great bulk of the protein extracted. This is due to fat which they were found to contain and which could not be wholly removed by extraction with ether, since in drying these preparations were converted into the horny condition which rendered penetration with ether impossible. Preparation 67 was evidently impure, as might be expected, since it con- tained all of the insoluble particles of the entire extract which had escaped nitration, and owing to its small amount these impurities produced a marked effect on its percentage composition. The same is true of 68, but as the quantity of this preparation is so much greater this contamination has pro- duced much less effect on its composition. Preparation 65 was obtained by evaporating the mother-liquor from 64 nearly to dryness and then extracting with absolute alcohol and ether. It would hardly be expected that under 53 the; proteins of the wheat kernel. such circumstances it would be entirely pure. The analyses of the other preparations are in good agreement, and it is evident that no fractional separation of the extracted protein is indicated. The preparations obtained from solution in pure water have the same composition as those from solu- tions in alcohol of 0.820 sp. gr., and also the same composition as the residue remaining after treatment with each of these reagents. The total amount of protein contained in these several preparations is 207.83 grams, being equal to 4.16 per cent of the flour. Table 14. — Gliadin extracted from flour zvitk dilute alcohol after extraction with 10 per cent sodium-chloride solution. Preparation 72. Preparation 73. I. II. Average. Ash-free. I. II. Average. Ash-free. P. ct. 52.72 6.86 17.89 o-95 21.58 Hydrogen . . Nitrogen.. . . Sulphur .... Oxveren .... P.ct. 52.61 6.82 17.62 1. 00 P.ct. 17.78 I.04 P.ct. 52.61 6.82 17.70 1.02 P. Ct. 52.69 6.84 17.73 1.02 21.72 P.ct. 52.65 6.85 17.87 o-95 P.ct. 0.94 P.ct. 52.65 6.85 17.87 o.95 Ash 0.16 0.13 100.00 100.00 Preparation 74. Preparation 75. Preparation 76. I. II. Average. Ash-free. I. Ash-free. I. Ash-free. Hydrogen . . Nitrogen.. . . Sulphur .... P.ct. 52.52 6.82 17.72 1. 10 P.ct. 52.67 6.76 P. ct. 52.60 6.79 17.72 1. 10 P. ct. 52.71 6.81 17-75 1. 10 21.63 P. ct. P. ct. P. ct. 52.62 6.83 17.78 1.08 P.ct. 52.65 6. S3 17.79 1.08 21.65 16.93 17.08 Ash 0.21 0.91 0.05 100.00 100.00 Extraction with Dilute Alcohol After Extracting the Flour with 10 per cent Sodium- Chloride Solution. Four kilograms of "straight flour" were extracted with 10 per cent sodium-chloride solution as long as anything was removed. After squeez- ing as dry as possible in a screw-press, the residue was treated with alcohol so as to yield with the water retained by the meal as nearly as possible a EXPERIMENTAL. 59 / solution containing 75 per cent of alcohol. After digesting 2 days with the solvent, the extract was squeezed out in a press and the process repeated three times. Four extracts were thus obtained. These were each concen- trated to small volume, cooled, and the solution decanted from the precipi- tated mass. This was then washed with distilled water. After removing the salts, the substance from extracts 1 and 2 dissolved to some extent ; that from extract 3 dissolved completely to a turbid solution. By adding sodium chloride the dissolved protein was precipitated. The residues remaining after washing with water were treated with abso- lute alcohol and digested with ether. The precipitates obtained from the water washings by adding salt were treated in the same way. From extract 1 preparation 72 was obtained, weighing 82 grams ; extract 2, preparation 73, weighing 57 grams ; from extract 3, after dissolving in water and precipitating with sodium chloride, preparation 74, weighing 11.3 grams; from extract 4 preparation 75, weighing only 1.35 grams, and from the united water washings of 72 and 73 preparation 76, weighing 5.8 grams. The total weight of these preparations was 157.45 grams, equal to 3.94 per cent of the flour taken. Their composition is shown by the analyses given in table 14 on the preceding page. Extraction of Gluten with Dilute Alcohol. Two kilograms of "straight spring- wheat flour " were made into a dough with distilled water of 200, and then washed in a stream of river water of 50. After washing until nearly all the starch was removed, the gluten was chopped up fine and digested with alcohol of 0.90 sp. gr. at a tempera- ture of about 200. This extraction was continued with repeatedly re- newed portions of alcohol of the same strength as long as anything was removed. The extracts were united, filtered perfectly clear, and concen- trated to about one-fourth their original volume. The residual solution was then cooled and allowed to stand over night to deposit the separated gliadin. The supernatant solution was poured off and the large amount of protein which had separated was then dehydrated by treatment with absolute alcohol. The decanted mother-liquor from which this protein had separated, and also the strong alcoholic solution which resulted from dehy- drating the precipitated mass, were each precipitated by adding a little sodium-chloride solution. The three products thus obtained were united, digested with fresh quantities of absolute alcohol in order to complete the dehydration, and then extracted with absolute ether. Dried over sulphuric acid, the preparation weighed 82 grams, and formed, therefore, 4.10 per cent of the flour taken. Dried at no°, this substance had the composition shown under the head " Preparation 77 " in the table on page 60. 6o THE PROTEINS OF THE WHEAT KERNEL- Thirty grams of preparation 77 were then dissolved in alcohol of o. 90 sp. gr. , and the clear solution evaporated to small volume, cooled, and, as no protein separated, strong alcohol was added until a considerable precipitate resulted, equal to about one-half the dissolved protein. This precipitate, which, if the extracted protein were a mixture, as stated by Ritthausen, would con- tain the bulk of the substances insoluble in strong alcohol, weighed 12 grams. This was marked preparation 78. The solution from which this substance had separated must have con- tained the chief part of the protein called by Ritthausen gluten-fibrin. It was then concentrated to small volume, cooled, water added until a con- siderable precipitate resulted, the solution then heated until all dissolved, and, after cooling, the mother-liquor was decanted from the separated protein. This process was repeated four times, and the precipitate finally obtained dehydrated with absolute alcohol and digested with ether. This preparation, 79, weighed 1.6 grams. Preparations 77, 78, and 79. Preparation 77. Preparation 78. Preparation 79. I. Ash-free. I. Ash-free. I. Ash-free. Carbon Hydrogen . . Nitrogen . . . Sulphur .... Oxygen .... P.ct. 52.33 6.63 17.57 1.08 P.ct. 52.58 6.67 I7-65 1.08 22.02 P. ct. 52.82 6-77 17.62 1.09 0.19 P.ct. 52.68 6.78 17.65 1.09 21.80 P.ct. 52.82 7.18 17.57 1 J 0.04 P.ct. 52.84 7.18 17-57 22.41 100.00 Ash 0.50 100.00 100.00 It is clear from these analyses that no separation into proteins of differing composition had thus been effected. Extraction of "Shorts" with Dilute Alcohol. Two kilograms of "shorts" from the spring-wheat flour were extracted with alcohol of 0.90 sp. gr. and the extract squeezed out with a screw- press. The extract, which was a deep red-brown in color, was filtered perfectly clear, and then concentrated by distillation to about one-third. On cooling, the protein separated, leaving the mother-liquor as a deep coffee-brown liquid. This was decanted, the precipitate dissolved in alcohol of 0.90 sp. gr., and again precipitated by concentration and cooling; the strongly EXPERIMENTAL. 61 colored mother-liquor was decanted and this process repeated. The pre- cipitated protein was then again dissolved in a little dilute alcohol, and the resulting solution poured into absolute alcohol, thereby precipitating the greater part of the protein and leaving the alcohol strongly colored. The precipitate was thus freed from a very considerable part of its coloring matter. After digestion with absolute alcohol, and finally with ether, this preparation, 80, was dried and analyzed with the result shown in the table below : Owing to the fact that this preparation, 80, was still contaminated with coloring matter, and also showed slight differences in composition from the protein extracted by similar treatment from the flour, it was subjected to further treatment with a view to its more complete purification. Apart of the preparation was dissolved in 150 cc. of alcohol of 0.90 sp. gr. and the solution poured into 1000 cc. of absolute alcohol. This produced a turbid liquid, which, on adding a drop or two of sodium-chloride solution, gave a heavy precipitate that rapidly settled, leaving the alcohol colored yellow. This precipitate was again dissolved in diluted alcohol, and pre- cipitated by pouring into ether in order to remove anything soluble in this liquid. No coloring matter was thus removed. The precipitate was then digested with absolute alcohol, yielding preparation 81. The strong alcoholic solution from which this preparation had separated on longer standing deposited a small amount of substance which, when dehydrated, yielded preparation 82. Analysis showed these two substances to have the composition shown in the following table : Preparations 80, 81, and 82. Preparation 80. Preparation 8 1. Preparation 82. I. Ash- free. I. II. Aver- age. Ash- free. I. II. Aver- age. Ash- free. Carbon. . . . Hydrogen . Nitrogen . . Sulphur. . . Oxveren . . , P. ct. 52.75 6.96 17.22 I.36 P. ct. 53-25 7.02 17-38 1-37 20.98 P. Ct. 52.57 6.74 17.39 p. a. 52.52 6.80 P.ct. 52.55 6.77 !7-39 P. Ct. 52.85 6.81 17.48 22.S6 P. Ct. 52.43 6.79 17-59 P.ct. 52.56 6.88 P. Ct. 52.50 6.84 17.59 P. Ct. 52.74 6.87 17-67 22.72 J o.54 0.46 Ash 0-95 100.00 100.00 100.00 These figures show that the protein extracted from the " shorts " has the same composition as that similarly obtained from the flour. 62 THE PROTEINS OF THE WHEAT KERNEL. Protein Extracted by Alcohol from Whole-wheat Flour. As Ritthausen and probably others used whole-wheat flour in extracting the various proteins described by them as soluble in dilute alcohol, it was thought best to carry out some extractions with meal obtained by grinding the entire wheat kernel in the laboratory. Preparations 83-88. Preparation S3. Preparation 84. I. II. Aver- age. Ash- free. I. II. Aver- age. Ash- free. Carbon Hydrogen... Nitrogen. . . Sulphur. . . . Oxygen .... P. ct. 52.88 7.00 17-47 i-43 P.ct. 52.68 6.97 17.48 P.ct. 52.78 6.98 17.48 1-43 P.ct. 52.90 6.99 I7.52 i-43 21.16 P.ct. 52.59 6.87 17.92 0.92 P.ct. 52-73 6.80 18.06 P.ct. 52.66 6.84 17.99 0.92 P.ct. 52.89 6.S7 18.06 0.92 21.26 Ash 0.23 o.45 - IOO.OO 100.00 Preparation 85. Preparation 86. I. II. Average. Ash-free. I. Ash-free. P.ct. 51-03 6-59 16.98 0.92 P.ct. 50.93 6.50 17.08 P.ct. 50.98 6.55 17.03 0.92 P.ct. 53-i6 6. S3 17-75 0.96 21.30 P.ct. 52.64 6.S6 17-49 I o.35 P.ct. 52.82 6.88 17-55 22.75 100.00 Hydrogen . . Nitrogen . . . Sulphur Oxygen Ash 4.11 4-cj 4.07 100.00 Preparation 87. Preparation 88. I. Ash- free. I. II. \verage. Ash-free. Carbon P.ct. 52.66 6.80 17.62 \ P.ct. 52.68 6.81 17-63 22.88 P.ct. 52-03 6.66 17.48 f 1.0S P.ct. 51-88 6.69 P.ct. 51-96 6.68 17.48 1.0S P. ct. 52.24 6.71 17-57 1.0S 22.40 Hydrogen Nitrogen .... Sulphur Oxygen .... 1 0.04 Ash , 0.52 100.00 100.00 EXPERIMENTAL. 63 One kilogram of whole spring- wheat flour, freshly ground, was made into a dough, and the gluten obtained from this by washing with water was then chopped fine and thoroughly extracted with alcohol of 0.90 sp. gr., the yellow extract concentrated, and the protein separated by cooling. The deposit thus produced was dissolved as far as possible in dilute alcohol, and the insoluble substance, which was coagulated protein, was washed with dilute alcohol, absolute alcohol, and ether. This was preparation 83. The solution filtered from 83 was poured into absolute alcohol and a small amount of protein separated ; this was treated with absolute alcohol and ether in the usual way, yielding preparation 84. The filtrate from 84 was concentrated to small volume and poured into absolute alcohol, whereby nearly all the protein was precipitated. This substance was dehydrated with absolute alcohol and digested with ether, giving preparation 85. These three bodies had the composition shown by the figures for preparations 83, 84, and 85 in the table on page 62. In a similar manner an extract was made of winter-wheat meal obtained by grinding the entire wheat kernel in the laboratory, the alcoholic extract concentrated to about one-third of its volume, cooled, and the solution decanted from the deposit. This was then dissolved in alcohol of 0.90 specific gravity and the coagulated protein filtered off, washed with dilute alcohol, digested with absolute alcohol and then with ether, giving prepara- tion 86. The solution filtered from this preparation was concentrated to small volume, cooled, and the protein separated was digested with absolute alcohol and with ether, yielding preparation 87. (See table on p. 62. ) The complete extraction of this protein from the gluten is very difficult, a little generally remaining in the insoluble residue after extracting with dilute alcohol. In one case the residue thus remaining was dissolved in 0.2 per cent potassium- hydroxide solution, and the resulting solution, after standing some time to deposit suspended impurities, was decanted and precipitated with dilute hydrochloric acid. This precipitate was washed by decantation with water and then digested for some time with dilute alcohol. The alcoholic solution was then filtered and concentrated to small volume and cooled. The protein separated was then digested with absolute alcohol and with ether and dried at 1 io° for analysis. From this analysis it is seen that the protein soluble in dilute alcohol is not changed in composition by solution with potassium hydroxide, nor is its solubility altered, so far as could be learned. In order to facilitate a comparison of these analyses they have been brought together in table 15. 64 THE PROTEINS OF THE WHEAT KERNEL. Table IS. — Composition of protein extracted by dilute alcohol. By direct extrac- tion After extraction with 10 per cent sodium - chlo- ride solution . From gluten spring - wheat flour From gluten whole spring wheat flour. From gluten whole winter wheat flour. . From whea "shorts".... Average of preceding figures Carbon. P.ct. 52.52 52.77 52.67 52.55 52.74 52.39 52.84 53-02 52.33 52.38 52.69 52.72 52.71 52.65 52.58 52.68 52.84 52.90 52.89 53- 16 52.24 52.82 52.68 53-25 52.85 52.74 52.72 Hydrogen. P.ct. 6.78 6.78 6.70 6.85 6-77 6-93 6.S1 6.79 6.91 7-13 6.84 6.86 6.81 6.83 6.67 6.7S 7.18 6.99 6.87 6.83 6.71 6.S8 6.S1 7.02 6.81 6.8- 6.86 Nitrogen. Sulphur. Oxygen. Total. P.ct. 17.18 17.26 17.64 17.77 17.66 17-94 17.62 17-31 17.67 17.32 17.69 17.70 17.82 17-73 17.89 17-75 17.79 17-65 17-65 17-57 17-52 18.06 17-75 17-57 17-55 17-63 I7-38 17.48 17.67 17.66 P.ct. 1.08 1.26 1.22 1. 21 I.23 1.38 I. II I.05 P.ct. 21.98 21.42 21.75 21-45 21.64 21.99 21.57 21.82 23.06 22.67 1.02 0-95 1. 10 1.08 ,08 .09 21.72 21.58 21.63 21.65 22.02 21.80 22.41 1-43 0.92 0.96 1.0S 21.16 21.26 21.30 22.40 22.75 22.8S 1-37 22.86 22.72 20.98 1. 14 21.62 P.ct. 100 100 loo IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO IOO 1 Omitted in average. THE PROPORTION OP GLUTAMINIC ACID YIELDED BY VARIOUS FRACTIONS OF THE ALCOHOL-SOLUBLE PROTEIN. The results of this extensive fractionation of the alcohol-soluble protein give no evidence, based on the ultimate composition of the many fractions, of the presence of more than one protein substance. This, however, does not justify the conclusion that only one such substance exists. In view of the recent positive statements of Kossel & Kutscher and of Kbnig & Rintelen it is important to obtain, if possible, further evidence based on the propor- tion of some of the decomposition products yielded by the various fractions of this protein. As gliadin yields a relatively large amount of glutaminic EXPERIMENTAL. 65 acid when decomposed by boiling with strong hydrochloric acid, determina- tions of the quantity of glutaminic acid obtained from various fractions have been made. The first difficulty encountered in examining the evidence that has been offered respecting the existence of several alcohol-soluble proteins in wheat flour lay in the impossibility of following Ritthausen's directions for pre- paring these, since many evidently important details are omitted in the description of his methods. Kossel & Kutscher state that their products were made according to Ritt- hausen's directions, but give no details, nor do they state which method they employed. Kutscher concludes another paper by the statement that ' ' the wheat gluten consists of gluten-casein wholly insoluble in cold 60 per cent alcohol; gluten-fibrin, but little soluble, and gliadin, easily soluble in cold 60 per cent alcohol." Although the writer has made a very large number of preparations repre- senting fractions of this protein substance dissolved by alcohol of various degrees of strength, he has never obtained any that were not either com- pletely soluble in cold alcohol of 60 per cent by volume, or else contained such insignificant quantities which did not dissolve that he has found it impossible to make from them a preparation of ' ' gluten-fibrin ' ' suitable for further examination. He has, therefore, been unable to repeat the work of Kossel & Kutscher and is entirely at a loss to understand how their preparation of " gluten-fibrin " was obtained. Konig & Rintelen describe their procedure in more detail. These inves- tigators extracted wheat gluten with absolute alcohol, added ether to the alcoholic extract, and united the precipitate produced with the extracted gluten. This latter was then extracted with 65 per cent alcohol, and to the extract alcohol was added until the mixture contained 88 to 90 per cent of alcohol. After decanting from the precipitate that had formed, the solution was filtered clear and evaporated to dryness on the water-bath, finely pulverized, and extracted with ether to remove fat. As all of the fat could not be thus removed, the mass was again dissolved in alcohol, to which some potassium hydroxide was added, and this solution shaken out several times with ether. The weakly alkaline solution was then exactly neutralized with hydrochloric acid and evaporated on the water-bath. The product thus obtained was their "gluten-fibrin." The precipitate produced by 88 to 90 per cent alcohol was washed with alcohol of the same strength and dissolved in a little 65 per cent alcohol. From this one-half of the alcohol was distilled off and the residual solution cooled, when a precipitate separated. From this the solution was decanted, leaving a mass of gliadin. The solutions which remained from several such 5 66 THE PROTEINS OF THE WHEAT KERNEL. precipitates were united and distilled until one-third of the solvent was removed. On cooling the residual solution, a deposit formed which they considered to be a mixture of gliadin and mucedin. The solution decanted from this deposit was evaporated to dryness, and yielded a considerable residue of " mucedin." The analyses of these products showed that the gliadin thus prepared had the same composition as that obtained by the writer, as well as that made by Ritthausen, while the "gluten-fibrin" and mucedin contained about i per cent less nitrogen and much more carbon. Konig & Rintelen obtained their mucedin from the nearly aqueous solu- tions remaining after separating the gliadin by evaporating to dryness. The writer assumes that the residue which remained was subjected to some fur- ther purification, but concerning this they say nothing. The solution con- tains many impurities, and when the protein substance in it has been properly purified it has the properties and composition of gliadin. Although large quantities of gliadin have been made in this laboratory and subjected to very careful and extensive fractionation, no evidence what- ever of the existence of " mucedin " has been obtained. Furthermore, it would seem improbable that our gliadin could be con- taminated by " gluten-fibrin " and " mucedin," which the writer certainly did not succeed in separating from it, and at the same time show so close an agreement in composition with that of Konig & Rintelen, from which they suppose that both of these proteins had been carefully removed. It is also improbable that Kjeldahl should have found the specific rotation uniform for numerous preparations of the alcohol-soluble protein if the material which he examined was a mixture of three different substances ; nor, if this were the case, could his determination of («)D — 92 ° be expected to agree 20° so closely with that made by Osborne & Harris,1 («) -=: 92. 30, and by Mathewson,2 («) -° 9i-95°- The composition of both "gluten-fibrin" and "mucedin" differs from that of gliadin just as one would expect if these former substances were slightly altered, and somewhat impure products obtained from gliadin. Until more convincing evidence of the existence of " gluten-fibrin " and "mucedin" as distinct protein substances is brought forward, they can not be considered to be original constituents of the wheat kernel. Kutscher has stated that there is a wide difference iu the proportion of glutaminic acid which is yielded by gliadin and gluten-fibrin. The writer 1 Osborne & Harris, Journal American Chemical Society, 1903, xxiv, p. 844. 2 Mathewson, Journal American Chemical Society, 1906, xxxviii, p. 1482. EXPERIMENTAL. 67 has therefore determined the amount of this amino-acid which was yielded by several preparations of the alcohol-soluble protein, one of which repre- sented the fraction of the protein soluble in the strongest alcohol, and should therefore, according to Ritthausen, consist chiefly of gluten-fibrin. The amount of glutaminic acid in gliadin was first determined by boiling 100 grams of gliadin, equal to 93 grams moisture free, for 14 hours with 200 cc. of concentrated hydrochloric acid. After standing on ice for 3 days, the entire solution solidified to a thick mass of crystals, which was sucked out with a pump and washed with ice-cold alcoholic hydrochloric acid. When dried over sodium hydroxide, this crude glutaminic acid hydrochloride weighed 58.33 grams. The filtrate and washings on concentration gave by similar treatment 2.58 grams more, making the total crude glutaminic acid hydrochloride 60.01 grams. This product was then dissolved in water freed from color with animal charcoal and recrystallized. After removing ammo- nium chloride by boiling with a slight excess of barium hydroxide and the barium with an equivalent quantity of sulphuric acid, 43.02 grams of pure glutaminic acid hydrochloride were obtained, which are equal to 34.42 grams of the free acid, or 37 per cent of the gliadin. Nitrogen: 0.5092 gram substance, dried at ioo°, gave NH3 = 3.86 HC1 (1 cc. HC1 = 0.01 gram N) = 7.58 p. ct. N. Calculated for C5HI0O4NCl, 7.64 p. ct. N. In confirmation of these figures this determination was repeated with two fractions of the alcohol-soluble protein of wheat gluten which had been separated from relatively strong alcoholic solutions, and should therefore have contained a large proportion of "gluten-fibrin" if the statements respecting the solubility of this substance are correct. Two portions of different preparations of the air-dry substance, equivalent to 18.62 and 14.65 grams dried at no°, were hydrolyzed as before, their solutions saturated with hydrochloric acid, and kept for some time on ice. The glutaminic acid hydrochloride which separated was filtered out, washed with alcoholic hydrochloric acid, and freed from ammonia by evaporating with an excess of baryta and from barium by an equivalent quantity of sulphuric acid. The solution was then decolorized with animal charcoal and evaporated with an excess of hydrochloric acid until crystallization began. The glutaminic acid hydrochloride which separated, when washed with ice-cold alcoholic hydrochloric acid and dried, weighed, respectively, 8.69 and 6.27 grams, equivalent to 37.33 and 34.2 percent of free glutaminic acid in the protein. Another attempt was made to isolate a fraction of ' ' gluten-fibrin ' ' by dissolving 200 grams of a preparation representing the total alcohol-soluble protein of wheat gluten in a mixture of 900 cc. of absolute alcohol and 68 the; proteins oe the wheat kernel. 600 cc. of water — that is, in alcohol of 60 per cent by volume. Although the solution was somewhat turbid, nothing was deposited, even on long standing. 300 cc. of water were then added to the solution, making the alcohol 50 per cent, but still nothing separated. The strength of the alcohol was therefore raised to 75 per cent by adding 1800 cc. of absolute alcohol, and a very large precipitate, A, at once separated. By adding 2000 cc. of absolute alcohol to the clear solution from which A separated another large precipitate (B) was produced, and in the filtrate from B a third precipitate (C) resulted, when a further large quantity of absolute alcohol was added. This last product weighed 20 grams and constituted only 10 per cent of the total protein. The solution from which C separated contained only traces of protein, and C, therefore, represented the fraction of the whole protein soluble in the strongest alcohol, and should consequently contain much "gluten-fibrin." In 17.5 grams of this substance dried at no° the gluta- minic acid produced by decomposing with hydrochloric acid was determined. By proceeding in the same manner as in the experiment last described 7.8914 grams of pure glutaminic acid hydrochloride were obtained, which are equiv- alent to6.2i3i grams of the free acid, or 35.5 per cent. Nitrogen: 0.7296 gram substance, dried at no0, gave NH3 = 5.66 cc. HC1 (1 cc. HC1 = 0.0100 gram N) = 7.75 p. ct. N. Calculated for C5H9N04HC1, 7.64 p. ct. N. This result seems to furnish conclusive evidence that there is no fraction, soluble in very strong alcohol, to be obtained from the alcohol-soluble pro- tein of wheat gluten that is characterized by yielding a relatively small proportion of glutaminic acid. Since Kutscher decomposed his proteins with sulphuric acid, while hydro- chloric acid was used in the preceding experiments, the following experi- ment was made in order to determine whether the higher yield obtained by us might not be due to this fact. Accordingly 50 grams of one of the preparations of gliadin, from which 37 per cent of glutaminic acid had pre- viously been isolated, were boiled for 14 hours with a mixture of 150 grams of sulphuric acid and 300 cc. of water. The resulting solution was treated with an excess of baryta and the ammo- nia expelled by evaporation. The barium was then removed by an equiv- alent amount of sulphuric acid and the filtered solution evaporated. Some tyrosine separated, which was filtered out, and the evaporation continued until the volume was quite small. On standing, an abundant quantity of crystals of free glutaminic acid separated, and from the mother-liquor, by further concentration and standing, a second crop of crystals was obtained. After recrystallizing several times, 8.48 grams of pure glutaminic acid were obtained, which contained 9.43 per cent of nitrogen. EXPERIMENTAL. 69 Nitrogen; 0.6094 gram substance, dried at no0, gave NH3 = 5.75 cc. HC1 (1 cc. HC1 = 0.0100 gram N) =9.43 p. ct.; calculated for C5H9N04) 9.48 p. ct. From the mother-liquors which yielded this glutaminic acid there were further separated, by saturating with hydrochloric acid and proceeding in the manner already described, 4.055 grams of well-crystallized hydrochloride, which contained 7.73 per cent of nitrogen. Nitrogen: 0.4983 gram substance, dried at no°, gave NH3 = 3.85 cc. HC1 (1 cc. HC1 = 0.0100 gram N) = 7.73 p. ct. N ; calculated for C3H9N04HC1, 7.64 p. ct. N. The free glutaminic acid corresponding to this amount of the hydro- chloride is 3.244 grams, making a total of 1 1.724 grams. Since the 50 grams of air-dry gliadin were equivalent to 46.33 grams dried at no°, this amount of glutaminic acid is equal to 25.3 per cent, or about the same proportion as Ritthausen found in his preparation of ' ' mucedin ' ' after decomposing with sulphuric acid, but much more than the 19.81 per cent found by Kutscher in the same substance. The amount, however, was only about two-thirds as much as that found after decomposing with hydrochloric acid, and although the separation was not complete there was no reason to suppose that more remained in the mother-liquors in one case than in the other. It is possible, however, that the very large precipitate of barium sulphate that formed on removing the sulphuric acid retained a considerable part of the glutaminic acid even after extensive washing with hot water. In conclusion, the results of these determinations are here brought together that they may be more readily compared : Table 16. — Percentage of glutaminic acid yielded by gliadin. Preparation. Decomposed by — HG1. H2S04. 1 P.ct. 37.00 37-33 34.20 35-5Q P.ct. 25-3 2 1. 4 s All these are minimal figures, since in each case some glutaminic acid still remained in the mother-liquors, but it does not seem probable that more than relatively insignificant quantities were thus lost. From these results it would appear that Kutscher' s determinations of glutaminic acid fall far short of the actual quantity of this substance yielded by the alcohol-soluble protein of wheat, and that they therefore afford no /O the; proteins of the wheat kernel. evidence which justifies the conclusion that this substance consists of two distinct protein bodies. Fractional precipitations of this alcohol-soluble protein yield practically the same large proportion of glutaminic acid, so that, in view of their very close agreement in composition and properties, both physical and chemical, there is every reason to believe that only one such protein is present. Gliadin yields a remarkable proportion of gluta- minic acid, much in excess of that from any other known protein and greater than that of any single decomposition product yet obtained in a pure state from any other true protein substance, the protamines, of course, excepted. HYDROLYSIS OF GLIADIN. There being no sufficient evidence that more than one alcohol-soluble protein occurs in the wheat kernel, no attempt has been made, in preparing large quantities of gliadin for the present investigation, to subject the pro- tein matter extracted by alcohol to any fractional precipitation, but it was separated as completely as possible from all other substances soluble in water, alcohol, and ether. The gliadin for this investigation was prepared entirely from gluten, as thereby the water-soluble constituents of the seed are more completely re- moved than by any other method of preparation which can be readily used on a large scale. The wheat flour was kneaded into dough in a domestic "bread-mixer," and then under water in a specially constructed kneading- machine. After frequently decanting and renewing the water, a thoroughly coherent gluten was obtained. This was washed practically starch free in a current of water and, while moist, was ground by passing through a special form of "drug-press," which was a ready means of reducing it to comparatively small pieces. The ground gluten was then extracted with alcohol of such strength that, with the combined water of the gluten, a solvent of 60 to 70 per cent by volume resulted. The extracts were filtered perfectly clear through thick felts of filter paper pulp, and the water-clear solution, free from any trace of opalescence or turbidity, was evaporated to a small volume on a water-bath. The thick sirup that resulted was cooled and then poured, with constant and rapid stirring, into a large volume of distilled ice-water containing a very little sodium chloride. The gliadin was thus precipitated as a filament, which, on stirring, united to a coherent plastic mass. This gliadin was next dissolved by stirring with strong alcohol until all had gone into solution, the water combined with the precipitated gliadin being sufficient to dilute the alcohol to the proper degree. The resulting solution was evaporated to a thick sirup, absolute alcohol being added from time to time in order to hold the gliadin in the solution, since this, during the evaporation, became constantly more aqueous. The thick EXPERIMENTAL,. 71 sirup was then poured, in a very fine stream, into a large volume of absolute alcohol under rapid and constant stirring. In this way a porous mass of protein was obtained which was at once reduced to small pieces and digested under fresh absolute alcohol. When well dehydrated, the gliadin was digested with ether, partially dried over sulphuric acid, ground to a coarse powder, and then dried thoroughly over sulphuric acid. When thus pre- pared, gliadin forms a snow-white, friable mass which is easily reduced to a powder. 1 100 grams of gliadin, equal to 998.6 grams dried at noc, were heated with a mixture of 1000 cc. of concentrated hydrochloric acid and 1000 cc. of water on the water-bath for several hours, until the gliadin had dissolved and frothing had ceased. The solution was boiled in an oil-bath having a temperature of 1 15° for 10 hours, cooled with ice, and saturated with gaseous hydrochloric acid. After remaining on ice for 2 days, the glutaminic acid hydrochloride that had separated was filtered out, washed with ice-cold alcoholic hydrochloric acid dissolved in water, the solution treated with bone-black and freed from ammonia by boiling with an excess of barium hydroxide. After removing the barium with an equivalent amount of sul- phuric acid, the glutaminic acid was separated as hydrochloride, and, when recrystallized and thoroughly dried, weighed 374.3 grams, equivalent to 300 grams of free glutaminic acid. This was converted into the free acid, which melted at 2020 to 2030. Carbon and hydrogen : 0.6218 gram substance gave 0.9276 gm. C02 and 0.3494 gtn. H20. Nitrogen : 0.4574 gram substance gave NH3 = 4.3 cc. HC1 (1 cc. HCl = o.oi gram N). Calculated for C5H904N,C 40. 78, H 6. 18, N 9.54 p. ct. ; found, C 40. 69, H 6.24, N 9.40 p. ct. The mother-liquor from the recrystallized glutaminic acid was added to the filtrate from the first separation of the glutaminic acid, and the entire solution concentrated to a sirup under strongly reduced pressure. 3 liters of alcohol, previously saturated with hydrochloric acid at a low temperature, were added to the sirup, and dry hydrochloric acid gas was passed into the solution until it was saturated. The mixture was again concentrated as before, under reduced pressure, the sirup again taken up in 3 liters of alco- holic hydrochloric acid, and after standing several hours again concentrated to a sirup, taken up a third time in alcoholic hydrochloric acid, and after some hours concentrated to a sirup on a bath of 400 and under a pressure of from 5 to 10 mm. The neutralization, extraction, and drying of the esters were conducted according to the method described by Emil Fischer.1 The undistilled residue, after distillation A, weighed 180 grams. The residue which remained after extracting the esters with ether was made strongly acid with hydrochloric acid, freed from sodium and potassium Fischer, Emil, Zeitschrift fiir physiologische Chemie, 1901, xxxiil, p. 151. 72 THE PROTEINS OF THE WHEAT KERNEL. salts by repeated evaporations with alcoholic hydrochloric acid and thor- ough extraction of the precipitated chlorides with the latter. The alcoholic solution of the chlorides of the amino-acids was evaporated to a sirup, and esterificatiou repeated as in the first instance. After extracting the esters with ether and drying them, the entire process was repeated, and the ethe- real solution of the esters resulting from this third treatment was united with that from the second, thus following the method applied by Abderhalden to oxyhemoglobin.1 Distillations A and B. Fraction. Distillation A. Distillation B. Temper- ature of bath up to — Vapor. Press- ure. Weight. Temper- ature of bath up to — Press- ure. Weight. I II Ill IV o 93 IOO 1 20 160 0 75-76 vim. 12.0 12.0 0.8 0.8 Grams. 28.18 4703 64.68 40.00 0 83 100 120 200 mm. 12.0 12.0 0.8 0.8 G> ams. 20.13 36 s. 62.70 33 °° 179.89 152.64 Fraction I. Distillation A. — This was saponified at once by evaporation with con- centrated hydrochloric acid on a water-bath, the residue taken up in alcohol, the solution saturated with dry hydrochloric acid gas, and a crystal of glycocoll ester hydrochloride added. After prolonged standing on ice, no separation occurred. The solution was then evaporated on the water- bath with concentrated hydrochloric acid, the latter removed with lead oxide and the lead with hydrogen sulphide. The amino-acids were subjected to fractional crystallization. Distillation B was treated in substantially the same way, but although several attempts were made to isolate the hydrochloride of glycocoll ester, none was found. By systematic fractionation there were obtained from fraction I of the two distillations, A and B, 6.68 grams alanine and 0.S6 gram leucine. The leucine, when recrystallized from dilute alcohol, decomposed at about 2980. Carbon and hydrogen: 0.1778 gram substance, dried at 1100, gave 0.3577 gram C02 and 0.160S gram H20. Calculated for C6H1302N, C 54.89, H 10.01 p. ct.; found, C 54.87, H 10.04 p. ct. 'Abderhalden, Zeitschrift fur physiologische Cheinie, 1903, xxxvn, p. 484. EXPERIMENTAL. 73 The alanine, when recrystallized by dissolving in a little hot water and gradually adding alcohol, decomposed at about 2900. Carbon and hydrogen: 0.2404 gram substance, dried at no°, gave 0.3571 gram CO,, and o. 1712 gram H20. Nitrogen: 0.3980 gram substance gave NH3 = 6.2 cc. HC1 (1 cc. HCl = o.oi gram N). Calculated for C3H702N, C 40.40, H 7.93, N 15.75 P- ct.; found, C 40.51, H 7.91, N 15.58 p. ct. ( Fraction II. Temperature of bath up to ioo°. (Pressure, 12 mm. Weight, 83.84 grams. Distillations A and B. — Each was saponified by boiling for 5^ hours with 5 parts of water. The solution was evaporated to dryness under reduced pressure, the dried residue boiled up with absolute alcohol, and 13.37 grams were dissolved. This solution was united with a similar one obtained from fraction III. The substance, insoluble in alcohol, after systematic fractional crystalliza- tion, gave — (I) 22.1 grams of leucine, which decomposed at about 2980 and had the following composition : Carbon and hydrogen : 0.3128 gram substance, dried at no0, gave 0.6310 gram C02 and 0.2794 gram H20. Calculated for C6H1302N, C 54.89, H 10.01 p. ct.; found, C 55.01, H 9.92 p. ct. (II) A fraction of 3.45 grams which by fractional crystallization could not be further separated and gave results on analysis which agreed best for a mixture of leucine and amino-valerianic acid. Carbon and hydrogen : 0.2857 gram substance, dried at no0, gave 0.5519 gram C02 and 0.2577 gram H20. Calculated for equal molecules of leucine and amino-valerianic acid, C 53.05, H 9.74 p. ct. ; found, C 52.68, H 10.02 p. ct. (III) 2.1 grams substance which had the properties and composition of amino-valerianic acid. Carbon and hydrogen: (I) 0.2115 gram substance, dried at 110°, gave 0.3957 gram CO, and 0.1870 gram H20 ; (II) 0.4597 gram substance, dried at no0, gave 0.8643 gram CO, and 0.3956 gram H20. Calculated for C5Hu02N, C 51.22, H 9.48 p. ct; found, (I) C 51.03, H 9.82 p. ct.; (II) C 51-27, H 9.56 p. ct. 20° Specific rotation. — Dissolved in 20 per cent hydrochloric acid, (a) — — = + 25. 790. E. Fischer and Dorpinghaus1 found + 25. 90 for their preparation from horn, and Schulze and Winterstein* found +28.20 and + 27.90 for preparations from lupine seedlings. 1 Fischer & Dorpinghaus, Zeitschrift fur physiologische Chemie, 1902, xxxvi, p. 462. 2 Schulze & Winterstein, ibid., 1902, xxxv, p. 300. 74 T II Ii PROTE] -S OF THE WHEAT KERNEL. The chlorine was removed from the solution used for determining the specific rotation and the substance racemized by heating with 20 cc. of water and 7 grams of crystallized barium hydroxide for 19 hours in an autoclave at 1750. The barium was quantitatively removed with sulphuric acid, and the «-naphthyl-hydantoic acid prepared according to the directions of Neuberg & Manasse.1 This crystallized in long needles and melted constantly on repeated recrystallization from 40 per cent alcohol at 1800 to 1810. Carbon and hydrogen: 0.3235 gram substance, dried at 900, gave 0.7936 gram C02 and 0.1909 gram H20. Calculated for C16H1803N2, C 67.07, H 6.35 p. ct. ; found, C 66.90, H 6.56 p. ct. By racemizing the remaining mixture of undetermined amino-acids we were unable to isolate any more amino-valerianic acid. (IV) 8.6 grams alanine. This was racemized by heating with an excess of barium hydroxide and coupled with a-naphthyliso-cyanate according to the directions of Neuberg & Manasse. The hydantoic acid, which crystal- lized in prisms, melted at 1970. Carbon and hydrogen: 0.3146 gram substance, dried at no°, gave 0.7480 gram C02 and 0.1560 gram H20. Calculated for CuH1403N2, C 65.06, H 5.48 p. ct.; found, C 64.84, H 5.51 p. ct. f Fraction III. Temperature of bath, up to 1200. ) \ Pressure, 0.8 mm. Weight, 127.38 grams. j This fraction was boiled for 5^2 hours with 8 parts of water. The solution, evaporated to dryness under reduced pressure, gave 98 grams of amino-acids or 80 per cent of the esters. Of this 59.94 grams were soluble in alcohol. From the part insoluble in alcohol, by systematic fractional crystallization, there were isolated 33.06 grams of leucine and 4.57 grams of alanine. The leucine decomposed at about 2980. Carbon and hydrogen: 0.2416 gram substance, dried at no0, gave 0.4872 gram C02 and 0.2166 gram H,0. Nitrogen : 0.2690 gram substance gave NH3 = 2.85 cc. HC1 ( 1 cc. HC1 = 0.01 gram N). Calculated for C6H1302N, C 54.89, H 10.01, N 10.70 p. ct ; found, C 54-99, H 9.96, N 10.59 P- ct- The alcohol- soluble substance from fraction II was united with that from fraction III. The solution was evaporated to dryness under reduced press- ure, and the residue taken up in water and boiled gently for about an hour 1 Neuberg & Manasse, Berichte der deutscben chemiscben Gesellschaft, 1905, xxxvni, P- 2359- EXPERIMENTAL. 75 with an excess of copper hydroxide. The filtered solution was evaporated to dryness under reduced pressure and the residue boiled with absolute alcohol. The undissolved part was dissolved in water, freed from copper by hydrogen sulphide, and the solution again evaporated to dryness under reduced pressure. The residue was boiled with absolute alcohol, in which all of it dissolved. The alcohol was evaporated off under reduced pressure, the residue dissolved in 500 cc. of water and again converted into the copper salt. By concentration 19.91 grams of crystalline racemic proline copper salt were obtained, which is equal to 13.98 grams of a-proline. This was recrystallized from water and dried in the air. Water : 0.7789 grain substance lost 0.0856 gram H20 at no°. Calculated for C10Hi6O4N2 Cu ■ 2H20, H20 11.00 p. ct. ; found, H20 10.98 p. ct. Carbon and hydrogen : 0.6S51 gram substance, dried at no°, gave 1.0310 gram C02 and 0.3450 gram H20. Copper: 0.2926 gram substance gave 0.0799 gram CuO. Calculated for C10H16O4N2 Cu, C 41. 11, H 5.54, Cu 21.79 P- ct \ found, C 41.04, H 5.59, Cu 21.81 p. ct. The r-proline copper salt was freed from copper with hydrogen sulphide, its solution evaporated to dryness, and the residue recrystallized from alcohol. After drying in vacuo over sulphuric acid the r-proline melted at 203 ° to 205V Carbon and hydrogen: 0.3373 gm. substance gave 0.6424 gin. C02 and 0.2453 gm- H20. Calculated for C5H902 N, C 52.12, H. 7.90 p. ct. ; found, C 51.94, H 8.08 p. ct. The solution of the alcohol-soluble copper salt was evaporated to dry- ness and left a residue of /-proline copper salt which, dried at 1200, weighed 71.62 grams, which is equal to 56.51 grams of free /-a-proline. Copper: 0.2850 gram substance, dried at 110°, gave 0.0760 gram CuO. Calculated for Ci0Hi6O4N2 Cu, Cu 21.79 P- ct- ; found, Cu 21.30 p. ct. One-half of this proline copper salt was freed from copper, and the pro- line racemized by heating with 150 cc. of water containing 80 grams of crystallized barium hydroxide for 5 hours at 1500. The barium was re- moved quantitatively with sulphuric acid, the solution concentrated, and the proline again converted into the copper salt. There was thus obtained 20. 1 grams of very nearly pure racemic a-proline copper. Water : 0.3356 gram substance, air-dry, lost 0.0369 gram H20 at 1100. Copper : 0.2640 gram substance gave 0.0640 gram CuO. Calculated for C10H16O4N2 Cu " 2H20, H20 11.00, Cu 19.40 p. ct. ; found, H20 11.00, Cu 19.35 p. ct. 1 Willstaeter, R., Berichte der deutschen chemischen Gesellschaft, 1900, xxxiii, p. 1160; Fischer, Emil, ibid., 1901, xxxiv, p. 458. 76 THE PROTEINS OE THE WHEAT KERNEL. The total crystalline raceraic copper salt was equal to 41.96 grams of r-"-proline. From the other half of /-proline copper salt the free proline was regen- erated and recrystallized from alcohol. A small part only was obtained in a crystalline condition, which melted at 2050 to 2060. Carbon and hydrogen : 0.2729 gram substance, dried over HaS04, gave 0.5198 gram C02 and 0.1977 gram K.O. Nitrogen : 0.2453 gram substance gave NH3 = 2.95 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for C5H902N, C 52.12, H 7.90, N 12.20 p. ct. ; found, C 51.95, H 804, N. 12.03 P- ct- From fraction III there were isolated 4.57 grams alanine, 33.06 grams leucine, and 70.49 grams a-proline, including in this last that from frac- tion II, which was not weighed separately. Fraction IV. Temper- ature of bath to — Press- ure. Weight. A 0 160 200 mm. 0.8 0.8 Grams. 39-96 33-oo B 72.96 This was treated with water and shaken out with ether according to the procedure described by Emil Fischer.1 The ether was carefully removed by evaporation and the residual phenyl- alanine ester saponified by dissolving in concentrated hydrochloric acid and evaporating on a water-bath. The phenylalanine hydrochloride weighed 29.14 grams, equivalent to 23.87 grams of free phenylalanine. The phenyl- alanine hydrochloride was recrystallized from strong hydrochloric acid. It was decomposed by evaporating with an excess of ammonia and the phenylalanine recrystallized from water. It melted, on slow heating, at 2630 to 265V Carbon and hydrogen : 0.3051 gram substance, dried at no°, gave 0.7322 gram C02 and 0.1792 gram HaO. Nitrogen : 0.3020 gram substance gave NH3 = 2.53 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for C9HnOsN, C 65.40, H 6.73, N 8.50 p. ct. ; found, C 65.44, H 6.53, N 8.38 p. ct. 1 Fischer, E., Zeitschrift fur physiologische Chemie, 1902, xxxvi, p. 274. 'Fischer, Emil, & Abderhalden, Zeitschrift fiir physiologische Chemie, 1902, xxxvi, p. 268 ; Erlenmeyer & L,ipp, Annalen der Chemie, 1883, ccxix, p. 197. Experimental,. yy The aqueous layer was heated with an excess of barium hydroxide on a water-bath for 5 hours. After standing some time, the barium salt that had separated was filtered out and decomposed by an equivalent amount of sul- phuric acid. The solution on concentration yielded 5.76 grams of aspartic acid. Carbon and hydrogen : 0.3866 gram substance, dried at no°, gave 0.5109 gram C02 and 0.1852 gram H20. Nitrogen : 0.3637 gram substance gave NH3 = 3.78 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for QH704N, C 36.09, H 5.26, N 10.53 P- ct- ; found, C 36.04, H 5.32, N 10.39 P- ct. The filtrate from the barium aspartate was freed from barium, concen- trated to small volume, and saturated with hydrochloric acid. On prolonged standing a trace of phenylalanine hydrochloride separated, but no glutaminic acid hydrochloride was obtained. After removing the hydrochloric acid with silver sulphate and the sulphuric acid with barium hydroxide, the solution was boiled with an excess of copper hydroxide, but no copper salt could be separated from it, even after concentrating to a very small volume. The copper was then removed and the solution treated with bone-black and, when concentrated in a vacuum over sulphuric acid, gave crystals which, on fractional crystallization from water, gave 0.42 gram of serine, which, in an open capillary, browned at about 2180 and decomposed to a brownish mass at about 240. ° Carbon and hydrogen: 0.2577 gram substance, dried at no°, gave 0.3224 gram C02 and o. 1602 gram H20. Calculated for C3H703N, C 34.24, H 6.73 p. ct. ; found, C 34.12, H 6.91 p. ct. In the filtrate from the serine there was obtained about 5.32 grams of crys- talline substance, from which nothing definite could be isolated. Distillation Residue. The residues from distillations A and B were dissolved in boiling alcohol and the solutions united. On cooling, 5.76 grams of long hair-like crystals separated. The filtrate from this substance was freed from alcohol and saponified by heating with 200 grams of crystallized barium hydroxide, the barium re- moved, the solution concentrated under reduced pressure to small volume, saturated with hydrochloric acid, and, after standing on ice for some time, yielded 74.21 grams glutaminic acid hydrochloride, equal to 59.46 grams.1 The glutaminic acid hydrochloride melted at about 1980 with effervescence. Carbon and hydrogen : 0.4028 gram substance, dried over H2S05, gave 0.4802 gram C02 and 0.2056 gram H20. Calculated for C5H10O4 NCI, C 32.67, H 5.50 p. ct. ; found, C 32.51, H 5.67 p. ct. 'Abderhalden & Wells, Zeitscbrift fiir physiologische Chemie, 1905, xi/vi, p. 31. /S Till- PROTEINS OF THE WHEAT KERNEE. Specific rotation. — Dissolved in 20 per cent hydrochloric acid, (a) -=- = + 31.47°. Fischer & Dorpinghaus found +31.910 for a preparation from horn, + 30. 450 from gelatin, and +28.200 for one from casein.1 The residue which remained after removing the esters with ether from the original solution of the products of hydrolysis was treated in the way described by Emil Fischer2 for the isolation of oxy-proline. The only substance, however, that could be isolated was serine, of which 0.87 gram was obtained, which browned at about 2190 and decomposed at about 2400. Carbon and hydrogen: 0.29S7 gram substance, dried at no0, gave 0.3743 gram C02 and 0.1844 gram H20. Calculated for C3H-03N, C 34.24, H 6.73 p. ct. ; found, C 34. iS, H 6.S5 p. ct. In the filtrate from this serine ,3-naphthalene-sulphone chloride failed to give any definite product. There were thus isolated in the two distillations from fraction I, 6.68 grams ; from fraction II, 8.59 grams, and from fraction III, 4.57 grams of alanine, or 19.84 grams in all; from fraction II, 2.1 grams of amino-vale- rianic acid ; from fraction I, 0.86 gram ; from fraction II, 22. 10 grams ; from fraction III, 33.06 grams of leucine — in all 56.02 grams ; from fractions II and III, 70.49 grams of proline ; from fraction IV, 23.47 grams of phenyl- alanine, 5.76 grams of aspartic acid, and 0.42 grams of serine, and from the residue which remained after extracting the esters 0.87 gram of serine, or 1.29 grams in all. From the main solution of the total products of hydrolysis there were obtained 300 grams of glutaminic acid, and from the residues after dis- tilling the esters 59.5 grams, or in all 3~9-5 grams. As the amount of proline found in this hydrolysis was so great, the result was confirmed by a second hydrolysis, and another effort made to obtain glycocoll, which, if present in very small amount, might have escaped detection. For this purpose 500 grams of gliadin, air-dry, equal to 439.6 grams dried at no°, were hydrolyzed in the same way as in the preceding h}rdrolysis. After esterifying and shaking out the esters three times, as before, the ether was removed by distillation at 760 mm. and the esters distilled. 'Cf. Fischer, Emil, & Dorpinghaus, Zeitschrift fiir physiologische Chemie, 1902, xxxvi, p. 475. ' Fischer, Emil, Berichte der deutschen chemischen Gesellschaft, 1902, xxxv, p. 2660 EXPERIMENTAL. 79 Tempera- - ,. ture of Fraction. bath up to — Pressure. Weight. I o 95 8o no no 180 200 mm. Grams. I7.OO 20.20 5.OO 46.08 2.00 44-14 O.88 15.70 O.88 54.7 O.78 29.69 II Ill \ A IV V 2IO. 51 Fraction I was immediately evaporated on the water-bath with hydro- chloric acid, the residue dissolved in alcohol, and the solution saturated with dry hydrochloric acid gas. The solution was concentrated to a small volume at a low temperature under a pressure of 10 mm., the residue taken up in alcohol, its solution cooled to o° and saturated with hydrochloric acid gas. On prolonged standing 0.22 gram of glycocoll ester hydrochloric separated, which melted at 1440 to 1450. When mixed with pure glycocoll ester hydrochloride, the melting-point was unchanged. Chlorine : 0.1058 gram substance gave 0. 1063 gram AgCl = CI 2X Calculated for C4HmO,NC1, CI 25.40 p. ct Neither fraction II nor the ether distilled from the esters gave evidence of glycocoll. This preparation of gliadin did, in fact, contain a very small amount of glycocoll, which is possibly due to a slight contamination with glutenin, in which there has since been found a notable quantity of this amino-acid. Fraction II was saponified, and the solution evaporated under highly reduced pressure from a bath at 400, and the residue extracted with alcohol, in which about 18 grams dissolved. Fraction III, by similar treatment, yielded 24 grams of alcohol-soluble substance. The alcoholic solutions were united and evaporated to dryness from a bath at 40c. The crystalline residue, when dried to constant weight in vacuo, weighed 39.59 grams. From this, by extraction with alcohol, 8.7 grams of substance insoluble therein were separated.1 The total proline thus found was 30. S9 grams, equal to 7.03 per cent. 1 Emil Fischer employs this method for estimating the proportion of pro-ine in pro- teins, but stated that the result obtained is too high ^Berichte d*er deutschen chemischen Gesellschaft, 19:6, xxxrx, p. 530). 80 THE PROTEINS OF THE WHEAT KERNEL. CYSTINE. 300 grams of gliadin were digested for 2 to 3 hours at 85 ° with 900 cc. of hydrochloric acid, sp. gr. 1.19, and the solution boiled for 3 hours. This was then concentrated to a sirup under diminished pressure, diluted to 900 cc. with cold water, and neutralized with 50 per cent sodium-hydroxide solu- tion. After boiling with a large amount of bone-black and concentrating to 800 cc, much substance separated, which was recr y stall ized from about 300 cc. of water. The recrystallized product was dissolved in 5 per cent sul- phuric acid and precipitated by mercuric sulphate solution.1 The mercury precipitate was decomposed by hydrogen sulphide, the solu- tion freed from hydrogen sulphide, neutralized with sodium hydroxide, and acidified with acetic acid. On standing, cystine separated from the solution in hexagonal plates, and by adding alcohol to the filtrate more was obtained. When no more cystine could be thus obtained, the precipitation with mercu- ric sulphate was repeated. By several repetitions of this process 1. 18 grams of cystine were finally isolated, which, when recrystallized by dissolving in dilute ammonia and acidifying with acetic acid, gave the following analysis : Carbon and hydrogen : 0.3063 gram substance, dried at no0, gave 0.3379 gram C0.2 and 0.1444 gram H20. Calculated for C6Hi204N2S2, C 29.96, H 5.04 p. ct. ; found, C 30.08, H5.23 p. ct. TYROSINE. 219 grams of gliadin, equal to 200 grams dried at no°, were treated with 600 cc. of concentrated hydrochloric acid, digested for some time on a water- bath, and the solution boiled for 12 hours on an oil-bath. The solution was freed from most of the glutaminic acid by saturating with hydrochloric acid, and the filtrate from the glutaminic acid was diluted, boiled with bone-black, and then concentrated strongly to remove as much hydrochloric acid as possible. The rest of the acid was neutralized with 50 per cent sodium- hydroxide solution. On standing, a considerable precipitate separated, which was filtered out and dissolved in ammonia. The resulting solution was boiled until most of the ammonia had been removed and the tyrosine that separated was filtered out. When dried, this weighed 2.4 grams, equal to 1.2 per cent of the gliadin. Recrystallized from boiling water, this gave the following results on analysis : Carbon and hydrogen: 0.3661 gram substance, dried at no°,gave 0.7981 gram C02 and 0.2160 gram H20. Calculated for C9Hn03N, C 59.62, H 6.13 p. ct. ; found, C 59-45. H 6.56 p. ct. Tyrosine separated from our hydrolysis solutions of gliadin with very great difficulty. Two other attempts to determine its proportion, which were made by hydrolyzing with sulphuric acid, gave lower results, and the 1 Cf. Hopkins & Cole, Journal of Physiology, 1901, xxvn, p. 418. EXPERIMENTAL. 8 1 solutions from which the tyrosine separated still continued to give a strong Millon's reaction. Kutscher1 found 2.09 per cent of tyrosine in gliadin, and Abderhalden & Samuely2 found 2.37 per cent. HISTIDINE. Fifty grams of gliadin, equal to 43.97 grams dried at 1 io°, were hydrolyzed according to the directions of Kossel & Kutscher3 and the determination of the bases carried out according to the method of Kossel & Patten.4 The solution of the histidine was made up to 500 cc. and nitrogen deter- mined in 100 cc. of it. 100 cc. of solution gave ammonia = 1.37 cc. HC1 (1 cc. HC1 = 0.01 gram N) = 0.0137 gram N = 0.0685 gram N in 500 cc. = 0.2524 gram histidine = 0.58 p. ct. of the gliadin. The identity of this histidine was not established, as the quantity was too small to permit the preparation of a satisfactory product. ARGININE. The arginine solution was made up to 500 cc. and nitrogen determined in 50 cc. of it. 50 cc. of solution gave ammonia=4.i5 cc. HC1 (1 cc. HCl = o.oi gram N) = 0.0415 gram N or 0.415 gram in 500 cc. = 1.39 gram arginine or 3.16 p. ct. of the gliadin. The remaining solution was treated as Kossel & Kutscher direct and the arginine obtained as carbonate. A portion of this carbonate was converted into the picrolonate according to the directions of Steudel.5 This melted at 2260 to 2270, while Steudel gives 2250. Nitrogen: 0.0832 gram substance, dried at no0, gave 18.8 cc. moist N2 at 765 mm. and 250. Calculated for C6Hu02N4 ■ C10H8O5N4, N 25.62 p. ct.; found, N 25.40 p. ct. LYSINE. The filtrate from the silver precipitate of arginine and histidine was freed from silver, precipitated with phosphotungstic acid, and lysine tested for with picric acid in the usual way, but none was found. Kossel & Kutscher6 found, in the three fractions of the alcohol-soluble protein of the wheat kernel which they examined, 1.2, 0.43, and 1.53 per cent of histidine. The writer's determination falls between these. Kutscher,7 in discussing the individuality of these three fractions, considers the differ- 1 Kutscher, Zeitschrift fiir physiologische Chemie, 1903, xxxviii, p. in. 2 Abderhalden & Samuely, ibid., 1905, xi.iv, p. 276. 3 Kossel & Kutscher, ibid., 1900, xxxi, p. 165. 4 Kossel & Patten, ibid., 1903, xxxviii, p. 39. 5 Steudel, ibid., 1902, xxxvn, p. 219. 6 Kossel & Kutscher, loc. cit. ' Kutscher, loc. cit. 6 82 THE PROTEINS OF THE WHEAT KERNEL. ences in the amount of histidiue found in them to be within the limits of accuracy of these determinations. Kossel & Kutscher also found 3.05 per cent of arginine in the fraction which they called gluten-fibrin, 2.75 per cent in their gliadin, and 3.13 per cent in their mucedin. In determining the amount of protein hydrolyzed they calculated the weight from the nitrogen in solution. If, as seems probable, only one alcohol-soluble protein exists in this seed, namely, gliadin, with 17.5 per cent of nitrogen, and that the nitrogen of their solu- tions belonged to this protein, the proportion of arginine found by them would be, respectively, 3.13, 2.79, and 3.25 percent, with which 3.16 per cent agrees very closely. The results of this hydrolysis are given in table 17. Table 17. — Gliadin. P.ct. Glycocoll 0.00 Alanine 2.00 Amino-valerianic acid 0.21 Leucine 5.61 a-proline 7.06 Phenylalanine 2.35 Aspartic acid 0.58 Glutaminic acid 37-33 Serine o. 13 Tyrosine 1.20 P.ct. Cystine 0.45 Oxy-proline Lysine 0.00 1 1 istidine 0.58 Arginine 3. 16 Ammonia 5. 11 Tryptophane present 65.78 PROTEIN INSOLUBLE IN WATER, SALINE SOLUTIONS, AND ALCOHOL — GLUTENIN. As already stated, extraction with the above-named solvents, applied suc- cessively, removed but a part of the total protein contained in the wheat kernel, that remaining being soluble only in dilute acids and alkalis. The following extractions were next made to determine the nature of this body. Protein Extracted by Dilute Alkaline Solutions after Extracting the Flour with 10 per cent Sodium-Chloride Brine and then with Dilute Alcohol. After completely extracting from 4000 grams "straight flour" all the protein soluble in 10 per cent sodium chloride solution, the residue was freed wholly from protein soluble in cold alcohol of 0.90 sp. gr. The re- sulting residue was then extracted twice with a large quantity of o. 1 per cent potassium hydroxide solution. After standing 3 days at a tempera- ture of 50 with frequent stirring, this extract was filtered off and allowed to stand in a cold room until the greater part of the fine starch and other impurities which had escaped filtration had settled. The solution, which was still turbid, was decanted from the sediment and neutralized as exactly as possible with 0.2 per cent hydrochloric acid, thereby producing a precipi- EXPERIMENTAL. 83 tate which settled rapidly, leaving the filtrate milky in appearance. This precipitate, after decanting the solution, was dissolved in 0.2 per cent potas- sium-hydroxide water and set aside to deposit the suspended impurities. After decantation from the sediment which resulted, the solution was filtered, but almost nothing was thus removed. It was then neutralized with 0.2 per cent hydrochloric acid and the precipitate produced washed by decan- tation, first with water and then with dilute alcohol, absolute alcohol, and ether. No attempt was made in this case to obtain the whole of the protein soluble in alkali, as the difficulties presented by slow and imperfect filtration rendered this impossible. The substance thus obtained formed a brownish, horny mass, which weighed 13 grams. This preparation, 89, when dried at no°, on analysis gave the results shown in the table at the bottom of this page. Subsequent preparations of this body led to the idea that it was far from pure, this supposition being supported by the fact that the final solution from which it had been precipitated was turbid. The remainder of the preparation was then dissolved as completely as possible in 0.2 per cent potassium-hydroxide solution, and by repeated filtration through very dense filter-paper obtained perfectly clear. A considerable insoluble residue remained, which appeared to consist largely of the coagulated form of this protein. This residue was washed by decantation with water, alcohol, and ether. It was found to contain but 13.68 per cent of nitrogen, showing that the preparation, 89, contained much non-nitrogenous matter. The filtra- tion of the dissolved protein proceeded very slowly, so that it was conducted at a temperature near o°. A portion of the filtrate first obtained was re- moved, precipitated by 0.2 per cent hydrochloric acid, washed with water, alcohol, and ether, and the preparation, 90, analyzed with the following results. A second portion similarly yielded preparation 91. Preparatio w 89, 90, and 91. Preparation 89. Preparation 90. Preparation 91. I. Ash-free. I. II. Average. Ash-free. I. Ash-free. Carbon .... Hydrogen . Nitrogen... Sulphur.. . . P.ct. 52.47 6.75 15-51 0.86 P.ct. 52.91 6.S1 15.65 0.86 23-77 P. ct. 51.10 6.46 17.01 0.91 P.ct. O.92 P. Ct. 5i-io 6.46 17.01 0.92 Pel. 52.29 6.61 17.41 0.94 22.75 P.ct. P.ct. 17.17 17-33 Ash. . 0.88 2.28 0.97 100.00 100.00 84 THE PROTEINS OE THE WHEAT KERNEL. These results show that it is absolutely necessary to filter the alkaline solution of this body perfectly clear before the final precipitation, since other- wise a considerable amount of non-nitrogenous matter will be precipitated with it. Another extraction was made by treating 200 grams of " patent flour " from spring wheat with 10 per cent sodium-chloride solution added in small quantities so as to make a dough. This dough was then washed with 10 per cent sodium-chloride brine until nearly all the starch was re- moved and a gluten obtained similar in all respects to that resulting from treating the flour with water. This gluten was then chopped fine, thoroughly extracted with alcohol of sp. gr. 0.90, and then dissolved at 200 in o. 1 per cent potassium-hydroxide water. The resulting solution was filtered, but, as only a part of the impurities was thus removed, the filtrate was placed in an ice-box in shallow dishes and allowed to deposit a considerable part of the suspended matter. The decanted solution, which was only very slightly turbid, was then precipitated by neutralizing with 0.2 per cent hydro- chloric acid, and the precipitate washed by decantation with water, extracted thoroughly with dilute alcohol, digested with absolute alcohol and then with ether. This preparation, 92, was a snow-white, light, porous mass, easily reduced to a powder. It had the following composition : Preparation 92. I. II. Average. Ash-free. P.ct. P. ct 52.18 6.90 16.90 1. 00 P. Ct. P. Ct. 52.50 6.94 17.22 1. 00 22.34 Sulphur . . 17. 11 17.01 Ash . 0.63 100.00 Protein Extracted by Dilute Alkaline Solutions from Gluten after Extracting the Latter with Dilute Alcohol. 2000 grams of ' ' straight flour ' ' from spring wheat were made into a dough with distilled water, and this was washed with river- water until the gluten was freed as completely as possible from starch. This gluten was then ex- tracted with 75 per cent alcohol as long as anything was removed. The insoluble residue was dissolved in 0.15 per cent potassium-hydroxide water and the resulting solution allowed to stand in a cold room for 48 hours. The solution was thus freed from but a part of the suspended matter. After decanting from the sediment, the solution was neutralized with dilute hydro- chloric acid, the precipitate produced, washed several times by decantation EXPERIMENTAL. 8; with water, thoroughly extracted with alcohol of 0.90 sp. gr., then with stronger alcohol, and finally with absolute alcohol and with ether. The precipitate was again dissolved in o. 1 per cent potassium-hydroxide solution and allowed to stand over night. It was then filtered, and a part of the clear filtrate first obtained was precipitated by neutralization with 0.2 per cent hydrochloric acid. This precipitate was washed with water, alco- hol, absolute alcohol, and ether, yielding preparation 93. A part of this preparation, 93, was redissolved in 0.2 per cent potassium-hydroxide water and found to contain a considerable amount of substance which had become insoluble in consequence of drying. This insoluble portion was filtered off, washed with water, alcohol, and ether, and gave preparation 94. The filtrate from this substance was precipitated with 0.2 per cent hydrochloric acid, and the precipitate filtered off and washed with water, alcohol, and ether. Through an accident this preparation dried on the filter and could not be removed from the paper. It was then again dissolved in dilute potassium- hydroxide water and treated exactly as before, yielding preparation 95. The following analyses show the composition of these three preparations : Preparations 93, 94, and 95. I. II. Average. Ash-free. Preparation 93 : Carbon P.ct. 51-59 6.72 17-34 \ P. ct. P.ct. 51-59 6.72 17.37 P.ct. 52.32 6.82 17.61 23-25 Hydrogen Nitrogen 17.39 Sulphur Oxygen V 1 I.40 I Ash Preparation 94 : Carbon 100.00 50.79 6.62 16.38 } 50.79 6.62 16.38 52.87 6.88 17.05 23.20 Hydrogen Nitrogen 16.38 Sulphur i 3-94 3-94 Ash Preparation 95 : Carbon 100.00 51-50 6.70 16.70 } 51-37 6.57 51.44 6.64 16.70 52.62 6.80 17.08 23-5o Hydrogen Nitrogen Sulphur Oxygen J 2.25 Ash 1 100.00 86 THE PROTEINS OF THE WHEAT KERNEL. Another lot of gluten made from 1000 grams of "straight flour" was treated in exactly the same way as that last described. This gluten was thoroughly extracted with alcohol of 0.90 sp. gr. and the residue dissolved in about 500 cc. of 0.2 per cent potassium-hydroxide water. After standing over night, a very turbid liquid was decanted from the sediment which had formed and treated with very dilute acetic acid added to slightly acid reac- tion. The precipitate produced was washed with water, alcohol, and ether and dissolved again in 0.2 per cent potassium-hydroxide water. The result- ing solution, filtered perfectly clear, was precipitated with 0.2 per cent hydrochloric acid, washed by decantation with water, then with dilute alcohol increased gradually in strength up to absolute alcohol, and finally with ether. When dried over sulphuric acid, a pure white, light mass was obtained. This preparation, 96, was analyzed with the results shown in the table below. Again, gluten was prepared in the usual manner and extracted with alcohol until everything soluble in that reagent was removed. The residue was then dissolved in 0.2 per cent potassium-hydroxide water and, in order to carry the operation to an end as rapidly as possible, the solution was at once thrown into a filter. As soon as the filtrate ceased to pass through turbid it was returned to the funnel and the filtration continued in an ice-chest. A considerable portion of the solution was obtained perfectly clear after 20 hours. This was then precipitated with o. 2 per cent hydrochloric acid and treated in the usual manner, giving preparation 97. Preparations 96 and gj. Preparation 96. Preparation 97. I. II. Average. Ash-free. I. Ash-free. Carbon Hydrogen . . Nitrogen. . . . Sulphur .... P. Ct. 52.07 6.71 17.23 1.07 P.ct. 52.23 6.88 17.42 P.ct. 52.15 6.80 17.33 I.07 P.ct. 52.54 6.85 17.46 1.07 22.08 P.ct. 52.07 6.77 17.49 1.24 P.ct. 52.38 6.81 17.59 1.24 21.98 100.00 Ash 0.74 0.76 o.75 0.61 100.00 In order to determine whether a loss of nitrogen occurred through pro- longed contact with the alkaline solution, the following experiment was tried, in which the conditions under which preparation 97 was made were repeated and preparation 98 obtained. The rest of the alkaline solution was kept in the ice-box for 3 days longer, during which time the ice melted and the EXPERIMENTAL. 87 temperature rose to 200. The preparation, 99, obtained by neutralizing the solution with hydrochloric acid contained the same amount of nitrogen as preparation 98, from which it is evident that prolonged solution in dilute potassium hydroxide solution caused no loss of nitrogen. Preparations 98, 99, and 100. Preparation 98. Preparation 99. Preparation 100. I. Ash-free. I. Ash-free. I. II. Average. Ash-free. Nitrogen . . . Ash P.ct. 17.32 1.23 P.ct. 17-53 P.ct. i7-5o 0.16 P.ct. 17.53 P.ct. 17.04 1.22 P.ct. 16.93 P. ct. 16.99 P.ct. 17.20 Another preparation of this substance (preparation 100) was made by completely extracting 200 grams of "patent flour" (from spring wheat) with large quantities of alcohol of 0.90 sp. gr. The flour was then washed with absolute alcohol and air-dried. The dry material was then powdered and made into a dough with distilled water which had considerable coher- ence, showing that the protein insoluble in alcohol played an important part in its formation. This dough was then washed on a fine hair-sieve under a stream of running water, but no coherent gluten resulted. The washings were then allowed to settle, and the sediment, after decanting the solution, was treated with 0.2 per cent potassium-hydroxide water. The solution so obtained, after standing over night, was decanted from the sediment and precipitated with o. 2 per cent hydrochloric acid and the separated protein allowed to settle. The solution was then decanted, the precipitate dissolved in 0.2 per cent potassium-hydroxide water, and filtered clear in the ice-box. The filtered solution was then precipitated and the separated substance treated in the usual manner. This preparation, 100, contained nitrogen, as shown in the above table. In order to learn whether any change in the protein occurred in conse- quence of contact with aqueous solutions before extracting with potassium- hydroxide water, the following experiment was made : 1000 grams of "straight flour" from spring wheat were repeatedly ex- tracted with alcohol of 0.90 sp. gr., and after removing everything soluble in that liquid the alcohol was squeezed out as completely as possible in a screw-press and the residue extracted with 0.2 per cent potassium-hydroxide water. It was found impossible to separate the solution from the undis- solved portion of the meal, either by filtration or subsidence, on account of the presence of gummy matter. An equal volume of alcohol of sp. gr. 0.820 88 THE PROTEINS OE THE WHEAT KERNEL. was therefore added, and on long standing the insoluble substance gradually settled, leaving a comparatively clear, yellow solution. This was then siphoned off and filtered. On account of its gummy character it filtered very slowly. The clear solution finally obtained was then precipitated with 0.2 per cent hydrochloric acid, the precipitate filtered off and dissolved in 0.2 per cent potassium-hydroxide water. The solution so obtained was filtered clear and precipitated with 0.2 per cent hydrochloric acid, the result- ing precipitate washed by decantation with water, alcohol — at first dilute, then gradually increased to absolute — and finally with ether. Preparation 101 was thus obtained, having the composition shown in the following table : Preparations 101 and 102. Preparation 101. Preparation 102. I. Ash-free. I. II. Average. Ash-free. Carbon Hydrogen . . Nitrogen . . . Sulphur .... Oxygen Ash P. Ct. 52.14 6.91 17-54 } 0. 10 P. ct. 52.19 6.92 17-56 23-33 P.ct. 51-59 6.S5 17.21 i 0.89 P.ct. 5I-67 6.S7 17.32 P.ct. 5I-63 6.86 17.27 0.89 P.ct. 52.19 6-93 17-45 0.90 23-43 1.07 100.00 100.00 This analysis shows that the protein extracted by potassium hydroxide water from the flour which has not been in contact with water has the same composition as that obtained in the other experiments. Protein Extracted by Dilute Alkaline Solutions after Complete Extraction with Dilute Alcohol of Gluten from Whole-wheat Flour. 1000 grams of flour made by grinding the entire kernel of spring wheat were made into a dough, washed with water till free from starch, and the gluten obtained extracted thoroughly with dilute alcohol. The residue was then dissolved in 0.2 per cent potassium-hydroxide water, and after the resulting solution had stood for some time it was decanted from the sedi- ment and precipitated by 0.2 per cent hydrochloric acid. The precipitate was washed by decantation with water, thoroughly extracted with dilute alcohol and then with absolute alcohol, and finally with ether, and then redissolved in 0.2 per cent potassium-hydroxide water. The solution, after filtering perfectly clear, was precipitated and the pre- cipitate treated in the same manner as all the preceding preparations. When EXPERIMENTAL. 89 analyzed, preparation 102 was found to have the composition shown in the table on page 88. The filtrate from this preparation, as well as all the others previously- described, contained a small amount of protein matter. This was then treated with a solution of copper sulphate and the small precipitate thereby- produced filtered off, washed with water, alcohol, and ether (preparation 103), and found to contain the following amount of nitrogen : Preparation 103. I. Ash-free. Nitrogen Ash P. ct. 13.28 23.88 P.ct. 17-45 In the same way as 102 a preparation of this protein was made from flour from the entire kernel of winter wheat. The composition of this preparation is shown by the figures in the following table for preparation 104. Preparations 104 and 105. Preparation 104. Preparation 105. I. II. Average. Ash-free. I. Ash-free. Hydrogen . . Nitrogen. . . . Sulphur .... Ash P. ct. 5I.7I 6.79 17.32 } P.ct. P.ct. 5I-7I 6.79 17.38 P.ct. 52.03 6.83 17.48 23.66 P.ct. 51-17 6.70 17-35 P.ct. 52.44 6.86 17.81 22.89 100.00 17.44 J 0.62 0.62 2.62 100.00 A second portion of the same solution from which preparation 104 had been precipitated was obtained 2 days later. This was then treated in the same way and yielded preparation 105, which was found to have the composition shown in the table above. If preparations 89 and 92 are omitted on account of being obtained from unfiltered solutions, 94 as an altered and insoluble product, and 95 as having been subjected to an exceedingly prolonged action of an alkaline solution, then will be had, in table 18, the analyses which, in the opinion of the writer, most nearly represent the true composition of this protein. 90 THE PROTEINS OF THE WHEAT KERNEL. Table 18. — Protein of the wheat kernel soluble only in dilute acids and alkalis. 90. 91. 93- 96. 97. 98 99- Carbon Hydrogen Sulphur P.ct. 52.29 6.61 I7-4I •94 22.75 100.00 P.ct. P. ct. 52.32 6.82 17.61 } 23.25 100.00 P. ct. 52.54 6.85 17.46 f 1.07 \ 22.08 100.00 P.ct. 52.38 6.81 17.59 1.24 21.98 100.00 P.ct. P. ct. x7-33 17-53 17-53 100. 101. 102. 103. 104. 105. Average. P.ct. P.ct. 52.19 6.92 17.56 } 23.33 P.ct. 52.19 6-93 17-45 23-43 IOO.OO P.ct. P.ct. 52.03 6.83 17.48 23.66 P. ct. 52.44 6.86 17.81 22.89 100.00 P.ct. 52.34 6.83 17-49 r 1.08 \ 22.26 Sulphur 17.20 17.45 Oxveen IOO.OO 100.00 100.00 There can be no doubt that the earlier analyses of this protein, with the exception of those made by Ritthausen and Chittenden & Smith, are incor- rect, for the substance analyzed contained a large part of the impurities of the gluten, since after extraction with alcohol the residue was not dissolved in any solvent, and the starch, bran, etc., contained in it filtered off. Ritt- hausen' s figures are the average of two analyses of preparations obtained from solutions in 0.2 per cent potassium-hydroxide water which had been filtered, one perfectly clear, giving a product containing 17.21 per cent N, the other very nearly clear and containing 17.08 per cent N. It is evident that the alkaline solution of the gluten that has previously been extracted with cold dilute alcohol contains something which requires filtration through very dense paper to effect its separation. Wheat gluten contains a considerable amount of phytocholesterin and lecithin, together with fat, and it is probable that all these bodies are held suspended in the solution of the dissolved protein as an emulsion, and can only be removed by pre- cipitating the first solution in potassium-hydroxide water, extracting the precipitate with alcohol and ether, redissolving, and filtering the resulting solution perfectly clear. The first alkaline solution before precipitating and extracting with alcohol and ether can not be filtered, as it either passes through turbid or does not run through at all. On the ground of priority and the fact that the relations to animal proteins which gave rise to the various names subsequently applied to this body have EXPERIMENTAL. 91 been proved to have no foundation, it would be desirable to return to Taddei's original name and in future call this protein zymon. Unfortunately this name is derived from the Greek word fo^, a ferment, and, as the results of this investigation show that the supposed ferment-changes do not occur in the formation of gluten, this name is undesirable. As this protein is espe- cially characteristic of gluten, it seems appropriate to call it glutenin, a name suggested by Prof. S. W. Johnson. Hydrolysis of Glutenin. The large quantity of glutenin which was required for the quantitative determination of its decomposition products was prepared from the residue of the wheat gluten after extracting the gliadin with alcohol. This residue was dried at room temperature and then ground to a powder, which was extracted first with absolute alcohol and then with ether as long as either solvent re- moved anything from it. The alcohol was then removed at room temperature and the residual powder treated with just enough 0.2 per cent solution of potassium hydroxide to dissolve it. The resulting turbid solution was then filtered perfectly clear and neutralized with very dilute hydrochloric acid. The precipitate produced was extracted with 70 per cent alcohol as long as any gliadin was removed, then thoroughly dehydrated with absolute alcohol, and dried over sulphuric acid. 940 grams, equal to 839.32 grams of glutenin, ash and water free, were hydrolyzed by heating for several hours on a water-bath with a mixture of 950 cc. of concentrated hydrochloric acid and 950 cc. of water. After standing over night, the solution was boiled on an oil-bath for 9 hours, and then saturated with hydrochloric acid gas. By the same treatment as that applied to gliadin (p. 71) 202.73 grams glutaminic acid hydrochloride, equal to 162.40 grams of the free acid, were obtained. Recrystallized once from concentrated hydrochloric acid, this melted at 1980. Chlorine: 0.5386 gram substance gave 0.421 1 gram AgCl. Nitrogen: 0.5911 gram substance gave NH3 = 4.53cc. HC1 (1 cc. HCl = o.oi gram N)„ Calculated for C5H10O4 NCI, CI 19.35, N 7.65 p. ct. ; found, CI 19.33, N 7.66 p. ct. The united filtrates and washings were concentrated to a sirup on a water- bath under reduced pressure and the hydrochlorides of the amino-acids esterified three times, as in the case of gliadin. The hydrochlorides of the esters were neutralized and the free esters shaken out with ether, as for gliadin. After drying the ether solution of the esters with potassium car- bonate, it was kept 2 days over anhydrous sodium sulphate. The ether was then removed by distillation from a water-bath at atmospheric pressure and the esters distilled, with the results shown in the table on page 92 for distillation A. The undistilled residue weighed 21 1.5 grams. 92 THE PROTEINS OF THE WHEAT KERNEL. Distillations A and B. Fraction. Distillation A. Distillation B. Temper- ature of bath up to — Pressure. Weight. Fraction. Temper- ature of bath up to — Pressure. Weight. I o 65 IOO IOO IIO 155 200 m m . 12.0 12.0 4.0 1.5 1.5 0.8 Gravis. 61.OO 43-74 45-94 33-Qi 43-32 32-56 I.. 0 65 88 IOO 120 180 mm. 12.0 12.0 10.0 0.8 0.8 Grams. 28.74 24.27 19.18 23-23 23.92 II II.. . ™{» IV IV.. V 259-57 H9-34 The residue, from which the esters had been removed by ether, was sub- jected to two or more esterificatious, as in the case of gliadin, and the esters distilled. (See distillation B in the table above.) The undistilled residue weighed 93 grams. The different fractions from the two distillations were worked up as follows: Fraction. Temper- ature of bath up to — Pressure. Weight. , r A 0 65 65 mm. 12 12 Grams. 61.OO 28.74 Mb This was saponified at once by evaporating to a sirup with concentrated hydrochloric acid and the residue dissolved in alcohol and esterified with dry hydrochloric acid gas. The glycocoll ester hydrochloride which sepa- rated weighed 6.68 grams. Recrystallized from alcohol, this melted at 1440. Carbon and hydrogen: 0.3179 gram substance gave 0.3984 gram C02 and 0.2151 gram H20. Calculated for C4H10O2NCl, C 34.39, H 7.23 p. ct.; found, C 34.18, H 7.52 p. ct. The filtrate from the glycocoll ester hydrochloride was saponified by evaporating on the water-bath with hydrochloric acid, the latter removed with silver sulphate, and the solution freed from sulphuric acid with an EXPERIMENTAL. 93 equivalent quantity of barium hydroxide. By fractional crystallization this solution yielded 11.83 grams of alanine, which melted at about 2900. Carbon and hydrogen: 0.2701 gram substance, dried at no0, gave 0.4001 gram C02 and 0.1946 gram H20. Nitrogen: 0.3460 gram substance, dried at no°, gave NH3 = 5.44 cc. HC1 (1 cc. HCl = o.oi gram N). Calculated for C3H702N, C 40.40, H 7.93, N 15.75 P- ct.; found, C 40.40, H 8.or, N 15.73 P- ct. Fraction. Temper- ature of bath up to — Pressure. Weight. «{tv:::::: 0 100 88 mm. 12 IO Grams. 43-74 24.27 The united esters were saponified by boiling with 10 parts of water for 5 hours, when their solution reacted neutral to litmus. This solution was evaporated to dryness under reduced pressure and boiled up with absolute alcohol, whereby 0.5 gram substance was dissolved. By repeated fractional crystallization of the substance insoluble in alcohol, 3.5 grams leucine were obtained. The isolated leucine decomposed at about 2980. Carbon and hydrogen: 0.240S gram substance, dried at no°, gave 0.4852 gram C02 and 0.2169 gram H2C Calculated for C6H1302N, C 54-89, H 10.01 p. ct. ; found, C 54.95, H 10.01 p. ct. After adding to the filtrate from this leucine the most soluble portion of fraction III, and further fractiouing, 25.76 grams of alanine and 3.84 grams of glycocoll were obtained. The latter was isolated as the hydrochloride of the ester, which melted at 1440. The alanine was racemized and the a-naphthyl-hydantoic acid prepared according to the directions of Neuberg & Manasse.1 When recrystallized from dilute alcohol, this melted at 1970. Carbon and hydrogen: 0.2527 gram substance, dried at 900, gave 0.5998 gram C02 and 0.1280 gram H20. Nitrogen: 0.3515 gram substance, dried at 900, gave NH3 = 3.74 cc. HC1 (1 cc. HC1 = 0.01 gram N). Calculated for CuH1403N,, C 65.06, H 5.48, N 10.87 p. ct.; found, C 64.74, H 5.61, N 10.61 p. ct. 1 Neuberg & Manasse, Berichte der deutschen chemischen Gesellschaft, 1905, xxxviii, P- 2359. 94 THE PROTEINS OF THE WHEAT KERNI'J.. Fraction. Temper- ature of bath up to — Pressure. Weight. ™{i o no 1 20 in m . 1-5 0.8 Grains. 78.95 42.41 These esters were saponified by boiling with 6 parts of water for 5 hours, when their solution reacted neutral to litmus. After evaporating to dryness under reduced pressure, the dried residue was extracted with boiling absolute alcohol. The part insoluble in alcohol, by a systematic fractional crystal- lization, gave 32.42 grams of leucine, which decomposed at about 2980. Carbon and hydrogen: 0.4255 gram substance, dried at no0, gave 0.8535 gram C02 and 0.3926 gram H20. Nitrogen: 0.1488 gram substance gave NH,= 1.60 cc. HC1 (1 cc HCl = o.oi gram N). Calculated for CGH130,N, C 54.89, H 10.01, N 10.70 p. ct.; found, C 54.71, H 10.251 N 10.77 P- ct- The filtrate from this leucine yielded 2 grams of substance, which had the composition and properties of amino-valerianic acid. Carbon and hydrogen: 0.2216 gram substance gave 0.4144 gram C02 and 0.1967 gram H20. Calculated for C5Hu02N, C 51.22, H 9.48 p. ct.; found, C 51.00, H 9.86 p. ct. Specific rotation. — Dissolved in 20 per cent hydrochloric acid, 0) -^ = + 25.63° A similar preparation from gliadin gave +25.790. E. Fischer & Dor- pinghaus1 found +25.90 for a preparation from horn, and E. Schulze & Winterstein2 found +28.20 and + 27.90 for a preparation from lupine seedlings. The solution used for determining the specific rotation was freed from hydrochloric acid with silver sulphate, and the amino-acids racemized by heating with an excess of barium hydroxide in an autoclave at 1750 for 19 hours. After removing the barium quantitatively with sulphuric acid, the substance was coupled with a-naphthylisocyanate. The hydantoic acid melted constantly, on repeated recrystallization from dilute alcohol, at 1830 to 'Fischer, E., & Dorpinghaus, Zeitschrift fur physiologische Chemie, 1902, xxxvi, p. 462. 2 Schulze, E., & Winterstein, ibid., 1902, xxxv, p. 300. EXPERIMENTAL. 95 1 84°. Heated side by side with the corresponding substance obtained from gliadin, the hydantoic acid from glutenin melted at 20 higher. Carbon and hydrogen: 0.3277 gram substance, dried at 8o°, gave 0.S067 gram C02 and o. 1891 gram H20. Calculated for C16H18OjN2, C 67.07, H 6.35 p. ct.; found, C 67.14, H 6.42 p. ct. From the filtrate from the amino-valerianic acid there was further obtained 1. 4 1 grams of alanine. The alcoholic solution which contained the proline was evaporated to dryness under reduced pressure and the dried residue again treated with boiling absolute alcohol. Even after several repetitions of this process no substance insoluble in alcohol could be obtained. The alcohol- soluble substance, when dried, weighed 35.54 grams. A copper salt was prepared from this in the usual manner, and its solution evaporated to dryness under reduced pressure. The dried residue was extracted with boil- ing absolute alcohol in order to remove the /-proline copper. The residue insoluble in alcohol, when recrystallized from water, gave 15.53 grams of racemic proline copper, equivalent to 10.9 grams of proline. Water : 0.361S gram substance lost at no° 0.0399 gram of H20. Copper : 0.3176 gram substance, dried at no°, gave 0.0856 gram CuO. Calculated for C10H16O4N2 Cu ■ 2H.O, H20 11.00 p. ct. ; found, H20 11.03 P- ct- Calculated for C10H16O4N2 Cu, Cu 21.79 P- ct- I found, Cu 21.54 p. ct. The alcoholic solution of the /-proline copper salt was evaporated to dry- ness, the copper removed, and the proline identified as the phenylhydantoin which was prepared according to the directions of Fischer.1 The substance thus prepared was at once pure and melted at 143°. Carbon and hydrogen: 0.2822 gram substance, dried in vacuo over H,S04, gave 0.6866 gram CO, and o. 1494 gram H20. Nitrogen: 0.1025 gram substance gave NH3 = 1.33 cc. HC1 ( 1 cc. HC1 = 0.01 gram N). Calculated for C12H1202N2, C 66.60, H 5.61, N 12.99 P- ct- \ found, C 66.36, H 5.88, N 12.98 p. ct. f Fraction IV, A. Temperature of bath, up to 1550. (Pressure, 1.5 mm. Weight, 43.32 grams. This fraction was shaken out with ether in the usual way and the ether allowed to evaporate spontaneously. No evidence of the presence of phenyl- alanine was obtained. The residue of ester was saponified with concentrated hydrochloric acid and the hydrochloride decomposed with ammonia. The free acid crystallized in the characteristic form of leucine and decomposed at 298 °. There were obtained 13.99 grams of leucine. Carbon and hydrogen: 0.2906 gram substance, dried at no°, gave 0.5825 gram C02 and 0.2690 gram H20. Calculated for C6H130,N, C 54.89, H 10.01 p. ct. ; found, C 54.67, H 10.2S p. ct. Fischer, E., Zeitschrift fur physiologische Chemie, 1901, xxxiil, p. 151. 96 the; proteins of the wheat kernel. The aqueous layer was'saponified by heating on a water-bath with an excess of barium hydroxide for 5 hours. The barium aspartate, which separated in considerable quantity on standing, was united with that obtained from frac- tion V and treated as will be described later. From the filtrate from the barium aspartate no definite substance could be obtained. It appeared to contain serine, but none could be isolated, even by the use of ,3-naphthalene-sulphone-chloride. Fraction. Temper- ature of bath up to — Pressure. Weight. V, A 0 200 180 mm. 0.8 0.8 Grams. 32-56 23.92 IV, B These esters were shaken 'out with ether, and the substance extracted was saponified with hydrochloric acid. The hydrochloride thus obtained, which weighed 20.21 grams, equal to 16.55 grams free phenylalanine, was converted into the free acid with ammonia and then into the copper salt by boiling its solution with copper hydroxide.1 Copper : 0.2099 gram substance, dried at 1100, gave 0.0425 gram CuO. Nitrogen: 0.22S3 gram substance gave NH3 = 1.63 cc. HC1 (1 cc. HCl = o.oi gram N)- Calculated for C18H20O4N2Cu, Cu 16.23, N 7.17 p. ct. ; found, Cu 16.18, N 7.14 p. ct. The free phenylalanine isolated from this copper salt melted at 2630 to 2650. Nitrogen: 0.1085 gram substance gave NH3 = 0.92 cc. HC1 (1 cc. HC1 = 0.01 gram N). T Calculated for C9Hu02N, N 8.50 p. ct.; found, N 8.48 p. ct. The aqueous layer was saponified by heating for 5 hours with an excess of barium hydroxide on the water-bath. The barium aspartate, which sepa- rated on standing, was united with that previously obtained from fraction IV, A, decomposed with an equivalent amount of sulphuric acid, and 7.12 grams of free aspartic acid were obtained from the solution. Carbon and hydrogen: 0.3293 gram substance, dried at no°, gave 0.4335 gram C02and 0.1625 gram H20. Nitrogen: 0.2997 gram substance gave NH3 = 3.i7 cc. HC1 (1 cc. HCl = o.oi gram N). ; Calculated for C4H704N, C 36.05, H 5.31, N 10.55 P- ct. ; found, C 35.90, H 5.48, N 10.58 p. ct. The filtrate from barium aspartate was freed from barium, concentrated under reduced pressure, and saturated with hydrochloric acid. On long standing on ice a trace of phenylalanine hydrochloride separated, but no glutaminic acid hydrochloride could be obtained. After removing the *Cf. Schulze & Winterstein, Zeitschrift fur physiologische Chemie, 1902, xxxv, p. 210. EXPERIMENTAL. 97 hydrochloric acid with silver sulphate, the solution was boiled with a solu- tion of copper hydroxide and 1.1 grams of copper aspartate were isolated. Copper : 0.3991 gram substance, air-dried, gave o. 1163 grain CuO. Nitrogen : 0.2066 gram substance gave NH.= 1.06 cc. HC1 (1 cc. HC1 =0.01 gram N). Calculated for C4H504N, Cu • 4^ H.,,0,1 Cu 23.06, N 5.09 p. ct. ; found, Cu 23.27, N 5.13 p. ct. The filtrate from this copper aspartate, when freed from copper by hydro- gen sulphide and concentrated to small volume, yielded, by fractional crys- tallization, 4.35 grams of nearly pure serine. Carbon and hydrogen: 0.2847 gram substance gave 0.3559 gram C02 ando.1751 gram H20. Calculated for C3H703N, C 34.29, H 6.67 p. ct. ; found, C 34.09, H 6.83 p. ct. This substance browned at about 213° and decomposed to a brownish mass at about 2430. Residue from Distillation. The residues which remained after distilling off the esters weighed 304.5 grams. These were dissolved in hot alcohol, and from their united solutions 3.16 grams of substance separated on cooling. The filtrate from this sub- stance was evaporated under reduced pressure, the residue dissolved in water and saponified by heating with an excess of barium hydroxide for 9 hours. After removing the barium quantitatively, the solution was concentrated and saturated with hydrochloric acid. On standing for some time on ice, 38.25 grams of glutaminic acid hydrochloride, which melted at 1980, were ob- tained. This was equivalent to 30.64 grams of free glutaminic acid, and, with that previously isolated, made a total of 193.04 grams, or 23 per cent of the glutenin. The glutaminic acid hydrochloride was decomposed with an equivalent quantity of potassium hydroxide, and the free glutaminic acid was recrystallized from water. It melted at 202 ° to 203 ° with effervescence. Carbon and hydrogen: 0.3504 gram substance, dried at 1100, gave 0.5275 gram C02 and 0.2003 gram H20. Calculated for C5H904N, C 40.82, H 6.12 p. ct. ; found, C 41.06, H 6.35 p. ct. Residue after Esterification. The residue which remained after the third esterification and extraction of the esters was treated in the same way as the corresponding residue from gliadin (p. 77), and the solution, freed from all mineral salts and bases precipitable by phosphotungstic acid, was concentrated under reduced press- ure to a small volume and then left for some time over sulphuric acid. 1Ritthausen, Die Eiweisskorper, etc., Bonn, 1872, p. 219. 98 THE PROTEINS OF THE WHEAT KERNEL. After removing a little tyrosine that first separated, 1.87 grams of serine crystallized out, which, when recrystallized from water, browned at about 2100 and decomposed at 2400. Carbon and hydrogen: 0.2308 gram substance, dried at no°, gave 0.2894 gram C02 and o. 1407 gram H20. Calculated for C,H703N, C 34.29, H 6.67 p. ct. ; foun.l, C 34.19, H 6.77 p. ct. This, with the serine obtained from fraction V, gives a total of 6.22 grams of serine isolated. The mother-liquor from the serine contained consider- able substance, but no oxy-proline or other definite substance could be obtained from it. cystine. 300 grams of glutenin were hydrolyzed in the way described for gliadin (p. 80). After evaporating at low pressure to a sirup, neutralizing the remaining excess of acid with sodium hydroxide, and decolorizing the solu- tion with bone-black, a considerable quantity of tyrosine separated out, which, on examination, was found to contain nearly all the cystine that could be detected in the solution. It was therefore dissolved in 5 per cent sulphuric acid and the cystine precipitated with mercuric sulphate. The mercury precipitate was decomposed with hydrogen sulphide, the solution concentrated somewhat, made alkaline with ammonia and then acid with acetic acid, and an equal volume of alcohol added. The cystine, which sep- arated on standing in characteristic hexagonal plates, weighed only 0.17 gram. No more could be obtained. This was dissolved in ammonia and reprecipitated by acetic acid. Sulphur : 0.0897 gram substance, dried at no0, gave 0.1730 gram BaS04. Calculated for C6H1,04N2S2, S 26.68 p. ct. ; found, S 26.53 P- ct. Although glutenin contains about the same amount of sulphur as gliadin, the amount of cystine obtained from the latter under similar conditions was very much greater. It would seem as if glutenin in fact yields less cystine, though the uncertainties attending the isolation of this substance will not permit of a positive conclusion. TYROSINE. 250 grams of glutenin were boiled with a mixture of 750 grams sulphuric acid and 1500 grams of water for 12 hours. The solution was freed from sulphuric acid by an equivalent amount of barium hydroxide, and after concentrating to 800 cc. allowed to stand for some time. A considerable quantity of tyrosine separated, which was filtered out, the filtrate boiled with barium carbonate in order to expel ammonia, and then concentrated to one-half its original volume. After cooling, the residue of barium carbonate EXPERIMENTAL. 99 and other substances which had separated were extracted with hot dilute ammonia and the filtered extract concentrated and cooled. On standing, a little more tyrosine separated, which was added to that first obtained. No more tyrosine could be isolated from the solution of the hydrolytic decom- position products. All of the tyrosine which had separated was dissolved in 5 per cent sulphuric acid, and phosphotungstic acid was added to the solution. Only a small precipitate resulted. After removing the phospho- tungstic acid with barium hydroxide, the solution was concentrated strongly and allowed to cool. After standing for some time, 9.62 grams of tyrosine were obtained, which is equal to 4.25 per cent of the glutenin. Carbon and hydrogen: 0.2922 gram substance, dried at uo°, gave 0.6370 gram CO, and 0.1634 gram H20. Calculated for C9HU03N, C 59.62, H 6.13 p. ct.; found, C 59.45. H 6.21 p. ct. Kutscher1 found 2.75 per cent of tyrosine in " gluten- casein." HISTIDINE. Fifty grams of glutenin, equal to 43.39 grams dried at no°, were hydro- lyzed, and the arginine and histidine separated in the same way as that described for gliadin (p. 81). The solution containing the histidine was made up to 500 cc. Nitrogen: 100 cc. solution gave NH3 =4.15 cc. HC1 (1 cc. HC1 =0.01 gram N) = 0.0415 gram N = 0.2075 gram in 500 cc. = 0.7645 gram histidine = 1.76 p. ct. of the glutenin. The amount of histidine in the remaining solution was too small for identification. arginine. The filtrate from the histidine precipitate yielded 500 cc. of solution con- taining the arginine, in which was found the following amount of nitrogen : Nitrogen: 50 cc. solution gave NH3 = 6.79 cc. HC1 (1 cc. HC1 = 0.01 gram N) = 0.0679 gram N, or 0.679 gram in 500 cc, or 2.107 gram arginine = 4.72 p. ct. of the glutenin. The remaining solution, treated as Kossel directs, yielded the arginine as carbonate. This was converted into the copper salt, which gave the fol- lowing results on analysis : Carbon and hydrogen: 0.2118 gram substance, air-dried, lost 0.0210 gram H20 at ioo°. Calculated for C12H28O10N10 Cu ■ 3 H20, H20 9.15 p. ct. ; found, H20 9.92 p. ct. Copper : (I) 0.1858 gram substance, dried at ioo°, gave 0.0275 gram CuO ; (II) 0.1808 gram substance, dried at ioo°, gave 0.0267 gram CuO. Calculated for C12H28O10N10 Cu, Cu 11.85 p. ct. ; found, Cu (I) 11.84, (II) 11.78 p. ct. 1 Kutscher, Zeitschrift fur physiologische Chemie, 1903, xxxvi, p. 114. IOO THE PROTEINS OF THE WHEAT KERNEL. Kossel & Kutsclier1 found in three separate determinations of arginine in glutenin 4.50, 4.02, and 4.54 per cent. They base their determinations on the supposition that glutenin contains 16.2 per cent nitrogen. If their results are recalculated to a basis of 17.5 per cent of nitrogen which has been found in this protein,2 they become 4.82, 4.52, and 4.84 per cent, with which 4.72 per cent agrees very closely. lysine. The filtrate from the arginine silver was treated as Kossel directs, and after precipitating the lysine with phosphotungstic acid it was converted into the picrate, and 2.33 grams, equivalent to 0.907 gram of free lysine, were obtained. This is equal to 1.92 per cent of the glutenin. Kossel & Kutscher found in the three determinations of this substance which they made in this protein 1.9, 2.29, and 2 per cent, or, recalculating to a basis of 17.5 per cent of nitrogen in this protein, 2.05, 2.15, and 2.40 per cent. Nitrogen: 0.1225 gram substance, dried at 1200, gave 23.1 cc. moist N2 at 760 mm. and 290. Calculated for C6Hu02N2 • C6H307N3) N 18.70 p. ct. ; found, N 18.76 p. ct. The results of this hydrolysis are given in table 19. Table 19. — Glutenin. P.ct. Glycocoll 0.89 Alanine 4.65 Amiuo-valerianic acid 0.24 Leucine 5.95 a-proline 4.23 Phenylalanine 1.97 Aspartic acid 0.91 Glutaminic acid 23.42 Serine 0.74 P.ct. Tyrosine 4. 25 Cystine 0.02 Lysine 1.92 Histidine 1 .76 Arginine 4.72 Ammonia 4.01 Tryptophane present Total 59.66 THE AMOUNT OF THE VARIOUS PROTEINS CONTAINED IN THE KERNEL OF WHEAT. 1000 grams of fine meal obtained by freshly grinding the entire kernel of spring wheat and a like quantity of similar flour from winter wheat were each extracted with 4000 cc. of 10 per cent sodium-chloride solution, and 2500 cc. of the clear extract were obtained from the spring-wheat flour and 2600 cc. from the winter- wheat flour. As 100 cc. of solution were used for every 25 grams of flour taken, the amount of extract obtained was approx- imately equal to that from 625 grams of spring- wheat flour and 650 grams of winter-wheat flour. These extracts were then dialyzed until all the chlo- 1 Kossel & Kutscher, Zeitschrift fur physiologische Chemie, 1900, xxxi, p. 165. 2 Osborne & Voorhees, American Chemical Journal, 1893, xv, 392. EXPERIMENTAL. 101 rides were removed, which required 5 days. The precipitated globulin was then filtered from each, washed with distilled water, alcohol, absolute alcohol, and ether, removed from the filter, and dried at no°. From the spring- wheat extract 3.8398 grams were obtained, equal to 0.624 per cent of the flour, and from the winter wheat 3.9265 grams, equal to 0.625 per cent. The filtrates from the globulin were then heated to 65 ° in a water-bath, and after being held at this temperature for some time the coagulum was filtered off and washed with hot water, alcohol, and ether, removed from the paper, and dried at no0. From the spring wheat 1.97 14 grams were obtained, being 0.315 per cent of the flour, and from the winter wheat 1.9614 grams, equal to 0.302 per cent. The solutions filtered from each of these coagula were next heated just to boiling and the resulting coagulum filtered off, washed thoroughly, and treated as the preceding preparations had been. The spring- wheat extract thus yielded 0.4743 gram, equal to 0.076 percent ; the winter-wheat extract 0.3680 gram, equal to 0.057 Per cent. The two extracts were next concentrated by boiling down over a lamp. They remained clear at first, but when somewhat concentrated the protein began to separate as a skin on the surface of the solution. When reduced to about one-fourth its original volume, the coagulum was filtered off, washed with boiling water, alcohol, and ether, and dried at 1 io°. The spring wheat thus yielded 0.8737 gram> equal to 0.139 per cent ; the winter wheat 0.8721 gram, equal to 0.134 per cent. The filtrates from these two preparations were evaporated very nearly to dryness on water-baths. On cooling, much substance separated, which, when treated with hot water, dissolved again. The insoluble coagulum was filtered from each, washed, and dried in the usual manner. The spring wheat gave 0.8149 gram of substance, being 0.130 per cent of the flour ; the winter wheat 0.5795 gram, being 0.089 Per cent. The total amount of protein coagulating on concentration was there- fore 0.269 Per ceQt for the spring wheat and 0.223 per cent for the winter wheat. The filtrates from these second coagula were then again evaporated to a sirup, and as no more insoluble matter separated they were each precipitated by pouring into strong alcohol. Large precipitates resulted in each case, which, after settling and decantation of the alcohol, were dissolved again in a little water and precipitated by pouring into strong alcohol. Much color- ing matter and sugar was held in solution, as proved by evaporation of the alcoholic mother-liquors. The precipitates were then thoroughly dehydrated with absolute alcohol, washed with ether, and dried at no°. The spring- wheat extract thus yielded 6.9289 grams of substance, the winter wheat 8.7517 grams. As these preparations were unquestionably very impure 102 THE PROTEINS OE THE WHEAT KERNEL. and no practicable method of purification existed which could be carried out without great loss of substance, the nitrogen in each was determined and the protein calculated by multiplying the result obtained by 6.25. The spring-wheat preparation was thus found to contain 3.07 per cent of nitro- gen, equal to 19.19 per cent of protein, and the winter-wheat preparation 5.15 per cent nitrogen, equivalent to 32.18 per cent. The amount of pro- teose and peptone thus found in the extract from the spring wheat was 1.3297 grams, which equals 0.213 per cent of the flour, and in the winter- wheat extract 2.8063 grams, equal to 0.432 per cent of the flour. The sodium chloride extract contained, therefore, the following amounts of protein : Spring wheat. Winter wheat. Globulin P.ct. 0.624 0.391 0.269 0.213 P. ct. 0.625 o-359 0.223 0.432 Albumin Coaguluru Proteose 1-497 1-639 The remainder of the protein matter of the seed forms the gluten. The proportion of gliadin and glutenin in this gluten was determined in the fol- lowing manner : 200 grams of spring-wheat flour made from the entire seed and a like quantity of a similar winter-wheat flour were each made into a dough and thoroughly washed with water as long as starch was removed. The gluten thus obtained, after freeing from loosely adhering moisture, was weighed and exactly one-half dried at 1100 to constant weight. The spring wheat was thus found to yield 12.685 Per cent dry gluten, the winter-wheat flour 11.858 per cent. The other half of the gluten was cut up very fine and extracted as thor- oughly as possible with alcohol of 0.90 sp. gr. The alcoholic extract was then evaporated to small volume, cooled, and the solution decanted from the precipitated protein. This precipitate was then extracted with ether and dried at no°. From the spring-wheat gluten 4.3379 per cent and from the winter wheat 4.2454 per cent of alcohol-soluble protein were obtained. The residue extracted with alcohol was then dried at no° and weighed. The spring- wheat gluten contained 7.80 per cent insoluble in alcohol, the winter wheat 7.504 per cent, reckoned on the wheat. Nitrogen was then determined in these residues dried at 1100, as well as in the dried gluten EXPERIMENTAL. 103 and also in the original flours. The washings of the glutens were collected in jars and allowed to settle, the sediments washed with water and with very strong alcohol and dried and weighed. The nitrogen in each was then determined. The results of these determinations and deductions drawn from them are given in table 20. Table 20. — Amount of the various proteins of the wheat kernel. Total nitrogen in the flour Total gluten in the flour Part of gluten insoluble in alcohol Per cent of nitrogen in gluten Total nitrogen in gluten in per cent of flour Total nitrogen in residue of gluten insoluble in alcohol Total nitrogen extracted by alcohol Gliadin (N X 5-68), assuming 17.60 per cent nitrogen in gliadin Gliadin by direct weighing Nitrogen in sediment from washing gluten Spring wheat. Winter wheat. P.ct. 1 -950 12.685 7.800 12.010 1.5222 0.8245 0.6977 3-9630 4-3379 0.2239 P.ct. 1.940 n.858 7-504 12.000 1.4230 0.7346 0.6884 3.9100 4-2454 O.I552 Glutenin Gliadin Globulin Albumin Coagulum Proteose From H20 washings of gluten Spring wheat. Winter wheat. Nitrogen. P.ct. 0.S245 X 5-68 0.6077 X 5.68 o. 124S 0.0657 0.0453 0.0341 0.2239 X 5 68 2.0050 Protein. P. a. 4.6S3 3-963 0.624 0.391 0.269 0.213 1.272 n.4'5 Nitrogen. P.ct. o.7346 X 5-68 0.6884 X 5 68 o.l 148 0.0603 0.0379 0.0791 0.1552 X 5-68 1.8703 Protein. P.ct. 4-173 3.910 0.625 o-359 0.223 0.432 c.881 10 603 THK FORMATION OF GLUTEN. Wheat, so far as known, is the only plant whose seeds contain protein matter separable in a coherent form from the other constituents by washing with water. When ground fine and mixed with a suitable quantity of water, it yields a dough from which a light, porous bread can be made. The im- portance of this fact in bread-making is so great that considerable attention has been paid to gluten by the chemists who have studied wheat proteins. Reference has already been made (p. 45) to the statements of Weyl & Bischoff that the protein matter of the wheat kernel is chiefly a globulin sub- stance, which in contact with water undergoes a change, presumably through 104 THE PROTEINS OF THE WHEAT KERNEL. the influence of a ferment by which gluten results. If the statements of these investigators are examined no evidence will be found to support their view. What their reasons were for concluding that ' ' myosin ' ' formed nearly all the protein of the wheat kernel does not appear. In view of the results obtained by the writer, this statement is certainly erroneous. Direct treatment of the meal with alcohol yielded extracts containing gliadin in exactly the same amount as obtained from the gluten made from an equal quantity of flour, and extraction of either flour or gluten with alcohol, after complete exhaustion with sodium-chloride solution, also gave the same pro- portion of gliadin. This substance must therefore have existed in the seed, and, as it forms one-half of the gluten, it leaves the other half only as possibly derived from a globulin body through the influence of a ferment. If Weyl & Bischoff's view were correct, treatment of the flour with 10 per cent salt solution ought to alter the character and quantity of the gluten obtained, if not altogether to prevent its formation. This is not so, for the usual amount of gluten can readily be obtained from flour made into dough with ioper cent sodium-chloride solution and then washed with the same until starch is removed. Weyl & Bischoff next state that " with the aid of a 15 per cent rock-salt solution the flour was extracted until no protein could be detected in the extract; the residue of the meal kneaded with water then gave no gluten. If the globulin substance is extracted, no formation of gluten takes place." It has been found that this is true if the flour is stirred up with a large quantity of salt solution, extracted repeatedly with fresh quantities of the same solu- tion until no more protein is dissolved, and the excess of solution removed by allowing the residue to drain on a filter as completely as possible. If, however, wheat flour is mixed at first with just sufficient salt solution to make a firm dough, this dough may be washed indefinitely with salt solution, and will yield gluten as well and as much as if washed with water alone. This difference is due to the fact that when large quantities of salt solution are applied at once the flour fails to unite to a coherent mass and can not after- ward be brought together, as is possible when treated with smaller quantities of solution. Weyl & Bi?choff then compare the formation of gluten to that of blood- fibrin from fibrinogen under the influence of a ferment. Sidney Martin next advanced a somewhat similar theory of the formation of gluten from the proteins contained in the seed. He states that alcohol extracts from gluten but one protein substance ; that this is soluble in hot water, but not in cold, and he therefore calls it an insoluble phytalbumose. The residue of the gluten not dissolved by alcohol is uncoagulated pro- tein, if the alcohol has not been allowed to act too long. This substance he EXPERIMENTAL. I05 names gluten-fibrin. Martin further says that gluten dissolves almost com- pletely in 0.2 per cent hydrochloric acid or 0.2 per cent potassium-hydroxide solution, leaving a small residue of fat. The solution gives a copious pre- cipitate when neutralized, but the supernatant liquid still contains a quantity of protein which is the dissolved insoluble albumose. The whole of the gluten-fibrin is reprecipitated by neutralization — that is, it is wholly con- verted into an albuminate. Martin states that by extracting flour with 76 to 80 per cent alcohol only fat is removed. This statement is certainly erroneous, for the writer has never failed in many experiments thus to extract this substance (gliadin) from the flour, and that, too, in the same amount and of the same properties and composition as from the gluten. Martin concludes that insoluble albumose is not present as such in the flour. He then says : Before proceeding to mention its precursor, it will be well to state that 10 per cent sodium-chloride solution extracts from flour a large quantity of globulin and of albumose. This globulin is of the myosin type, coagulating between 550 and 6o° C. , and precipitated by saturation with sodium chloride and ammonium sulphate. Both the globulin and albumose are present in a much smaller quantity in the watery extract of the flour. From this it is evident that Martin has fallen into the same error as Weyl & Bischoff, mistaking the albumin for a myosin-like globulin, and being greatly misled as to its amount. Continuing, Martin says : The direction of the evidence is to show that the insoluble albumose is formed from the soluble. Moreover, I think that the globulin is transformed into the gluten-fibrin, for I have been able to obtain from the globulin in solution a body having the same reactions as the gluten-fibrin. What this evidence is which by its direction shows that the insoluble albumose is derived from the soluble is not clear, and Martin makes no further statements on this point. That a body should be obtained from the solution containing the globulin which had the same reactions as the "gluten-fibrin" is not surprising, for the insoluble products derived from nearly all globulins have no characteristic reactions, being merely soluble in dilute acids and alkalis and precipitated by neutralization in the same way as " gluten-fibrin." Martin then states his theory of the formation of gluten in the following scheme : Gluten = I Gluten-fibrin — precursor, globulin. " \ Insoluble albumose — precursor, soluble albumose. This can not be a correct representation of the formation of gluten, for it has been shown to be founded on two erroneous observations — first, that alcohol does not extract protein matter from the flour when applied directly, and, second, that at least one-half the protein matter of the seed is a myosin- like globulin. 106 THE PROTEINS OF THiC WHEAT KERNEL. The results obtained by the author and described in this paper have led to the conclusion that no ferment-action is involved in the formation of gluten ; that but two protein substances are contained in the gluten, glu- tenin and gliadin, and that these exist in the wheat kernel in the same form as in the gluten, except that in the latter they are combined with water in an amount equal to about twice the weight of the dried protein. The reasons for this opinion are, first, that alcohol extracts the same protein and in the same amount, whether applied directly to the flour, to the gluten, or to flour previously extracted with 10 per cent sodium-chloride solution ; second, that 0.2 per cent potassium-hydroxide solution extracts glutenin of uniform composition and properties from flour which has been extracted with alcohol or with 10 per cent sodium-chloride brine and then with alcohol as it extracts from gluten which has been exhausted with alcohol. Both glutenin and gliadin are necessary for the formation of gluten, as may be seen from the following experiments : A portion of flour was washed completely free from gliadin by means of alcohol of 0.90 sp. gr., next with stronger alcohol, finally with absolute alcohol, and air-dried. The residue was then rubbed up fine until all lumps were removed, and water caref ully added and a dough made of the mass. A tolerably coherent dough was thus obtained, but much less elastic and tough than that pro- duced from the untreated flour. This dough was then washed with water on a sieve, using every precaution to obtain a gluten, but none was formed. In another experiment 7.5 grams of very finely ground air-dried gliadin were mixed intimately with 70 grams of fine corn-starch and distilled water added. A plastic dough was thus produced, but it had no toughness. On adding a little 10 per cent sodium-chloride solution, the dough became at once tough and elastic. This was then washed with great care on a sieve with cold water, a little 10 per cent sodium-chloride solution being added from time to time, but in spite of every precaution no gluten was obtained. The following experiment shows that the gliadin used was capable of forming gluten when glutenin was present, and also that salts have a marked influence on the toughness of the resulting dough. Two portions of flour weighing 100 grams each were taken, and after adding 5 grams of gliadin to one both were made into dough with the same quantity of water. The two doughs presented marked differences. That to which gliadin had been added was much tougher and more yellow than the other. They were then washed with water as long as starch separated. The gluten was dried superficially by wiping with a cloth and weighed in the moist state. That from 100 grams of flour to which 5 grams of gliadin had been added weighed 44.55 grams ; that from 100 grams of flour alone weighed 27.65 grams. The moist glutens were dried at 1 ro° to constant weight, and both yielded EXPERIMENTAL. 107 the same proportion of dry gluten, viz, 34.6 per cent. The yield of dry- gluten was accordingly in the first case 15.41 grams and in the second 9.56 grams. The difference, 5.85 grams, shows that the added gliadin was fully recovered in the gluten. The figures show that these proteins combine with about twice their weight of water in forming gluten. The fact that the added gliadin entered so readily and completely into the formation of gluten indicates that it exists in the seed as such and undergoes no chemical change during extrac- tion and drying. The properties observed in testing the separated gliadin show how it acts in forming gluten and explain many of the points observed by others and attributed to a ferment-action. When treated with distilled water in small amount, the fine-ground air-dry gliadin at once forms a sticky mass, which, on adding more distilled water, dissolves to a turbid solution. If, however, a very little sodium chloride is added to distilled water and this applied to gliadin that has been first moistened with pure water, a very coherent, viscid mass results, which adheres to everything it touches and can be drawn out into long threads. If the gliadin is moistened with 10 per cent sodium-chloride solution and then treated with a larger quantity of this solution, the substance unites to a plastic mass, which can be drawn out into sheets and strings, but is not adhesive. From this it is evident why Ritthausen, in washing flours which gave a fluid gluten, obtainable only in small quantity and with great diffi- culty, found that the addition of calcium sulphate to the wash-water ren- dered the gluten much more coherent and easily obtainable. The gliadin is thus proved to be the binding material which causes the particles of flour to adhere to one another, thus forming a dough. But the gliadin alone is not sufficient to form gluten, for it yields a soft and fluid mass, which breaks up entirely on washing with water. The insoluble glutenin is probably essen- tial by affording a nucleus to which the gliadin adheres and from which it is not mechanically carried away by the wash-water. The behavior of the gliadin toward 10 per cent sodium-chloride solution shows why no gluten was obtained by Weyl & Bischoff from flour extracted with this solvent. The gliadin had under these conditions no adhesive quali- ties, and therefore was unable to bind the flour into a coherent mass. If, however, the salt solution is added in small quantities and the flour kneaded and pressed, the particles are brought together and then adhere tenaciously. IOS THE PROTEINS OF THE WHEAT KERNEL. SUMMARY. The proteins of the wheat kernel are gliadin, insoluble in neutral aqueous solutions, but distinguished from all the others by its ready solubility in neutral 70 per cent alcohol ; glutenin? a protein having a similar elementary percentage composition to gliadin, soluble in very dilute acid and alkaline solutions, but insoluble in dilute alcohol or neutral aqueous solutions and yielding a wholly different proportion of decomposition products when boiled with strong acids ; leucosin, an albumin-like protein, freely soluble in pure wrater and coagulated by heating its solution to 500 to 6o° ; a globulin similar in composition and properties to many globulins found in other seeds, and one or more proteoses which are present in very small quantity. It has also been shown that the proteins obtained from the embryo of the wheat are the globulin, albumin, and proteose above mentioned, and that these form nearly all of the protein substance of this part of the seed. It thus appears that these three proteins are contained chiefly in the embryo, and that gliadin and glutenin form nearly the whole of the proteins of the endo- sperm, or over 80 per cent of the total protein matter of the seed. It is possible that a part of the albumin, globulin, and perhaps minute quantities of the proteose are contained also in the endosperm, for these proteins are always found in flour from which, in the milling process, the embryo is very nearly completely separated. The uncertainty, however, as to the completeness of this separation makes it questionable wmether or not the small amount of these proteins found in the best flour is not due to the presence of more or less of the embryo that escaped separation in the milling process. The flour of wheat differs from that of other seeds in forming a dough when moistened with sufficient water, which, when washed with more water, loses its starch, and finally yields a tough elastic mass, long known as wheat gluten. This gluten contains the greater part of the protein matter of the seed, together with a little starch, fat, lecithin, and phytocholesterin, and possibly some carbohydrate substance or substances of as yet unknown char- acter. These non-protein substances are probably not united with one another in the gluten, but are mechanically mixed. The quantity of starch that remains in the gluten depends on the thoroughness of the washing, wmile the other substances owe their presence largely to their insolubility in water. The chief constituents of the gluten are the two proteins, gliadin and glu- tenin, the relative proportions of which vary with the variety of wheat from 1 This is the protein which Ritthausen called " gluten-casein." SUMMARY. IO9 which the flour is made. The character of the gluten and the commercial value of the flour depend to a large extent on the proportion of gliadin to glutenin. In the moist gluten these proteins are present combined with about twice their weight of water, which is gradually lost on exposure to dry air or at an elevated temperature. The gliadin and glutenin are present as such in the seed and are not, as was formerly supposed, derived from other protein substances through the action of an enzyme. This is shown by the fact that they may be obtained directly from the flour by the same treatment as that which yields them from the gluten and under conditions which preclude the action of an enzyme. GLIADIN. Gliadin is the most important of the five proteins above mentioned, not only on account of its influence on the character of the gluten, and therefore on the quality of the flour for domestic purposes, but also on account of its unusual physical properties and chemical constitution. Gliadin, unlike most other protein substances, is freely soluble in relatively strong ethyl alcohol. Although gliadin is wholly insoluble in absolute alcohol and but slightly soluble in water, it dissolves in dilute alcohol, the solubility increasing with increasing concentration of alcohol until a certain strength is reached, when the solubility diminishes until, by absolute alcohol, it is no longer dissolved. Exactly what strength of alcohol dissolves the largest proportion of gliadin has never been determined, but the maximum solubility is attained with about 70 per cent of alcohol by volume. Gliadin is also soluble in other alcohols, as methyl, propyl, and benzyl alcohols, in phenol and paracresol, and also in glacial acetic acid. Gliadin is somewhat soluble in pure water, but less so in water containing salts, though not wholly insoluble in solutions containing 10 per cent of sodium chloride. Very dilute acid or alkaline solutions dissolve gliadin to solutions which, by neutralizing, yield the gliadin apparently unchanged. By stronger solutions of acids or alkalis the gliadin is altered, as is the case with all other native proteins. All the usual color tests given by other protein substances are obtained with gliadin, which therefore contains the groups that give rise to these re- actions. Gliadin is much less easily converted into insoluble products than are most other proteins. Its solution in 70 per cent alcohol can be boiled for an indefinite time, and even concentrated until much of the alcohol has been removed, without forming insoluble products. On heating with very weak alcohol or with water this protein is gradually altered and becomes insoluble in stronger alcohol, but the coagulated gliadin thus formed is in no THE PROTEINS OF THE WHEAT KERNEL. appearance unlike the heat coagulum formed by most other proteins. Pro- teins of similar behavior to gliadin are found in the seeds of other cereals, such as rye, barley, maize, oats, and sorghum. That found in rye is prob- ably identical with the gliadin of wheat, for a rigid comparison has not yet revealed any difference. Both have the same ultimate composition, the same solubility, the same physical properties, and yield the same amount of ammonia and glutaminic acid on hydrolysis. The identity of the two, how- ever, is not certain, and with our present knowledge can not be established. The proteins of the other cereals above mentioned are distinctly different, though similar, proteins. Proteins characterized by such ready solubility in strong alcoholic solu- tions as are those found in these cereals have not been obtained from the seeds of any other plants. The ultimate composition of gliadin has been fixed within narrow limits by the accordant analyses of several investigators as follows : Giiusberg. Ritthausen. Osborne & Voorhees. Nasmith. Konig & Rintelen. Carbon Hydrogen . . Nitrogen Sulphur P.ct. 52.67 6. S3 17.62 P. Ct. 52.76 7. 10 18.01 0.85 21.28 P. Ct. 52.72 6.86 17.66 1.03 21.73 P. Ct. 52.39 6.84 17-47 1. 12 22.18 P.ct. 52.70 7.62 17.77 095 20.96 Oxygen. .... 22.88 100.00 100.00 100.00 100.00 100 00 Gliadin yields on hydrolysis a larger amount of ammonia, glutaminic acid, and proline than any other protein yet examined, no glycocoll or lysine, and a relatively small amount of histidine and arginine, as the following analysis shows : Products of hydrolysis of gliadin. P.ct. Glycocoll 0.00 Alanine 2.00 Amino-valerianic acid 0.21 Leucine 5.61 Proline 7.06 Phenylalanine 2.35 Aspartic acid 0.58 Glutaminic acid 37-33 Serine o. 13 P.ct. Tyrosine 1 . 20 Cystine 0.45 Lysine 0.00 Histidine 0.61 Arginine 3. 16 Ammonia 5.11 Tryptophane present Total 65.81 Although gliadin gives a strong Molisch reaction, which is commonly con- sidered to indicate the presence of a carbohydrate group, it gives no furfurol when distilled with hydrochloric acid, as does ovalbumin, which has been SUMMARY. Ill proved to contain such a group. Other evidence than that given by the Molisch reaction is required before the presence of a carbohydrate group in this protein can be assumed. Nearly two-thirds of the total sulphur in gliadin is split off as sulphide by boiling with alkalis, which would indicate the possibility that all the sulphur is contained in a cystine complex. The amount of cystine isolated is far below that required by such an assumption. The determination, however, of this substance is not quantitative, and the figure given does not necessarily prove that all of the sulphur may not be contained in a cystine- yieldiug complex. The specific rotation of gliadin in 80 per cent by volume ethyl alcohol, ac- cording to determinations by the writer, which agree closely with those of 20° others, is («) -^- = — 92.30. The specific rotation in solvents of alcoholic nature has been recently determined by Mathewson, who gives the following data : («) 40° D Methyl alcohol, 70 per cent. — 95. 650 Ethyl alcohol, 70 per cent. — 91-95° Ethyl alcohol, 60 per cent. — 96. 66° Ethyl alcohol, 50 per cent. — 98.450 Propyl alcohol, 60 per cent. — 101.100 »£ Phenol, 70 per cent — 123. 15 Phenol, anhydrous — 131.77 Paracresol — 121.00 Benzyl alcohol Glacial acetic acid. — 53-ioc — 78.6oc The amount of gliadin in different varieties of wheat differs to a large extent, as shown by a considerable number of determinations that have been recorded. Thus Teller found in a sample of Canadian white winter wheat 2.74 per cent, in one of Oregon white winter wdieat 2.85 per cent, and in a red spring wheat from South Dakota 8.15 per cent. In other varieties of wheat he and others have found quantities falling between these figures. Although the total quantity of gliadin in the sample of red wheat above mentioned was nearly three times as great as in that of the white wheat, its percentage of the total proteins was much more uniform, being 42.5 and 34 per cent respectively. This appears to be generally true for most of the wheats that have been examined. Thus Shepard has found in 13 samples of durum wheat that the gliadin formed from 40.2 to 48.6 per cent of the total protein (N X 5.7), in 9 sam- ples of Northwestern spring wheat from 44.2 to 49 per cent, in 4 samples of Kansas hard winter wheat from 40 to 50.7 per cent, and in 3 samples of soft winter wheat from 40 to 47.4 per cent. The figures given by others are similar, though a few fall between somewhat wider limits. It would seem safe to say that in the majority of wheats the gliadin forms from 40 to 50 per cent of the total proteins, and that its absolute amount depends chiefly on the total protein content of the seed. 112 THE TROTEINS OF THE WHEAT KERNEL. GLUTENIN. This protein was first described by Taddei under the name of zytnom. Iyiebig, as well as Dumas & Cahours, named it " plant-fibrin ; " Ritthausen called it ' ' gluten-casein ; ' ' Weyl & Bischoff considered it to be an albumi- nate form of a myosin-like globulin. On the ground of priority and the fact that the relations to animal proteins, which gave rise to the various names subsequently applied to this body, have been proved to have no foundation, it would be desirable to return to Taddei 's original name and in future call this protein zymom. Unfortunately this name is derived from the Greek word tofty, a ferment, and as the results of this and all other subsequent investigations of this subject show that the supposed ferment-changes do not occur in the formation of gluten, it seems appropriate to call it glutenin, a name suggested by S. W. Johnson. Glutenin is next in importance to gliadin, for in most varieties of flour it is present in nearly equal amount. It is so nearly insoluble in water and alcohol that it is a question whether the slight traces that are dissolved by these solvents from carefully purified prep- arations are not due to traces of gliadin, which it is very difficult to separate from it. In hot dilute alcohol glutenin is slightly soluble and separates from such solutions on cooling. When freshly precipitated and in the hydrated condition, it is very readily dissolved by extremely dilute acids or alkalis and is precipitated from such solutions on neutralization. Glutenin can be separated from the other constituents of the seed only by solution in dilute alkalis or acids, and it is necessary to filter the solutions, from which it is finally precipitated, perfectly clear in order to separate the associated non-protein substances. This filtration is accomplished with great difficulty unless it is preceded by thorough extraction of the crude glutenin with alcohol and ether, whereby fats and lecithins are removed. Glutenin contains all the groups which give rise to the usual color reac- tions of the proteins. No protein similar to glutenin in physical and chem- ical properties has yet been found in any other seed. Although its ultimate composition is nearly the same as gliadin, the proportion of the various products of hydrolysis which it yields is very different, as may be seen from the following analyses : The ultimate composition of glutenin, P. ct. Carbon 52.34 Hydrogen 6.83 Nitrogen 17-49 Sulphur 1.08 Oxygen 22.26 100.00 SUMMARY. 113 The following products of hydrolysis have been obtained by boiling glutenin with strong hydrochloric acid : Products of hydrolysis of glutenin. P.ct. Glycocoll 0.89 Alanine 4-65 Amino-valerianic acid 0.24 Leucine 5.95 ^-proline 4.23 Phenylalanine 1.97 Aspartic acid 0.91 Glutaminic acid 23.42 Serine 0.74 P.ct. Tyrosine ... 4.25 Cystine 0.02 Lysine 1.92 Histidine 1.76 Arginine 4.72 Ammonia 4-nI Tryptophane present 59.66 Glutenin is qualitatively distinguished from gliadin by yielding both glycocoll and lysine, and quantitatively by yielding less proline, glutaminic acid, and ammonia and more alanine, tyrosine, and arginine. In other respects the differences are not great. The amount of glutenin varies greatly in different samples of wheats, but usually forms about 40 per cent of the total protein of the seed. Teller has found in a sample of Canadian white winter wheat 3.64 per cent of glutenin and in an Oregon white winter wheat 3.82 per cent, while in a red spring wheat from South Dakota he found 8.04 per cent. Although the total amount of glutenin in these wheats was very different, the relative amounts were nearly the same, being 45.2 and 46.5 percent of the total proteins in the white winter wheats and 42 per cent in the red spring wheat. Other winter wheats differed but little from other spring wheats both in the total and relative amount of glutenin which they contained. In most of the wheats that he examined the glutein formed from 40 to 45 per cent of the total proteins. Shepard, who analyzed a much larger number of samples, found a wider variation in the relative proportion of glutenin, from 34.2 to 46.8 per cent of the total proteins, and a narrower range in its total amount, the latter falling between 3.41 and 7. 10 per cent. LEUCOSIN. Leucosin forms about 0.3 to 0.4 per cent of the wheat kernel. Although the entire seed contains but little leucosin, the embryo contains a relatively large proportion, since about 10 per cent of the commercial " germ meal " consists of this protein. As this ' ' germ meal ' ' contains more or less of the endosperm and outer coats of the seed, the embryo contains somewhat more than 10 per cent. Leucosin is an albumin, for it is soluble in pure water and coagulated by heating its solution. It has been mistaken by several observers for a myosin-like globulin, owing to the fact that its temperature of coagulation falls near to that of such globulins found in animal tissues, 8 114 THE PROTEINS OE THE WHEAT KERNEL. and also to the fact that its solutions are precipitated by saturating them with sodium chloride or with magnesium sulphate. That it is soluble in pure water and not held in solution by small quantities of salts is shown by the fact that a solution containing a considerable quantity of leucosin when sub- jected to prolonged dialysis and then evaporated to dryness, and the leucosin burned off at a low temperature left a residue of mineral matters weighing less than a milligram. Leucosin is much more readily precipitated by am- monium sulphate than are the albumins of animal origin, for by adding an equal volume of a saturated solution of this salt to solutions containing leu- cosin the latter is almost completely precipitated. Leucosin resembles the animal proteins in ultimate composition, in the proportion of its products of hydrolytic decomposition, and in its physical properties more closely than it resembles most of the seed proteins yet studied. It seems not improbable that this is because it is a constituent of the tissues of the embryo, for its physiological functions are unquestionably different from those of the stored-up food proteins of the endosperm. Proteins having the same ultimate composition and, so far as known, the same properties as leucosin are found in the seeds of other cereals. Whether these are identical or not can not be determined by any means now available. The ultimate composition of leucosin is shown by the following figures, which are the average of accordant analyses of several preparations : Composition of leucosin. P.ct. Carbon 53-02 Hydrogen 6.84 Nitrogen . 16.80 Sulphur 1.28 Oxygen 22.06 100.00 Leucosin yields the following amounts of the several products of hydrolysis when boiled with strong hydrochloric acid : Products of hydrolysis of leucosin. P. ct. Glycocoll 0.94 Alanine 4 . 45 Amino-valerianic acid o. 18 Leucine 1 1 . 34 a-proline 3.18 Phenylalanine 3.83 Aspartic acid 3-35 Glutaminic acid 6 . 73 P. ct. Tyrosine 3-34 Lysine 2 . 75 Histidine 2 . 83 Arginine : 5-94 Ammonia 1 .41 Tryptophane present 50.32 Leucosin probably contains a relatively large proportion of tryptophane, for in a series of comparative tests made with a large number of different SUMMARY. 115 proteins the intensity of its glyoxylic acid reaction was greater than that of any of the other proteins. L,eucosin also gave the strongest reaction with the Molisch test of any of a large number of different proteins examined under like conditions, but whether or not it contains a carbohydrate group must be shown by other evidence. THE GLOBULIN. The seeds of wheat contain about 0.6 per cent of a protein which is insol- uble in water, but soluble in neutral saline solutions. This globulin is chiefly contained in the embryo from which 5 per cent was extracted by sodium- chloride solution. The preparations of the globulin that were made from the embryo contained nucleic acid, while those from the whole seed con- tained none. These nucleic acid compounds from the embryo behaved like nucleates, for their proportion of nucleic acid was not constant, but varied with the conditions of preparation. It seemed most probable that the nucleic acid combines with the basic protein to form salts in the same manner as other acids are known to do, and that such combinations with the globulin still retained the solubility characteristic of the globulin. It is quite possi- ble that other insoluble compounds containing more nucleic acid may have existed in the embryo or have been formed during extraction, so that the amount of globulin obtained may not have equaled that actually present in the embryo. The fact that the globulin from the whole seed was free from nucleic acid, while that from the embryo was not, was probably due to the presence of a much larger proportion of protein insoluble in salt solution in the whole seed compared with that in the embryo, so that the nucleic acid united with the insoluble protein instead of with the globulin, as happened when the embryo was extracted. This globulin is very similar to, if not identical with, that found in the seeds of rye and barley. It contains over 18 per cent of nitrogen, and re- sembles, in composition and properties, many of the globulins found in large proportion in many other seeds. Owing to the difficulty encountered in preparing large quantities of this globulin, the products of its hydrolysis have not been studied. Its ultimate composition is shown by the following figures, which are the average of several analyses of preparations from the whole seed : P. ct. Carbon 5x-03 Hydrogen 6.85 Nitrogen 18.39 Sulphur o. 69 Oxygen 23.04 100.00 n6 THE PROTEINS OF THE WHEAT KERNEL. This globulin is precipitated by saturating its solutions with magnesium sulphate, but not with sodium chloride. Dissolved in 10 per cent sodium- chloride solution, it is partly coagulated by boiling, but is not coagulated at temperatures below ioo°. THE PROTEOSE. The wheat kernel yields a very small amount of proteose when extracted with water. Whether this is an actual constituent of the seed or is formed from the other proteins during its extraction and isolation was not definitely ascertained. This proteose, like leucosin and globulin, is chiefly found in the embryo. No preparations were made from the entire seed in sufficient quantity to permit of analysis or examination of its properties. The proteose obtained from the embryo appeared to be a mixture of two or more substances. A part was precipitated by saturating its solution with sodium chloride ; a part was not. The analyses of these two parts showed the low percentage of carbon characteristic of proteoses obtained by peptic digestion. Analyses of prepara- tions thus obtained gave the following results : Proteose in- soluble in saturated NaCl solution. Proteose soluble in saturated NaCl solution. Hydrogen Nitrogen P.ct. 49-94 6.80 17.08 I.24 24.94 P.ct. 48.99 6. 85 1689 1. 10 26.17 IOO.OO 100.00 THE GLUTEN. The proteins of the wheat kernel differ from those that have been found in any other seed by the fact that they may be largely separated as a coherent elastic mass by washing the dough made from the flour under a gentle stream of water. This protein mass has long been known as gluten. It consists chiefly of gliadin and glutenin in combination with about twice their weight of water, together with more or less starch, which can not be wholly removed by washing, and also some fat, cholesterin, and lecithin. The non-protein constituents usually form about 20 to 25 per cent of the crude gluten. The protein constituents of the gluten, except for the fact that they are combined with water, are present in the same form in the gluten as in the grain. SUMMARY. 117 Their proportion is also nearly the same, for practically the same amount of gliadin can be obtained from the gluten as from the flour, if allowance is made for the small amount of gluten that is mechanically carried away in the process of washing out the starch. This loss is largely made up by the presence of non-protein constituents in the gluten, so that the weight of the dried gluten usually corresponds closely with the amount calculated for the sum of gliadin and glutenin, as computed from the nitrogen belonging to these proteins. The gluten is not formed from globulins by the action of a ferment in a manner analogous to the formation of fibrin from fibrinogen, as was asserted at one time. The observations on which this supposition was founded were incorrect. The glutenin probably forms the nucleus to which the gliadin adheres, and thus binds the gluten proteins into a coherent elastic mass. Both gliadin and glutenin are necessary for the formation of gluten, for a dough made with starch and gliadin, or one made with flour from which the gliadin has been extracted with alcohol, yields no gluten when washed with water. That the gliadin is capable of taking part in gluten formation under such conditions is shown by the fact that when dry and finely ground glia- din was added to wheat flour the amount of gluten obtained on subsequently washing out the starch was increased by the full amount of the gliadin that was added. THE NUTRITIVE VALUE OF THE WHEAT PROTEINS. It has recently been shown that in the process of digestion the protein molecule is very largely broken down into amino-acids, and that the animal forms from these, by the processes of assimilation, the proteins of its blood and tissues. How this change is effected is not known, nor is it known whether the food protein is converted into the body protein and then ox- idized and eliminated, or is partly converted into body protein and partly burned directly in the form of amino-acids. It is also a question whether or not the animal has the power to convert one amino-acid into another, and thus obtain material suitable for the construction of its own body protein. Directly connected with these important problems are the facts presented by the determination of the relative amounts of the different amino-acids yielded by the proteins of wheat flour, for these are used in enormous quan- tities as food by man, and, as an examination of the analyses of their decom- position products show, present marked differences from similar analyses of all the other food proteins thus far examined. As gliadin and glutenin together form about 85 per cent of the proteins of wheat flour, they deserve especial consideration in this respect. If it is n8 THE PROTEINS OF THE WHEAT KERNEL. assumed that these two proteins are present in equal quantities in wheat flour, an assumption that in most cases is nearly correct, the relative propor- tion of the amino-acids yielded by the proteins of wheat flour is closely shown by the following figures, which are the mean of those found for gliadin and glutenin. In comparison with these are given the figures, at present available, for some of the other forms of food protein. Such a com- parison is at present unsatisfactory, owing to the lack of data now available, but it will serve a useful purpose in indicating the wide differences between these food proteins, and will emphasize the importance of obtaining more information in regard to this question. Products of hydrolysis of some food proteins. Glycocoll Alanine Amino-valerianic acid. Leucine Proline Phenylalanine Aspartic acid Glutaminic acid Serine Tyrosine Cystine Lysine Histidine Arginine Ammonia Tryptophane Wheat gluten. Milk casein. Egg albumin. Zein, maize. Excel- sin, Brazil nut. Phase- olin, ■white bean. Beef muscle. Halibut muscle. P.ct. 0.44 3-33 0.23 5-78 5-65 2.16 o.75 30.38 0.44 2-73 0.24 0.96 1. 19 3-94 4-56 present P.ct. 0.00 0.90 I. OO 10.50 3-IO 3.20 I.20 11.00 0.23 4-50 0.06 5-8o 2-59 4.84 i-95 1.50 P.ct. O.00 2.IO P.ct. P.ct. O.60 2-33 8.70 3-65 3-55 3.85 12.94 P.ct. 0.55 I.80 I.04 956 2.77 3-25 5-24 14-54 0.38 2.17 P.ct. P.ct. 6.IO 2.25 4.40 I.50 9.10 16.87 II. I 8.9 I. IO 0.20 I.63 present 3.16 O.OO 0.81 I.82 3.61 O.OO 1.64 1-47 16.02 1.80 present 3-59 1.97 4.72 2.06 present Owing to the lack of data, it is not yet possible to compare the relative proportion of the amino-acids which the food proteins yield except in respect to glutaminic acid, ammonia, arginine, histidine, and lysine. The amount of glutaminic acid which the gluten proteins yield is far greater than that yielded by any of the other food proteins, with the exception of gliadin from rye and hordein from barley. The proteins of the legumes and nuts which are used as food yield from 15 to 20 per cent of glutaminic acid, so that the mean amount of this amino-acid from the wheat protein is nearly twice as large. The same also is true of ammonia. The proportion of arginine from wheat gluten is relatively small compared with that from most other seed proteins, many of which yield from 10 to 16 per cent of this base. SUMMARY. 119 The proportion of lysine is likewise small, especially compared with that obtained from the leguminous seeds. The amount of histidine, however, does not differ very greatly from that of the other seed proteins. What significance these differences have in respect to the nutritive value of these different proteins must be determined by future investigation, for it has only very recently been discovered that such differences exist. That a molecule of gliadin can have the same nutritive value as one of casein would seem impossible if one molecule of food protein is transformed into one of tissue protein, for in the former lysine is wholly lacking, and glutaminic acid, ammonia, and proline are in great excess over the amount required to form any of the tissue proteins of which we know. It would seem probable that either the animal requires a variety of food, so that the relative proportion in which the amino-acids are available for its use shall correspond more nearly to its requirements, or that only a small part of these amino-acids are converted into its tissue proteins and the rest oxidized as such. It is possible that feeding experiments with proteins of known character in respect to the relative proportions of their decomposition prod- ucts will throw light on these important questions. w MBL WHOI LIBRARY wh ifin y. D H- A 7 lUl " m ■ SEE* K ££22= KS ii Sssssg ■■■.'-• SBsagaga pang