The procedures commonly employed for the determination of lignin in plant materials used as animal feeding-stuffs are reviewed with particular reference t o the possible sources of error. A study of the reproducibility of each stage in the determination of lignin by a method employing 7 2 y 0 sulphuric acid is reported, and it is concluded that the variability of the final lignin yields is partly due to the presepce of alcohol-soluble substances not completely removed before the main treatment with 72.96 sulphuric acid. The need is indicated for further investigation of various ' humin ' precipitates obtained during lignin determination.

The method proposed by the authors (see l'art I1 of this series) for the determination of ' acid-insolublc lignin ' excludes thc humin material which dissolvcs in 7 2 0 0 ( w i w ) sulphriric acid but is reprecipitated by dilution. Furthcr work has shown this huinin material t o possess thc propertics of lignin. Although i t is t o bc expected that thc ' acidinsoluble ' lignin will hc resistant to the digestive processes and its dctcrmination should *

Examination of humin residues obtained a t various stages during lignin determination and of the alcohol-soluble fraction of pre-treated food materials has indicated the necessity for modifying the procedures used in lignin determination. Factors affecting the dispersion and hydrolysis of cellulose are reported, and it is shown that in order t o determine as lignin only material insoluble in concentrated acid it is necessary t o filter without dilution after the main treatment. A method for determining ' acid-insoluble lignin ' is proposed and shown t o give satishctory agreement between replicates.* Throughout this paper the term ' concentrated hydrochloric acid ' refers to the ordinary laboratory reagent (containing about 3696 HCl), whereas ' fuming hydrochloric acid ' is a more concentrated product (containing 41-4396 HC1) prepared by treating the ordinary reagent with hydrogen chloride.

No abstract

Ethyl acetoacetate was found t o be a satisfactory solvent for lignin, but not highly selective. The purest ' ester lignins ' were obtained from the residues after alkali extraction, and when mature plant materials were used these lignins showed little variation in methoxyl content. It is suggested that the percentage of methoxyl in pure lignin is not affected by plant maturity and that the true lignin present in a crude preparation may be determined from the methoxyl content by the use of an appropriate conversion factor. IntroductionThe determination of the lignin contents of plant materials by the attempted removal of all other substances and the weighing of the residues gives results that are far from satisfactory. Even when corrected for ash, there is still an unpredictable degree of contamination in the residue, due to both nitrogenous and non-nitrogenous substances, so that the true lignin content remains highly conjectural. From the results given in an earlier paper1 it appears that the total methoxyl in the crude lignin residue may be the best measure of the true lignin content, but although this may be useful in comparative work, such as digestibility measurements, it cannot be used to determine the absolute lignin content of any sample without a knowledge of the methoxyl content of the pure lignin. Moreover, it is generally believed that the methoxyl content of lignin is not constant, but varies with the age of the ~l a n t~-~ and with the species;5 the evidence on these points is far from conclusive, however. It was considered that, before further progress could be made with methods for the determination of lignin, some lignins of as Ggh a purity as possible should be prepared for reference purposes.Methoxyl and nitrogen contents were taken as the most reliable criteria of the purity of the various preparations. Crude lignin invariably has a high nitrogen content, but until there is satisfactory evidence to the contrary it is reasonable .to assume that pure lignin is nitrogen-free.Meyer & Bondi6 believe that the lignin molecule of plants contains nitrogen to the extent of 1-15%, but we have made about 20 different preparations of acid lignin from rye-grass hay and found them to have nitrogen contents varying from 0.3 to 5.6% according to the pretreatment, the nitrogen content of the material treated with 72% sulphuric acid and the degree of contamination with non-nitrogenous substances ; the nitrogen in lignin preparations is thus as variable as the mineral content and may equally be regarded as being due to contamination. Nitrogen content alone, however, is an inadequate index of purity, since contamination with carbohydrates may be quite considerable. If suitable pretreatments are employed to remove methoxyl-containing substances other than lignin, then the methoxyl content of the final preparation may serve as an index of freedom from carbohydrate contamination, and preparations with maximal methoxyl and minimal nitrogen contents are presumably the most pure.Brauns' extraction procedure for wood7 d...