358 THE ANALYST. ON THM NITROGEN COMPOUNDS PRESENT I N THE CEREALS. By G. W. WTGNEB, F.C.S. (Third Pqer.) IN the previous notes on this subject I omitted to notice that the determinatione of nitrogen had all been made by the soda-lime process. When I commenced the in- vestigation I was of opinion that the results obtained by this process were sufficiently correct for the determination of such nitrogenous eubstances as are met with i n the cereals, provided of course that the proper precautions were taken, and especially that the quantity of the grain burnt was not too large.Working in this way on 10 or 15 grains of the sample, I had found that with proper care the proportion of gluten shown by this process varied as the maximum error less than 0.2 in different analyses of the same sample.But on carrying these investigations further I am convinced that there are certain disturbing circumstances due to the samples themselves which render the results liable to a greater error than this. No alteration in the mode of procedure in the analysis is sufficient t o account for some of the differences found, and I cannot a t present trace any relation between the non-coagulable constituents and the irregular results, but the fact remains that I have found certain samples which show in some cases nearly one per cent. more or less albuminous matters by the soda-lime process than I should have expected them to do, and yet while rcpeats give nearly identical results, the percentages found diff ereci materially from those given by the absolute nitrogen process.It seems to me probable that some of the non-coagulable constituents may be the cawe of this disturbing action, but a t present all I can say is that the residues which haw been coagulated hy acid give results which are more uniformly correct than those which are obtained from the original meal or flour. It may be possible for me on a future occasion to point out the exact differences i n the resultp of every sample, but for the present I give the results by soda-lime process only.THE ANALYST.359 Before passing to the results obtained from the analysis of the flour it will be convenient if I give some more general particulars as to these samples. They were, as I have said before, representative not average ones, and were selected from soils of different kinds, and from good and bad crops.Taking the weight per bushel as the standard of comparison most generally referred to, I have the following figures, for which I am again indebted to my friend, Professor Tanner. I also give in another column a determination of the actual weight of 100 grains or kernels of the sample. These weights were taken in such a way that I think the results are quite correct, and they may furnish some clue to the variations found between the different samples.I also give the specific gravity of the whole grain determined by direct weighing. Mark. Weight in lbs. per Specific gravity of ~~g~~~~~~~ bushel. whole grain, A ... 66 ... 1.3G B .*I 61 ... 1.26 C ... 67 ... 1.33 D ... 66 ... 1.31 E ...67 ... 1-24 F ... 65h ... 1.28 G ... 66 ... lt39 H ... 63 ... 1.31 J ... 63 ... 1-28 K '.. 64 ... 1.27 L ... 63 ... 1.25 Ed ... 64 ... 1.24 N ... 62 ... 1.27 0 ... 64 ... 1.30 P ... 64 ... 1.35 The weight of the corns will be seen to Far7 greatly, 100 to 81.8 grains per 100, the average of the whole is 100 corns. of wholeugrain. ... 69.2 ... 65.5 ... 66.2 ... 60.4 ... 44.1 ...63.8 .. . 67.0 ... 78.1 ... 81.2 0.. 81.8 ... 80-9 ... 46.1 ... 63.2 .I. 63.7 ... 65.0 namely, from 44.1 grains per 65.75 grains as the weight of As a further guide in considering the results it may be noted that Proft?ssor Tanner has kindly informed me that sample L was a very large crop, and sample E was a small crop ; sample A was a very fine white wheat; sample B was harvested during very bad weather ; sample J was grown on clay land after clover, and sample H on gravelly soil after clover.As I have already referred to the results obtained from the whole meal and bran, I have next to consider those obtained from the samples of flour. It is necessary to bear in mind that these samples of flour were made by grinding the wheat by hand in a coffee mill, and dressing through a sieve with 80 holes per linear inch.The result of this mode of grinding and dressing is to mix the flour with pulverized bran to a much larger extent than is the case with commercial samples of flour which have been ground between stones and dressed through silk. The soda-lime combustions of these samples were dl made on small quantities of flour, as I found the results obtained from say 10 grains more uniform than those obtained when larger quantities were taken.Slow combustion seemed to give the highest and most consistent results, and was therefore adopted, SO that most of the samples were more than an hour and a half in the furnace, which is Griffin's gas. The samples were coagulated by carbolic acid, acidified with a few drops of metaphosphorio acid solution, and the filtrates saved for subsequent examination.360 THE ANALYST.The following were the results obtained, the nitrogenous constituents being calculated as before by multiplying the nitrogen found by 6.33 ; the results are given in percentages :- FLOUR. Percentage of true m a ~ ~ r ~ ~ ~ ~ ~ t e d Nitrogenous matter gluten, calculated mat teru.by carbolic acid. not coagulated. on total nitrogenous matters. Total nitrogenous Mark. A B C n E F G H J K L M N 0 P ... ... ... ... ... ... ... ... ... ... ... ... ... ..* 12.97 7-97 7.85 7-60 8.86 8.73 9.43 8.86 9.49 8.41 7.78 11.28 8.40 9 18 12 13 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 11.64 7.32 7.60 7.34 8.10 8.10 7.59 8.35 6 * i 0 7.72 6.73 10.12 7.88 8.67 11-13 ... ...... ... ... ... ... ... ... ... ... ... ... ... ... 1.33 *65 -25 *26 -76 -6 3 1-54 - 5 1 2-79 #69 1.05 1-16 *62 -6 1 1.00 ... ... ... ... ... ... ... ... ... ... ... ... ... ... 89.7 91.8 97-8 96.6 91-4 92.8 80.5 94.2 70.6 91.8 86.5 89.6 9 3.8 93.4 92.0 The most discordant results are those shown by the J samples. Now on referring to TRE ANALYST, pages 305 and 306, it will be seen that this sample of wheat showed in the whole meal the smallest but one percentage ratio between the true and non- coagulable albuminoids, namely 73.7 per cent., and that the bran from i t showed a high ratio, namely 88.0 per cent.The proportion of flour obtained from this sample waB 82.5 per cent. leaving bran 17.5 per cent. Now multiplying the results up wc hare :- Bran .. . . . . . . . 17.5 X 88.0 per cent. = 15.40 per cent. Flour . . . . . . . . . 82.5 X 70.6 ,, = 58.24 ,, Total . . . . . . . . . 73-64 ,, Found in whole meal ... 73.70 ,, - This shows very good agreement in the results inter se, milch better than prevails in some other cases. In the G sample, which shows the next smallest percentage, the figures worked out as above show by calculation from the bran and flour 83.3 per cent, of the nitrogenous matter present as true gluten, and the whole meal result was 83.9 per cent.I n the case of the highest percentage the bran sample of C, after washing with carbolic acid, was spoilt, but the whole meal result was also one of the highest. It mill be more useful t o show these results next in another form, putting side by eide the results obtained from the whole meal, bran and flour.PEUCENTAGE OF TRUE GLUTEN CiLCULATED ON THE TOTAL NITHOGENOUS OONSTITUENTS. Mark, Whole meal. Bran. Flour. A ... 87.9 ... 42 4 ... 69.7 B ... 80.9 ... 59.6 91.8 C ... 92.5 ... - ... 97'8 1) ... 91.9 ... 40.8 96.6 E ... 65-9 ... 76.6 91.4 F ... i6.4 I.. 86.8 ... 92.8 G ... 83.9 ... 89.8 .-. 80.5 H ...81.0 ... 70.1 ... 94.2 J ... 73.7 ... 88.0 ... 70.6 K ... 79.4 ... 64.3 ... 91.8 L ... 74.0 ... 66.4 ... 86.6 M ... 90.0 ... 78.8 ... 89.6 N ..* 95.3 ... 81.5 ... 93.8 0 ... 76 2 ... 68.1 ... 93.4 P ... 79.8 .*. 65.7 .*. 92.0THE ANALYST. 36 1 It will be seen that the percentage ratio of true gluten found in samples M and N* is greater in the whole meal than in either the bran or flour when sifted.It seems impoasible to explain this result on any other supposition than the uncertainty of the soda-lime process already referred to. Samples G and J give a higher ratio of true gluten in the bran than in the flour, and they are the only samples which do so. The average of all these ratios is as follows :- They have already been referred to. Whole meal .. . . . . . . . . . . . . . . . . . . . 81.9 per cent. Bran . . . . . . . . . . . . . . . . . . . . . . . . 68-1 ,, Flour . . . . . . . . . . . . . . . . . . . . . . . . 90.2 ,, 1 have in each case determined the nitrogen in the form of nitrates and nitrites in the filtrate from the carbolic acid process. To render these results readily comparable with the previous ones I must now give the percentages of actual nitrogen found by combustion side by side with those found existing as nitrates and nitrites :- BRAN.N. as nitrates Mark, Total N. by Coamlable N. and nitrites by Difference= combustion. by combuation Difference* aluminium N* in Other forms. process. A 1.. B ... c ... D ... E ... F ... G ... H ... J ... K ... L ... M ... N ... 0 ... P ... 2.1 19 2.147 1.858 1.369 1.173 1.473 1-669 1.738 1.840 1.587 1.775 1.300 1-460 2.473 1 * i O O ...... ... ... ... ... ... ... ... ... ... ... ... ... ... 0.900 1.280 0.560 0.900 1,280 1.500 1-219 1.618 1.020 1.178 1'340 1-059 0.769 1.378 - ... ... .., ... ... ... *.. ... ... ... ... I.. ... ... ... 1.219 0,867 0,809 0.273 0.193 0.169 0.519 0,222 0.567 0.597 0.360 0.241 0-70 1 1.095 - ... ...... ... ... ... ... ... ... ... ... ... ... ... ... 0.064 0.0Sl 0.017 0.016 0 055 0,028 0.016 0.037 0.037 0.022 0.022 0-020 0.015 0.015 0.021 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 1.255 0.846 0.793 0.218 0.165 0'153 0.482 0.185 0.545 0.575 0.340 0-226 0.686 1.074 -_ These results show an average of 0.027 per cent. of nitrogen as nitrates and nitrites still unaccounted for, and although there are considerable variations in the samples, the maximum proportion of -064 per cent.is still only a small one. Taking the flours in the same way we get the following results :- FLOUR. Mark, Total N. by Coagulable N. combustion. by combustion 4 ... B ... c ... D ... E ... F ... G ... H ... J ... K ... L ... M ... N ... 0 ... P ... 2.049 1-269 1.240 1.200 1.400 1.378 1.489 1,400 1 ,500 1.330 1.230 1.781 1-326 1.450 1.915 ...... ... ... ... ... ... * .. ... ... ... ... ... ... ... 1.838 1.156 1.200 1.1 60 1.279 1.279 1.198 1-318 1.058 1-219 1.063 1.599 1-245 1.354 1.758 ... ... ... ... ... ... ... ... ... ... ... ... ... ... Difference. 0.21 1 0.103 0.040 0 040 0.121 0.099 0.231 0.088 0.442 0.111 0.167 0.182 0,081 0.096 0.157 N.a8 nitrates and by Difference = aluminium E. in other forms. process, ... ... ... ... ... ... ... ... ... ... ... ... ... ... 0.019 0.017 0.022 0.025 0 022 0.018 0-019 0-021 0.022 0.016 0.021 0.017 0.018 0.020 0.018 ... ... ... ... ... ... ... ... 0 . . ... ... .. ... ... ... 0.1 92 0.086 0.018 0.015 0,099 0.081 0.272 0 067 0.420 0.095 0,146 0,165 0.065 0.076 0,139 * The N sample was from the continuous growth of wheat for 9 pare out of 10.362 THE ANALYST.The average of this series shows :- Nitrogen not coagulated . . . . . . . . . . . . . . . 0.149 per cent. Ditto a8 nitrates and nitrites . . . . . . . . . . . . 0.019 ,, Ditto in other forms . . . . . . . . . . . . . . . 0.130 ,, So that in the case of the flours only about one-eighth part of the nitrogen is accounted for by this determination.The variations here do not seem to show any special law, except that the proportion present is, in nearly every case, less than that found in the corresponding bran. The net result so far then is that in the bran we have an merage of 0.390 per cent, of nitrogen, and in the flours an average of *130 per cent of nitrogen, m-hich is not present as true gluten or as a nitrogen acid, Part of this quantity is no doubt present as glutin, and it seema probable that another portion may be in a body of the asparagin class.I am carryihg this investigation further, but cannot report upon it at present. My object in taking up this matter was to form a true opinion as t o flesh-forming values, and it is quite clear t o me that at present it will not do to assume that any nitrogenous bodies other than those coagulated by carbolic acid are really of value.Averaging the samples again I find that the whole meals show an average of 8.09 per cent. of true gluten, and the flours 8.33 per cent., or 0.24 per cent. of real flesh-formers in favour of the flour, and it must be borne in mind that this would have been larger if the bran had been more perfectly separated. I come therefore to the conclusion that the supposed greater feeding power of wholo meal is not at present borne out by the facts.I hope t o give shortly a few similar facts as to oats and barley, which, as far as I can judge at present, show similar results. At a nieeting of the Scarborough Town Council, on the 26th October, Mr. James Baynes, jun., was appointed public analyst for that borough. The other candidates were Mr. Thomas Fairley, T4eeds ; Mr. George Jarmain, Huddersfield; and Dr. Proctor, Pork. Mr, Louis Siebold, F.C.S., editor of the “Year-book of Pharmacy,” has been appointed Public Analyst for the County of W estmoreland. hlr. J. Baynes has been appointed public analyst for the Borough of Hanley. Mr. A. H. Allen has been appointed public analyst for Barnsley. Dr. Drinkwater, F.C.S., has been appointed Lecturer on Chemistry in the Edinburgh School of Medicine. FARADAY LECTURE.-Prof. Wurtz will deliver his lecture 11 Sur la Constitution de la Matiire A 1’Etat Gazeux,” on Tuesday, November 12th, at half-past eight o’clock p.m., in the Theatre of the Royal Institution, Fellows and visitor’s tickets can now be had on application to Mr. Hall. In consequence of the Chemical Society’s Dinner having been fixed for the 13th November, the next meeting of the Society of Public Analysts has been postponed to the 20th November.