ON THE ESTIMATION OF THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES. Lauri Paloheimo and Irja Paloheimo. Department of Animal Husbandry, University of Helsinki. Received 22. 11. 1948. Membrane substances and cell enclosure substances. The microscopic examination of vegetable materials suggests a partition of plant substances which proves very useful as well for the chemist as for the nutri- tionist. This is the partition into 1) cell enclosure substances or, plainly, enclosure substances and 2) cell wall substances or membrane substances. Contrary to the animal tissues the vegetable cell walls are easily distinguishable under the microscope. Also a good deal of the cell contents is detectable without using any dye or other reagent. With the exception of water, some inorganic salts, and pectin compounds which occur both in the cell walls and in the enclosure, the partition is sharp. Substances occurring in the intercellular spaces are regarded as enclosure substances. So far as the Phanerogames are concerned the two main categories of plant substcnces include the following subdivisions: Enclosure substances sugars dextrins starches inulins proteins amino acids amides amines lipids organic acids and their esters and salts alcohols hydrocarbons chlorophylls glucosides alkaloids inorganic salts water https://c-info.fi/en/info/?token=skYDUdGAI0M8mEaj.tk4R2G0wvF8TpQO-3bQNWg.1qaxNo7-f3HXo17qd0QOxn6QfhCKOJ3nWas_DhqrKCbjt4cLKLHK8StMRm3VYEBbksfk3Zr0roEB5vx45o8Oinhoj9YJnosA5I31pief8-FK3l_AVONtF0u8wNekS80Y9srocOSx8gEwIO5-zaYQ5B5to1-BopDVoGgVMiqmaIyju9DoRqJvQKJKmImIjxc6 LAURI PALOHEIMO AND IRJA PALOHEIMO Membrane substances . celluloses and other pectins membrane hexosanes lignins pentosanes suberins hexo-pentosanes cutins polyuronides This classifying list does not mean to be complete. It also contains some con- ventional faults; thus e.g., the water and the inorganic salts are ranged only under the Enclosure substances, whereas the pectins are placed under the heading Mem- brane substances. The different vitamins find their places in definite subdivisions of the enclosure substances. From the point of view of the nutritionist it is important to notice that, with eventual exception of certain pectin compounds, which possibly are decomposed by the hydrochloric acid in the stomach, the membrane substances can’t be disin- tegrated by the digestive fluids secreted to the alimentary canal from the glands. On the other hand, the substances belonging to the cell enclosure are either absorb- able as such or can be disintegrated by the digestive fluids into absorbable compo- nents. There are only few and insignificant exceptions from this rule. That a part of the enclosure substances escapes the digestion does not usually depend on the inherent character of the compounds in question but on the protective effect of the cell mambranes. The usefulness of the above partition is not impaired by the fact that the horse and ruminants can digest the membrane carbohydrates and pectins with the aid of bacteria; for the disintegration of the plant cell walls in the alimentary canal of the named animals is far from being complete, and a high content of membrane substances means altogether a low feeding value of the fodder in question. Con- cerning the foods of the man and omnivorous animals the pertinence of the partition of vegetable materials into enclosure substances and membrane substances is obvious: the total of the membrane substances approximately represents the indigestible part of the food. In the teaching of fodder chemistry the above partition has proved very in- structive. The one of the authors has also adopted it to his textbooks on animal nutrition (21, 23). Earlier attempts to determine the total of membrane sub- stances. It is more than likely that just the microscopic examination of plant tissues led the naturalists of the 18th century to the conception that the cell walls consisted of a single substance, the wood substance, la matiere ligneuse (2, p. 561). In spite of Payen’s investigations which led to the discovery of »cellulose» and »les matiéres incrustantes» from the wood (ref. Dumas, 3, p. 52, and Czapek, 2, p. 522, 561), the erroneous conception of the unity of the wood substance lasted in many labo- ratories over the middle of the 19th century. THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES H. Davy was probably the first to propose a method for estimation of the total of membrane substances. He treated the woody materials to be investigated with boiling water and boiling alcohol, but it is worthy of mention that he did not term the remainder as wood substance but crude fibre (9, p. 303). It is self-evident that Davy’s method was not fitted for investigation of mate- rials containing appreciable amounts of starch or protein. Carl Sprengel was presumably the first who described a method for analyzing such materials (26, p. 251). He treated the sample to be analyzed successively with water, alcohol, ether, diluted hydrochloric acid, diluted potassium hydroxide and chlorine water. The residue obtained was called Holzfaser. Obviously Sprengel had the erroneous concept that this product, as obtained from a definite raw material, was a chemical unity and contained the whole cell wall matter of the raw material in question, although, according to his view, there were different kinds of Holzfaser depending on the plant products analyzed, as there are different starches and sugars. Spren- gel did not know that by the treatment with acid, alkali and chlorine, not only starch, protein and certain coloured substances but also a great part of membrane substances were removed. The principle proposed by Sprengel was adopted to use in many European laboratories during the next three decades, until the end of 1850’s. Only the concentration of the acid and alkali varied in different modifications of the method, and treatment with chlorine was usually omitted. Only the methods of Boussin- gault, Wolff (ref. Henneberg and Stohmann, 6, p. 511), Moser, Dietrich (ref. Lebbin, 14, p. 214) and Henneberg and Stohmann (6, p. 367) may be men- tioned on this paper. Several methods from the same period which were intended for the estimation of cellulose must here be left unconsidered. It is however, import- ant to remember that there was in Germany in those times much confusion about the term Holzfaser. Many writers used it as a synonym with the words Zellstoff and Cellulose, where as some others meant by it the total of membrane substances. It is likely that Henneberg and Stohmann to begin with shared the latter view. Later, as their own investigations proved the erroneusness of this view, they re- placed the term Holzfaser by the term Rohfaser, crude fibre. It is a very common opinion that the usual, so-called Weende method for crude fibre determination were an invention of Henneberg and Stohmann. As a matter of fact the method, nearly in its final form, was used at the Weende expe- rimental station already before Henneberg and Stohmann. The reason why the method just in the form Henneberg and Stohmann gave to it has made its way through the world, is the fact that particularly in this form it was used at Weende in exeptional numerous analyses, many of which were joined to carefully performed digestion experiments. Instead of being the inventors of the Weende crude fibre determination method the named workers have by a scientifically accurate way investigated the behaviour of the vegetable cell membranes in the different treatments belonging to this method. Before 1856, however, Poggiale (ref. Kraut, 12, p. 19) had already shown that by the acid- and alkali treatments of the usual wood fibre determination methods LAURI PALOHEIMO AND IRJA PALOHEIMO4 a part of the cellulose goes into solution. As the wood fibre (Holzfaser) of the fodder plants was considered to be identical with the wood substance of the proper wood which, in return, was regarded as a prototype of indigestible plant material, it is only natural that even wood fibre of the fodder materials was considered as indi- gestible. Henneberg and Stohmann (6, pp. 513, 514) were presumably the first who established that the crude fibre (wood fibre) of the hay or straw was not indi- gestible. They stated that the digestibility of the crude fibre of different fodders varied from 45.5 to 60.4 %. After estimating the elementary composition of the crude fibre of different fodders, of the nitrogen-free extract 1 and of the digestible and indigestible portions as well of the crude fibre as of the nitrogen-free extract of the same fodders, the said investigators were able to draw some important conclu- sions. They concluded: 1) that in the crude fibre determination the cell wall sub- stances are rather arbitrarily divided into two portions, the one constituting the crude fibre and the other belonging, together with sugar and starch, to the nitrogen- free extract; 2) that the crude fibre is not pure cellulose but contains also encrusts rich in carbon (lignin, suberin and cutin are named), which remain among the feces together with the indigestible part of cellulose, whereas the digestible part of crude fibre is pure cellulose; 3) that in the crude fibre determination the bulk of the lignin and a part of cellulose are dissolved and thus fall into the nitrogen- free extract; 4) that indigestible part of nitrogen-free extract is composed mainly of lignin (6, pp. 367, 511 —514; 7, pp. 331—335; see also Paloheimo, 16, pp. 9 -13). The criticism directed to the Weende method by Henneberg and Stohmann is so accurate that the later investigators have scarcely been able to add to it any- thing essential. The Weende method had proved to be no method for estimation of the total of cell wall substances but a very rough and imperfect cellulose deter- mination method. This view was not altered after introduction of the pentosane determinations according to Tollens (28, p. 3575—3585), for it was shown that the main part of the pentosanes falls into the nitrogen-free extract (König, 11, p. 93). Thus it appeared that not even the lignin-free portion of the nitrogen-free extract was composed merely of cell enclosure substances. Of the later criticism of the Weende method the papers of Hoffmeister (8, p. 243), Düring (4, p. 87), König (1. c.), Honcamp and Ries (9, pp. 306—317), Magnus (15, p. 34), Palo- heimo (19, p. 281) and Norman (ref. Ellis, Matrone and Maynard, 5, p. 285) 2 may be mentioned. Paloheimo (1. c.) points to the fact that cellulose, though resistant as well against diluted acid as against diluted alkali, is partially rendered soluble in alkali when preliminarily treated with acid. Poggiale (1. c.) not only criticized the usual crude fibre determination methods but also developed a new method for estimation of the total amount of membrane substances. In his method, which was developed for the analysis of vegetable 9 The complex of protein-, fat- and ash-free material dissolved by the treatments belongin to the crude fibre determination method. ) The original paper: Normann A. G. 1935. The composition of Crude Fiber. Agr. Res., 25 529—540 THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES 5 food-materials, the matter to be analyzed is successively treated with cold water, ether and diastase. From the residue the share of crude protein and ash is to be removed. The remainder was called fibre. Besides his diastase method Poggiale proposed a modification in which the treatment with diastase is replaced by boiling in 2 per cent, (about 0.3 normal) hydrochloric acid. In our opinion the latter treatment is far too vigorous and causes dissolving of a great lot of membrane constituents. In our own method a boiling in 0.05 normal acid is used. The ferment method of Poggiale, although obviously the most accurate of the fibre determination methods of the last century, did not win any success in the time of its inventor and later it was entirely forgotten. The authors who 30, 70 and 80 years later developed analogous methods seem not know the work of Poggiale. The first who after Poggiale proposed a ferment method were Stutzer and Isbert (27, p. 93). They used not only an amylase-containing solution but also pepsin- and pancreatin solutions. It is obvious that the method is very time-con- suming, and it is questionable, nevertheless, whether the protein can be quantita- tively disintegrated and extracted by the ferment solutions. Remy (25, p. 2), Andersen (1, p. 52), Williams and Olmstead (ref. Horwitt, Cowgill and Mendel, 10, p. 264) h Horwitt, Cowgill and Mendel (1. c.) and Williams, Wicks, Bierman and Olmstead (29, p. 595) have later proposed similar methods with only non-essential modifications. The principle and general features of the new method. The principle and the general features of our method for determining the total of vegetable cell membrane substances were described in a preliminary report by L. Paloheimo (22, p. 19). A short review may, however, be appropriate. Our method was developed without knowledge of the methods of Poggiale (1. c.) or the later modifications of his diastase method, which we came to know only afterwards. It is evident, however, that the new method belongs to the named line of analytical research. The immediate starting point of it was the starch deter- mination method of L. Paloheimo (18, p. 150; 20, p. 391; 24, p. 109) and parti- cularly, the observation that a boiling with 0.05 normal acid was sufficient to disintegrate all the starch and to liberate it from the plant cells. Using samples previously extracted with suitable lipid solvents it was obwiously possible after boiling with weak acid to obtain undissolved residues which contained, besides the membrane substances, only proteins and ash constituents. It seemed probable that the treatment would be very cautious as far as the membrane substances are con- cerned. As the determination of the protein and ash content of the residue is an easy task the approximate membrane content could be estimated by subtracting *) The original paper: R. D. Williams and W. H. Olmstead 1935. A biochemical method for determining indigestible residue (crude fiber) in feces: lignin, cellulose and non water soluble hemi- cellulose. J. Biol (Them., vol. 108, pp. 653—666. LAURI PALOHEIMO AND IRJA PALOHEIMO from the residue after boiling with 0.05 normal acid its protein and ash. Thus the general features of the method turned out to be as follows. A sample extracted with an efficient lipid-solvent is boiled with a large volume of 0.05 normal hydrochloric acid solution. The residue separated from the solution by filtering is dried, weighed and incinerated. The nitrogen determinations are made from the lipid-free material and from the filtrate, the nitrogen content of the residue is calculated as difference. The corresponding protein content is cal- culated using the coefficient G.25 and the share of the protein is subtracted from the loss of incineration. The difference obtained is the amount of membrane sub- stances. The nitrogen-free organic substances dissolved by the acid treatment are called by the authors easily soluble carbohydrates. The percentage of this fraction actually composed as well of easily soluble as of easily hydrolyzable nitrogen-free non-lipid substances can be calculated by subtracting from 100 the percentages of water, ash, lipids, crude protein and membrane substances. It is obvious that for securing a correct value for the group of easily soluble carbohydrates the lipid- solvent used must not dissolve any non-lipid substances. In many cases it is advis- able to fit an after-extraction with a lipid-solvent in between the filtration and incineration. It is important that the volume of the acid solution is large enough to prevent any significant change of the H-ion concentration. A detailed description of the method. In the following the method is described chiefly in that form the authors have employed it. Some suggestions however are made to other possible manners of procedure. Preliminary treatment and preservation of the samples. Materials containig water more than about 15 per cent are dried at 60° C with efficient motor-driven ventilation. The air-dry material obtained is ground by a Wiley mill with 1 mm screen. In some cases the grinding is easier if the material is quite dry. Then the air-dry material is dried in a vacuum oven at 60° C before grinding. After grinding the material is dried in a vacuum at 60° C for about 10 hours and placed in a pre- servation pot with tight stopper; if it need not be stored for a longer time it can be placed air dry in the pot. Determination of water. If not only the percentage composition of the dry matter, but also that of the original watercontaining material is to be determined, the above manipulations must be completed by some weighings fitted in approp- priate points. When in the later course of the analysis dry-matter determinations are to be made it is preferable to perform the drying in a vacuum oven at 70° C for 10 hours. This is recommendable especially for the sake of fructose. Extraction of lipids. The most efficient lipid-solvent the authors have tested is the ethanol-benzene (32 parts by weight ethanol and 68 parts by weight benzene). It can however be used in our method only when the material is preliminarily THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES extracted with cold water. The authors have, in fact planned and applied a more complete scheme of analysis to which belongs the extraction with cold water, and by means of which the group of easily soluble carbohydrates can be divided into two subsections: 1) water soluble carbohydrates and 2) easily hydrolysable carbo- hydrates. 1 In that system the extraction with ethanol-benzene is appropriate. Usually the authors have made the first extraction with ether and the after- xtraction (see above p. 6) with ethanol or with ethanol-benzene. As well the ether as the sample to be extracted (8—10 g) must be quite dry. Either Soxhlet- or Twisselmann-apparatus has been used; The after-extraction has always been made with Soxhlet-apparatus, the middle part with the siphon being isolated with paper. By this arrangement it is possible to retain the temperature of the solvent around the sample quite near the boiling point. In the after-extraction the residue after the acid-treatment need not be dry. If one prefers the drying before the extrac- tion, it may be done even in a common drying oven at 100—105° C. The drying of the extracts is performed in a vacuum at 70° C. In some instances, especially when the after-extract contains water, it is advisable to put in the flask some slips of filter paper which have been dried and weighed together with the flask before the extraction. When only the determination of membrane substances is to be made, without paying attention to the other groups of substances, materials poor in lipids may be analyzed omitting the preliminary extraction with ether. Boiling with acid, filtration. The material extracted with ether is spread on a watch glass or a Petri disk and let stay over-night for attaining an air-dry state. Then three samples are weighed: 1) for determination of dry matter, 2) for nitrogen determination, and 3) for boiling with acid. For the last purpose I—21 —2 g are weighed; of materials rich in mambrane subs- tances 1 g is adequate. The sample is placed in a 600 ml beaker and mixed with some cold distilled water. If the material contains hard particles with compact formations of solid gels of starch, pectin or protein, the weighed sample must be crushed in a mortar before transferring into the beaker. 2 The beaker has a mark at 400 ml. Boiling water is added to the beaker not quite up to the mark and the mixture is brought up to boil. 20 ml of 1-normal hydrochloric acid is added and made up to 400 ml with boiling water. The mixture is boiled for 30 mi- nutes by compensating the loss of evaporation with boiling water. When examining succulent materials poor in lipids, such as potatoes, beets or different fruits, analysis can be made of the fresh material directly. A sample of s—lo5—10 g of grated pulp is weighed for the acid treatment. Generally it is advi- sable to crush the weighed sample in a mortar before boiling with acid. Immediately after boiling, the mixture is filtered through a coarse sinter of glass or quartz covered with a layer of quartz sand. A water-jet vacuum pump is used. The filtration must go rapidly; if a pellicle rendering the filtration difficult is formed on the surface of the sand layer, it must be teared up with a rough-edged glass rod. The residue on the filter is washed with about 70 ml of boiling water; l ) This system of analysis will probably be published in the near future Many prepared foods, e.g. bread, belong to this category LAURI PALOHEIMO AND IRJA PALOHEIMO8 the filtrate with the washing waters (30 ml for the washing of the filter-flask) is transferred in to a 500 ml measuring flask and reserved for nitrogen determination, eventually also for starch determination according to the iodine colorimetric prin- ciple (24). After-extraction, incineration. After-extraction, the ethanol or ethanol- benzene extraction of the residue from the boiling with acid, is indispensable also, and especially, in those cases where the preliminary extraction with ether is omitted. The residue with quartz sand is transferred into a thimble made of filter paper. Also may the whole sinter crucible be put into the Soxhlet apparatus if only the dimensions of the apparatus permit it. Further description of the extraction pro- cedure is given already on page 7. After extraction the residue with the sand is transferred into an ignited cru- cible. The crucible with contents is dried, weighd and placed in an incineration oven. After incineration at red heat the loss of incineration is estimated. If a sinter crucible of quartz has been used and the after-extraction made without removing the residue from the crucible, also the incineration can be made in the same crucible. Protein correction, calculation of results. The loss of incineration is compozed of membrane substances and of the protein not dissolved by the acid treatment. The latter component can be estimated by subtracting from the amount of crude protein before the boiling the amount dissolved by boiling with acid. As mentio- ned above, the filtrate was reserved for nitrogen determination in a 500 ml mea- suring flask. After making the flask up to the volume an aliquot, at least 200 ml, is taken for the nitrogen determination. The calculation of the percentage of membrane substances needs no further description. For attaining a more comprehensive picture of the nature of a food or feed the percentagesof water, ash, lipids (crude fat), crude protein and easily soluble carbohydrates must be known. For computing the percentage of lipids the amount resulting from the after-extraction is to be added to the yield from the preliminary extraction. The amount of the easily soluble carbohydrates is obtained by sub- tracting from 100 the percentages of the other constituents. If a part of filtrate after boiling with acid is used for starch determination the fraction of easily soluble carbohydrates can be divided into two subdivisions: starch and other easily soluble carbohydrates (see p. 7). Criticism and applications. The authors have not as yet performed any detailed studies on the behaviour of different plant substances in the proposed method, and especially in the treat- ment with weak acid. However, some plain observations and simple experiments are rather encouraging as to the usefulness of the method. By testing with iodine the residues resulting from the acid treatment of different vegetable products it was proved that only insignificant traces of starch were still present among the membrane substances. It may therefore be concluded that the THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES 9 Table 1. Content of membrane substances, lipid-, protein-, and ash-free residue after boiling with water, and crude fibre of different vegetable materials poor in starch. (The figures are percentages on dry matter basis.) Lipid- protein-, Membrane and ash-free Crude substances residue after fibre boiling whit water Filter paper 99.6 99.9 89.8 Spruce wood 90.9 96.3 73.8 Birch wood 90.8 94.3 59.5 Hulls of sunflower seeds 87.5 90.5 64.6 Cork 85.1 86.1 22.3 Peat litter 85.1 88.3 42.9 Peanut hulls 83.8 85.1 71.9 Rye straw 82.1 89.1 51.1 Oat straw 73.8 83.3 47.0 Corn cobs 72.0 90.1 36.1 Rye chaff 66.8 77.5 34.3 Barley chaff 58.6 72.4 27.9 Red clover stalks .... 55.3 64.1 40.6 Oat chaff 51.3 62.7 25.7 Alfalfa hay 42.5 50.1 32.0 complex of membrane substances obtained by the method in question is free from cell enclosure carbohydrates. It is obvious that for the competence of the method the inviolability of the cell walls is not less important than the complete dissolving of the cell enclosure carbohydrates. Considering the great number of cell wall constituents and their very variable susceptibility to acid hydrolysis it is to be expected that even so weak an acid as 0.05 normal may have a dissolving effect upon some cell wall subs- tances. In the authors’ opinion it is even impossible to find an acid treatment by which it were feasible to dissolve all the enclosure carbohydrates without affecting the integrity of the complex of membrane substances. It is, however, essential that a method planned for the determination of the complex of membrane substances, as unbroken as possible, does not give yields that include also a part of cell enclosure carbohydrates. Thus, although boiling with water may be considered a treatment by which the membrane substances are dissolved hardly in any degree, this treat- ment is inadequate to liberate all the starch from the tissue. However, so far as materials poor in starch are concerned, it is obvious that the lipid-, protein-, and ash-free portion of the residue after boiling with water must not be much greater than the corresponding result of the membrane subs- tance determination. In tables 1 and 2 the comparison in question is presented comprising a series of vegetable materials poor in starch. The tables show also the corresponding percentages of crude fibre. Tables 1 and 2 suggest that the plant cell walls do not remain intact in the treatments of our method. It is also obvdous that the margin between the pereen- Table 2. Content of membrane substances, lipid-, protein-, and ash-free residue after boiling with water, and crude fibre of some grasses at different stages of growth. (The figures are percentages on dry matter basis.) Lipid-, protein, Membrane and ash-free resi- Crude substances due after boiling fibre with water Phleum pratense leaf stage I 50.3 53.1 22.0 leaf stage fl 54.2 58.0 26.9 at beginning of bloom 62.1 65.6 32.8 Alopecurus pratensis leaf stage 47.7 53.1 23.3 with ears, before blooming 55.9 59.0 26.7 at full bloom 58.4 64.9 28.9 Poa pratensis leaf stage 51.9 55.4 21.3 with panicles, before blooming .... 56.6 61.5 28.0 at full bloom 60.0 67.1 29.2 Festuca pratensis leaf stage 42.0 50.6 22.4 with panicles, before blooming .... 42.8 54.4 23.1 at beginning of bloom 53.7 60.7 26.8 A vena elatior leaf stage 47.1 49.0 22.3 with panicles before blooming .... 52.8 58.4 28.6 at beginning of bloom 65.1 68.8 32.6 Trifolium repens leaf stage 26.8 31.6 15 6 at beginning of bloom 27.7 35.8 21.5 after full bloom 43.1 49.7 32.9 Table 3. Behaviour of pentosanes in boiling with 0.05 normal acid (The figures are percentages on dry matter basis.) Pentosanes dissolved Material Pentosanes ; total m V 2 hour’s in 1 hour’s boiling With acid White clover grass 11.7 4 5 5 1 Wheat bran I 24.0 17.1 18.3 Wheat bran II 25.4 18.0 19 1 tages of the water-insoluble fraction and the »membrane substances» in many cases is so broad that it cannot be covered with the whole pectin content of the material in question, not to mention the fact that a part of pectins is dissolved already by the hot water. It is conspicuous that the said margin is especially broad in some very coarse feed materials as in different kinds of straw and chaff whereas 10 LAURI PALOHEIMO AND IRJA PALOHEIMO THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES 11 it is rather narrow in different sorts of hay and even in samples of hay cured at a very early stage of growth. On the other hand, the margin is narrow also in some coarse materials not suitable for feeds. At the present state of our researches it is difficult to decide what are the substances which, in addition to a part of pectins, are soluble in boiling 0.05 normal acid although insoluble in boiling water. Pentosane determinations according to Pollens cannot, owing to the non-specificity of the method, afford much light on this question. Table 3 shows that at least in some cases a rather great portion of furfurol-forming substances is soluble in the acid treatment of our method. Con- cerning the wheat bran II we know that the percentage of »pentosanes» dissolved in boiling water is 7.3. Tables 1 and 2 not only disclose the incompleteness of our method but, on the other hand, reveal its superiority as compared uith the crude fibre determination. The difference between the percentage of membrane substances determined by our method and the percentage of crude fibre is, as a rule, very great, in many cases about a half of the amount of membrane substances. As we may conclude that our method gives residues free from lipids and cell enclosure carbohydrates, and as the protein correction cannot involve any appreciable errors, the membrane substance values in Tables 1 and 2 may not prove to be too high. On the contrary, they may be regarded as minimum values, the actual values being as a rule some- what higher. Thus they, compared with the corresponding crude fibre figures, reveal the very violent manner in which the treatments of the crude fibre determination attack the vegetable cell wall. Especially striking is the low crude fibre value of cork, a result of the disinte- grating effect of the boiling with alkali. A considerable part of the margin between membrane substances and crude fibre may, as a rule, be covered with the alkali- soluble fraction of lignin. The fact that the said margin in spruce wood is relatively narrow depends on the resistance of coniferous lignin against diluted alkali (see Paloheimo, 17). It is obvious that our method is to a certain extent conventional. As the 0.05 normal acid is capable of dissolving a portion of the cell wall substances during a half an hour’s boiling, it is to be expected that a doubling of the acid concentration or of the duration of the boiling will result in a lower yield compared with the result obtained by the conventional boiling procedure. Table 4 shows, however, that lengthening the duration of the boiling with half an hour does not appreciably reduce the yield; neither has the doubling of the acid concentration any great effect on the result of analysis, as seen from table 5. It is especially remarkable that a prolonged boiling appears to result in a rather slight increase in the amount of dissolved pentosanes of clover grass and wheat bran (see table 3). As the cell wall substances of the wheat bran as a whole evidently are exceptionally sensitive to the boiling with acid, it remains an interesting problem to solve what are those sensitive components. Obviously the size of the sample to be boiled in 400 ml of acid solution is not without importance. Therefore we have arranged comparative experiment’s with LAURI PALOHEIMO AND IRJA PALOHEIMO12 Table 4. Effect of the duration of boiling with 0.05 normal acid (The figures are percentages on dry matter basis.) The protein-, ash-, and fat-free residue after Material % hour’s 1 hour’s boiling with 0.05 normal acid Oat straw 73.8 69.9 Alopecurus pratensis 62.9 59.5 Palm kernel cake 51.7 51.7 Alfalfa 48.8 48.5 Trifolium repens 43.1 40.3 Coconut cake 32.2 31.5 Red clover leaves 30.7 29.0 Wheat bran 23.0 19.0 Linseed cake 22.9 21.8 Wheat bran I 21.5 16.5 Cottonseed cake 17.4 16.5 Sunflower-seed cake 19.9 15.5 Carrots 128 122 Rutabagas 12.7 12.0 Red beets 10.7 9.2 Fodder beets 9.9 9.1 Potatoes 3.9 3.6 Table 5. Effect of the raising of acid concentration from 0.05 normal to 0.10 normal. (The figures are percentages on dry matter basis). The protein-, ash-, and fat-free residue after % hour’s boiling in Material 0.05 normal 0.10 normal acid solution Oat straw 73.8 68.7 Wheat bran 111 1 31.5 25.7 Red clover leaves 30.7 29.3 air-dried samples of 1,2, and 3 g. The results which are seen in table 6 show that the method is not very sensitive to such variations. From the results of the last-mentioned experiments also the conclusion can be drawn that the pH of the acid solution may not be essentially influenced by the small sample of vegetable material. This circumstance the authors have directly studied only with a sample of young red clover grass. While the blind test gave the pH value 1.27, the solution with 1 g of air-dry grass sample gave the value 1.33. A sample boiled with 400 ml destilled water only gave the value 6.62. 1 A part of the starch removed by washing with cold water THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES 13 Table 6. Effect of the size of the sample to be boiled with TOO ml 0.05 normal acid. (The figures are percentages on dry matter basis.) Air-dried sample of Material Ig2 g 3 g Hay, mixed 49.8 50.5 Wheat bran 22.1 22.1 Barley corns, with hulls 18.7 18.6 Spinach 18.0 18.4 Cabbage, head without stem 14.0 14.6 Apples, fruit flesh 6.3 6.5 Potatoes, peeled 2.8 3.1 Wheat flour, Graham 4.8 4.7 Wheat flour, white 0.5 - 0.8 Table 7. Solubility of crude protein by the boiling with 0.05 normal acid. Material Crude protein, dissolved % Trifolium repens, leaf stage 33.7 Alfalfa hay 49.3 Rye straw 40.4 Cabbage 75.9 Carrots 78.8 Potatoes 92.3 Apples 93.8 Wheat flour, Graham 82,5 Soya meal 70.1 Cottonsead meal 89.7 Aspergillus niger 20.9 Yeast, baker’s 100.0 Pseudomonas fluorescens 98 2 Bacillus subtilis 100.0 The importance of quick filtration has already in a previous chapter been pointed out. The fulfilment of this requirement seems not cause any trouble, for all materials analyzed by the authors have given by boiling with 0.05 normal acid readily filtrable mixtures. Nor has the boiling with acid met with any noteworthy difficulties. The protein correction, which is an essential point in our method, contributes in many instances to the inaccuracy of the results. The authors have conventio- nally, and certainly erraneously, assumed that the nitrogen content of the crude protein of all materials analyzed, as well in the fraction dissolved by the 0.05 nor- mal acid as in the fraction of the residue, is the same, viz. 16 %. In fact these circumstances are rather unsufficiently known and we have no idea especially of the possible differences in the nitrogen content of the nitrogenous compounds of AND IRJA PALOHEIMOAURI PALOHEIMO the two fractions named above. The fraction of crude protein soluble in the 0.05 normal acid during a half an hour’s boiling is greatly variable depending on the botanical nature of the sample and also on its degree of fineness. Table 7 shows the percent solubility of the crude protein of some vegetable materials of different type. (It is worth to remember that the filter used by the authors was rather coarse.) Usually the error rising from the protein correction is relatively the smaller the greater the ratio membrane substances/undissolved protein. In cases where the matter to be analyzed contains chitin substances it is perhaps advisable to omit the protein correction. It is obvious that all errors attached to the different determinations belonging to our system of analysis ha\’e their effect upon the percentage of »easily soluble carbohydrates». To these errors belongs also a conventional error connected with the determination of lipids: the sum of the preliminary extract obtained with ether and of the after-extract obtained with ethanolbenzene is always somewhat smaller than the total of lipids because a part of lipids falls into the filtrate after boiling with acid. It is also evident that omitting the nitrogen of the lipids causes an error in the percentage of the total crude protein. This circumstance as well as the con- ventional errors attached to the determination of ash are universallv known. Conclusions. In the preceding chapter the authors have criticized rather strictly their own method. Its defects should, however, not be exaggerated. Compared with the Weende crude fibre estimation method and with its modifications the new method has the great advantage that the plant substances are fractionated by it, if not exactly, at least rather accurately, into histologically natural groups. It helps to give a correct picture of vegetable materials and offers a rather useful basis for a deeper going and more detailed analysis Summary. The authors refer to the chapter »The principle and general features of the new method» and to the »Conclusions». REFERENCES (1) Andersen, A. 1934. Die Futtermittelanalyse und die Bestimmung der Verdaulichkeit der Futtc mittel. Skandinavisches Archiv für Physiologie, 69. (2) Czapek, Fr. 1905. Biochemie der Pflanzen, I. Jena. (3) Dumas, J. B. A. 1839. Rapport ?ur un memoir de M. Payen, relatif åla composition de la matiére ligneuse. Comptes rendus hebdomaires des seances de I’Academie des Sciences, VIII (4) Düring, Fr. 1897. Über den Pentosangehalt verschiedener Futtermittel und deren Rohfaser Journal für Landwirtschaft, 45. Ellis, G., Matrone, G. and Maynard, L. 1946. A 72 percent H2S0 4 method for the determina tion of lignin and its use in animal nutrition studies. Journal of Animal Science, 5. (6) Henneberg, W. und Stohmann, J. 1859. Über das Erhaltungsfutter volljährigen Rindviehes. Journal für Landwirtschaft, 7. 7) Henneberg, W., Stohmann, J. und Rautenberg, J. 1864. Berichte über die auf der landwirt- schaftlichen Versuchsstation zu Weende ausgeführten Versuche. Journal für Land- wirtschaft, 12. (8) Hoffmeister, W. 1888. Die Rohfaser und einige Formen der Cellulose. Landwirtschaftliche Jahrbücher, 17. (9) Honcamp, F. und Ries, F. 1914. Untersuchungen über die verschiedenen Stroharten mit beson- derer Berücksichtigung der Zusammensetzung und Verdaulichkeit unter dem Einflüsse der Witterung. Die Landwirtschaftlichen Versuchsstationen, 84. (10) Horwitt, M., Cowgill, G., and Mendel, L. 1936. The availability of the carbohydrates and fats of the green leaf together with some observations on crude fiber. The Journal of Nutrition, 12. (11) König, J. 1897. Die Notwendigkeit der Umgestaltung der jetzigen Futter- und Nahrungsmittel- Analyse. Die Landwirtschaftlichen Versuchsstationen, 48. (12) Kraut, E. 1858. Jahresbericht über die Untersuchungen und Erfahrungen auf dem Gebiete der Landwirtschaft. Journal für Landwirtschaft, 6 11. (13) KÜhn, J. 1897. Die Zweckmässigste Ernährung des Rindviehes. Dresden. (14) Lebbin, G. 1897. Über eine neue Methode zur quantitativen Bestimmung der Rohfaser. Archiv für Hygiene, 28. (15) Magnus, H. 1919. Theorie und Praxis der Strohaufschliessung. Berlin. (16) Paloheimo, L. 1926. Lignin Determination by Acid Hydrolysis. Acta Agralia Fennica, IS. (17) —l>—■ 1929. Beiträge zur Ligninbestimmung mit Säurehydrolyse. Biochemische Zeitschrift, 214. (lg) —)>— 1930. Zur Verwendbarkeit der jodkolorimetrischen Prinzips bei Stärkebestim- mung. Biochemische Zeitschrift, 222. (19) — i»— 1930. Mitä ovat raakakuitu ja typettömät uuteaineet? Maatalous, 23. (20) —»— und Paloheimo, I. 1931. Beiträge zur Jodkolorimetrie der Stärke nach der methode von Paloheimo. Biochemische Zeitschrift, 238. (21) -—>>— 1932. Kotieläinoppi. Jyväskylä. (22) —»— 1945. Determination of total quantity of cell wall constituents in foods and feeds. Maa- taloustieteellinen Aikakauskirja, 17. (23) —»— 1947. Kotieläinhoidon perusteita. Jyväskylä. (24) — k— 1948. Determination of starch according to the principle of iodine colorimetry. Maa- taloustieteellinen Aikakauskirja, 20. (25) Remy, E. 1931. Experimentelle Studien zur Biochemie und Biologie der Rohfaser. Biochemische Zeitschrift, 236. (26) Sprengel, C. 1832. Chemie für Landwirthe, Forstmänner und Cameralisten, 11. Göttingen. (27) Stutzer, A. und Isbert, A. 1888. Untersuchungen über das Verhalten der in Nahrungs- und Futtermitteln enthaltenen Kohlehydrate zu den Verdauungsfermenten. Zeitschrift für physiologische Chemie, 12. (28) Tollens, 8., Günther, A. und de Chalmot, G. 1891. Über die Bestimmung des Furfurols und der in Vegetabilien enthaltenen Penta-Glycosen und Pentosane. Berichte der deutschen chemischen Gesellschaft, 24. (29) Williams, R. D., Wicks, L., Bierman, H. R. and Olmstead, W. H. 1940. Carbohydrate values of fruits and vegetables. The Journal of Nutrition, 19. 15THE TOTAL OF VEGETABLE MEMBRANE SUBSTANCES 16 LAURI PALOHEIMO JA IRJA PALOHEIMO SELOSTUS KETTOAINEIDEN KOKONAISPALJOUDEN MÄÄRITTÄMINEN KASVIKUNNAN TUOTTEISTA. Lauri Paloheimo ja Irja Paloheimo Kotieläintieteellinen laitos, Helsingin yliopisto Käytännössä oleva weendeläinen raakakuidunmääritysmenetelmä on kovin sovinnainen ja luon nontieteellisesti epäpätevä. Karkein puutteellisuus tässä menetelmässä lienee, että pääosa ligniniai neista, jotka muodostavat ruoka- ja rehuaineiden ravinnollisesti negatiivisimman osan, liukenee kuitu määrityksen käsittelyissä ja joutuu siten typettömien uuteaineiden joukkoon yhdessä sokerin ja tärk kelyksen kanssa. Tekijät ovat suunnitelleet menetelmän, jolla kasvikunnan tuotteisiin s'sältyvier kettoaineiden kokonaispaljous saadaan melko tarkoin määxitetyksi. Tähän menetelmään kuuluu 1) rasva-aineiden poistaminen tutkittavasta näytteestä, 2) näytteen keittäminen runsaassa määrässä 0,01 normaalista suolahappoliuosta, 3) happokeitossa liukenematta jääneen osan eroittaminen siivilöimällä 4) liukenemattomaan osaan sisältyvän raakavalkuaisen määrittäminen rasvattoman näytteen raaka valkuaismäärän ja happokeitossa liuenneen raakavalkuaisen eroituksena ja 5) liukenemattoman jään- nöksen hehkutuskevennyksen määrittäminen. Kun hehkutuskevennyksestä vähennetään siihen sisäl tyvä, edellä esitetyllä tavalla laskettu raakavalkuaisen määrä, saadaan selville kettoaineiden paljous Saadut kettoaineprosentit ovat useimmiten 150—200 % korkeammat kuin vastaavat raakakuitupro sentit.