Historic, archived document Do not assume content reflects current scientific knowledge, policies, or practices. Washington, D. C. Octcber, 1923 FLAX-STEM ANATOMY IN RELATION TO RETTING. By Rosert L. Davis, Assistant Plant Breeder, Office of Fiber Investigations, Bureau of Plant Industry. CONTENTS. Page TINTS 0 2) 6c i ee = = Ee See aes Wicks Bnet il The retting objective and some factors that may feiertene: Be deh are 2 ee vig ngey ies 2 The progress of retting in the different tissues of the stem_ Hx nen 5 Some macroscopic and externa] changes in the flax stem during Rettig: Sas 1 Mechanical methods of testing wet flax stems for completion of retting__________ 16 SUN Ws oe ea Stas eas ors Steep. at): Weameee Reh gre ey dl 26 INTRODUCTION. The flax stem consists of two main parts, the central wooden core, hned with pith, and the cortex or bark, which completely encircles the wooden core. Between the wooden core and the bark hes the cambium layer, commonly known as the green inner bark. The fiber bundles are located in the cortex. The process of retting has the double object of separating the cortex from the wooden core and the fiber bundles from the nonfibrous portions of the cortex. Numerous efforts to ret flax by chemical methods have thus far failed to be sufficiently successful to supersede the natural methods of dew retting and water retting. In dew retting the stems are spread out on a meadow for four to six weeks, and irregular weather conditions may result in a very uneven product. In water retting the flax stems are submerged in water, and retting is accom- plished by the activity of bacteria, which dissolve the gums that hold the fibers to the surrounding tissues. Water retting may be controlled more satisfactorily than dew retting, 1s a more rapid process, and usually results in a more uniform product than dew retting. The scope of this bulletin is confined to water retting. One of the chief difficulties experienced in water retting is that of deciding when to discontinue the process. If the stems are re- moved before the critical point is reached the fiber will not be well separated from the cortex and will be both coarse and harsh. If the process is carried beyond the critical point of complete retting the 55159—23——_1 2 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. fibers will be seriously weakened through the dissolution of the @ pectin or gum that holds together the fiber cells forming the strands — ! or commercial fibers. 3 Many practical retters assert that the only reliable test for the completion of retting is to remove several handfuls from different parts of the retting tank and examine the quality of the fiber after it is dried. Many retters depend upon the Belgian or loose-core test made on wet stems freshly removed from the retting tank; in — this case retting is said to be complete when the wooden core may be pulled out from the fibrous cylinder without any resistance to | the pull. The objection to the first or dry-straw method is that fiber may be injured by artificial drying unless proper humidity is maintained in the drying room. It is impracticable to have to wait for the sample to dry naturally in the sun on account of the danger of overretting with the remainder of the flax stems in the retting tank, and it is rather expensive to equip a drying room so as to maintain proper humidity. The second method, that of the loose- core test, is very undependable in inexperienced hands and _ not entirely dependable in the hands of experienced retters. A positive loose-core test indicating that retting 1s complete may at times be obtained when retting is far from complete and when the fibers are still firmly attached to the rest of the cortex. The study reported | | in this bulletin was undertaken to determine whether there existed a relation between the flax-stem anatomy and retting that would be helpful in testing wet flax stems for the completion of retting more accurately than can be done through the use of the loose-core test. THE RETTING OBJECTIVE AND SOME FACTORS THAT MAY INTERFERE. The object of retting is to obtain perfect fiber and to terminate the process so that the best balance of softness and strength of fiber results. Flax fiber, in general, is graded according to strength, fine- ness, weight, length, color, and freedom from shives, or pieces of the wooden core. The best quality of flax fiber is strong, very fine in diameter, soft, feels heavy to the hand, averages 20 inches or more in length, and is entirely free from shives. Light-cream and silvery- gray colors are common with good flax fiber, but the color is not so important as the other qualities. A silken luster is considered de- sirable, as this indicates that the fiber is well retted. If the flax stems are underretted the fiber is apt to be green in color, due to adherent chlorophyll from the cambium layer, and the fibers may not be separated from the rest of the cortex. Overretted flax fiber is enerally dull gray, lacking in luster, and so very weak that it may Be pulled to pieces by the hands with little effort. . After the retting process is completed the flax stems are dried and the mechanical separation of the wooden core from the retted fiber is performed by the breaking and scutching processes. In | breaking, the flax stems are run through a series of corrugated rollers. This breaks up the wooden core into short pieces known as shives. Well-retted flax fiber is soft, strong, and pliable, so that it bends between the corrugations of the rollers without breaking pA Boe ered Liladistese nding ; A FLAX-STEM ANATOMY IN RELATION TO RETTING. 3 and comes through the machine uninjured. In scutching, the fiber, which has been run through the brake, is evened up and exposed to the blows of a paddle wheel which beat off any shives that remain clinging to the fiber. If the fiber is much overretted, it will be weak and go to pieces as it passes through the brake. If it is under- retted, the shives will cling persistently to the fiber, and much long fiber will be wasted in the scutching process before the shives ean be beaten off. The factors that interfere with the retting objective may be divided into those of variable straw quality and those due to variable retting conditions. The freedom from shives, and the luster, the color, and the softness of the fiber are almost entirely determined by the management or mismanagement of the retting process, but nearly all the other characters on which flax fiber is graded may be influenced by the straw qualty, which is, in turn, influenced by the manner in which the crop is cared for, as well as by the kind of seed used, the quality of the soil, and the season or climate. STEM FACTORS AFFECTING RETTING. Strength of fiber may be influenced by the variety of seed used or by the soil. Different varieties of fiber flax differ considerably in the strength of fiber. Stems of White Blossom Dutch have less fiber strength than stems of Blue Blossom Dutch of equal diameter, and the fiber in the stems of Saginaw, a variety selected for superior stem length and strength of fiber, has greater strength than that in either of the preceding varieties. Flax stems produce weaker fiber when grown in*marshy or sandy soils? than when grown in clay- loam soil. The strength of fiber may be influenced also by season or climate; a dry climate or season produces fiber that is relatively weak. The fineness of fiber is influenced both by heredity, as represented by differences in the varieties of seed used, and by environment, as represented by the growing conditions. The fiber of White Blos- som Dutch is coarser than that of Blue Blossom Dutch, due to hereditary differences.’ According to hand-lens examinations of fiber from the 1922 crop at East Lansing, Mich., the fiber of White Blos- som Dutch is coarser than that of Blue Blossom Dutch, even wher stems of equal diameter are compared. The thickness of the stand of flax as controlled by the rate of seeding or by injuries from frost or disease determines the size and the uniformity of the stem diam- eter. A thick stand that has not been injured by frost or disease causes crowding of the flax plants and produces stems of fine diam- eter. Stems of fine diameter in turn produce fine fiber. Figure 1 shows at the left unretted fibers taken from a stem of Blue Blossom Dutch 1.3 millimeters in diameter. Compare the fine quality of this fiber with the relatively coarse fiber shown at the right from a stem 1 Davis, Robert L. Pedigreed fiber flax. U.S. Dept. Agr. Bul. 1092, p. 18. 1922. lpoyier Gustaf. Inhemska fibervixter. Jn K. Landtbr. Akad. Handl. och Tidskr., arg. 60, p. 388-424, 13 fig. 1921. (Manuscript translation by Ebba Oleson Campbell, p10. "On file in the Office of Fiber Investigations, Bureau of Plant Industry.) *Tammes, Tine. Die blau-bliimigen und weiss-bliimigen Flachsarten und ihre Bedeu- tung. In Forsch.-Inst. Sorau Verbandes Deut. Leinen-Industrieler Mitt., Jahrg. 2, Nr. 6-7. 1920? (Manuscript translation by Lucile Reinbach, p. 9. On file in the Office of Fiber Investigations, Bureau of Plant Industry.) 4 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. of the same variety 1.7 millimeters in diameter. The photograph here reproduced represents the average size, enlarged six times, of fiber bundles that may be found in stems of these respective diam- eters and shows how close the association is between fine stems and fine fiber. Measurements were made of the fiber bundles from stems of a selection in which the variations due to heredity have been eliminated and the variation in fiber size may be regarded as due almost en- tirely to the differences in stem diameter. The fiber bundles of this selection, Yale (10), a strain selected at Yale, Mich., in 1920, measured from 0.06 to 0.07 millimeter in stems 1.5 millimeters in diameter. The fiber bundles in stems 1.1 millimeters in diameter taken from the same selection were less than 0.04 millimeter in diam- eter. The term diameter in reference to fiber bundles means the width measured tangent to the circumference of the stem. The average number of fiber bundles is not materially less in a stem of small diameter, and the fiber bundles are necessarily smaller in the fine stems, as they are crowded together into a smaller girth of cortex. Since the opposite is also true, a thin stand of flax results in stems of large diameter that produce coarse fiber. For best results the stems should be uniform and should not be more than 1.4 millimeters or much less than 1.2 milli- meters in diameter. It is only when large and small stems that fall outside of this range are included that there is much variation in the num- ber of fiber bundles Fic. 1.—Variation in the size of fiber bundles of flax due SAF to differences in the diameter of the stem. The fibers clue to the stem di on the left were taken from a Sed te eee es ameter variation. diameter, and the relatively coarse fibers on the rig # were taken from a stem 1.7 millimeters in diameter. — Part of the lack of uniformity in fiber fineness in Blue Blossom Dutch is probably due to the great variety of strains of flax that are present in this variety. Strains that vary in stem length and in fineness of fiber have been selected from this variety in Ireland? and in the United States.° The possibilities that are present along this line of selection for increased fiber fineness are indicated by a study of the fiber bundles in two closely related pure lines. The influence of heredity has given 36 fine fiber bundles to one pure line selected from Blue Blossom Dutch, while to stems of the same diameter of another pure line it has given only 24 relatively coarse fibers which are fully half as broad again as those of the former. It is not easy to untangle all the interrelating factors of environ- ment and heredity, but it has been shown that the quality of the Hunter, H. Improvement of the flax crop by propagation from selected plants. Jn Jour. Dept. Agr. and Tech. Instr, Ireland, v. 15, p. 245-246. 1915. Stewart, John W. The Irish flax-growing industry and how it may be improved, p. 126-129. Coleraine, Ireland. [1922?.] : 5 Davis, Robert L. Op. cit., p. 19. - FLAX-STEM ANATOMY IN RELATION TO RETTING. 0 fiber may be influenced indirectly by frost resistance. The coarse fiber produced by White Blossom Dutch is in part due to its marked susceptibility to frost, as its stand may be depleted by frost and coarse stems developed when varieties such as Blue Blossom Dutch and Saginaw escape with little injury.® ; It has been shown that straw quality may influence the retting product in strength and fineness of fiber and that the hereditary quality of the seed as well as the soil, climate, and season all share in exerting this influence. Therefore the objective of retting, that of producing fiber of the best quality, can not be attained unless pains be taken to assure uniform straw quality, It is important that the stems be fine and of uniform diameter. If possible a strain of flax selected for strength and fineness of fiber should be used, so as to eliminate the variation that may occur even in stems of the same diameter. These precautions will eliminate most of the vari- ables that affect retting and are due to straw quality, so that the variations that arise may be controlled through management of the retting process. FACTORS OF THE RETTING PROCESS AFFECTING THE QUALITY OF THE FIBER. ‘The variables in the retting process itself are the nature of the retting bacteria, the temperature of the water, the rate of water circulation, the water quality (whether hard or soft), and the dura- tion of the ret. Of these factors the one that causes most trouble is the duration of the ret, or making the decision as to when retting is completed. A study of the process of disintegration of the flax stem as retting proceeded was undertaken in order to find out what changes in the flax stems were closely associated with the comple- tion of retting. THE PROGRESS OF RETTING IN THE DIFFERENT TISSUES OF THE STEM. The rapidity and order with which the different tissues in the flax stem are retted depends on the digestibility or solubility of the cementing substances that bind the tissues together, their accessi- bility, and the relative amounts of them present in the different tis- sues. The tissues of the flax stem that lie in the cortex may be divided into the fiber bundles, the phloem parenchyma lying be- tween the fiber bundles and the cambium layer, the parenchyma lying between the fiber bundles themselves, the epidermis that in- closes the stem, and the outer parenchyma that les between the epidermis and the fiber bundles. (Fig. 2.) Retting takes place first in the cambium layer (tissue 1, in fig. 2) where the pectin is quite soluble and where because of the very thin cell walls the layers of pectin or cementing substance are corre- spondingly thin. When flax stems are sterilized in water at 115° C. much of the pectin in the cambium layer is dissolved, and the cortex is so well loosened from the wooden core that the loose-core test is of no value at all as a positive indicator that retting is completed. The pectin in the tissues of the cortex is less soluble than that in the ® Davis, Robert L. Frost resistance in flax. U. S. Dept. Agr., Dept. Cire. 264, 8 p., 5 fig. 1923. FEROS TT SOF WE 6 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. cambium layer and is not much affected by the solvent action of the hot water. It is not entirely because of a difference in accessibility that the cortex is retted after the cambium layer, as at the exposed surfaces of cut stems where all tissues are equally exposed to the attack of the bacteria, retting 1s not materially hastened in the cortex. It is true, however, that the cambium layer, located on the inside of the stem, is somewhat more accessible to the attack of the bac- teria than the tissues lying to the outside in the cortex. The stems = Pieris * CELL HALLS OF 11/, CUTIN INTERCELLULAR. XYLEM oR FIBER CELLS VA IALAYER GU/t OF F/BER BUNOLE WOOD CELLS lig. 2.—Cross section of flax stems, showing the consecutive order of the retting processes. The numbers indicate the order in which the tissue is destroyed by the bacteria. 1, The cambium layer; the cortex becomes loosened from the central wooden core. 2, The phloem parenchyma is disintegrated, exposing the insides of the bundles > to the bacteria. 38, The parenchyma between the bundles is destroyed, separating the edges of the bundles and exposing the outer parenchyma to contact with the retting liquid. 4, Most of the outer parenchyma goes next, leaving the fiber bundles held together by the thick-walled epidermis. 5, The epidermis and the remainder of the outer parenchyma and the fiber bundles are now isolated. Note that the thickness of the intercellular gum between the fiber cells is greater than that of the cell walls in the adjacent parenchyma. (Camera-lucida drawing, enlarged 480 times.) are partially protected from the outside by the waterproof nature of the cutin in the outer wall of the epidermis. The areas where the leaves drop off are waterproofed by the formation of leaf scars where suberin is deposited. At harvest time the waterproof cover- ing on the outside of the stem is complete with the exception of the stomatal openings. Part of the stomata, as microscopic examination shows, are closed and made water-tight by the formation of cork cambium. The retting liquid and bacteria, however, can find their way through the epidermis, as the flax stems which have been paraf- fined at both ends will ret. The closing up of the ends, however, does FLAX-STEM ANATOMY IN RELATION TO RETTING. 7 distinctly retard retting, indicating that the bacteria enter less readily through the epidermis than at the stem ends. ‘These con- clusions were drawn from the following experiment: DETERMINING THE MODE OF ENTRANCE OF BACTERIA INTO THE RETTING FLAX STEM. This experiment was made with the advice and assistance of R. A. Tweed and Miss Antoinette Trevithick in the department of bacteriology, Michigan Agricultural College, East Lansing, Mich. Two sets of test tubes containing both sterilized and unsterilized straw were set up in series of six. Both the tops and the bottoms of the straw in three of each set were paraflfined, while the other three were unparaffined. One set of six was inocculated and incubated under anaerobic conditions and the other under aerobic conditions for nine days at 30° C. After this period the straw was examined, and it was found that the fiber was weak and overretted in the case of the unparaffined in both aerobic and anaerobic sets. The paraffined sets showed the fiber to be strong and well retted under both aerobic and anaerobic conditions. These resultS were similar with both sterilized and unsterilized straw. This experiment was run on four different occasions, the only difference be- ing that unsterilized straw, as well as sterilized straw, and an aerobic culture were used in the first experiment while sterilized straw only and a facultative anaerobic culture were used in the other three experiments. The results just described would indicate that the paraffined ends obstructed the entrance of the bacteria and retarded retting. As the culture used in three of the experiments consisted of a facultative organism, the results would further indicate that the retardation in retting was due to obstruction of the entrance to bacteria rather than the obstruction of the entrance to air. Inasmuch as the unparaffined sets retted more-rapidly than the paraffined sets, it would indicate that the bacteria found more ready means of access through the ends of the stem than through the epidermis. As the flax stems are but loosely filled with pith and are nearly hol- low it is reasonable to assume that they contain a large quantity of air relative to the quantity present in the retting water in the test tubes. This would tend to disprove that the restriction in the air cir- culation was responsible for the retardation in retting. The bacteria enter the flax stems through the broken ends of the roots; the inside of the flax stem is but loosely filled with pith cells at the center and is nearly hollow, so that once inside the root ends, the retting liquid containing the bacteria can find ready access through the entire length of the stem. The bacteria can then pass out into the cambium layer along the medullary rays between the sectors of wood or xylem. The cambium layer is thus most accessible to the attack of the bacteria, because it lies nearest the pith of the stem where bacteria find their way unobstructed and the almost com- plete waterproof covering on the outside of the stem forces the bac- teria to first attack the ‘stem from the inside and adjacent to the ecambium layer. The cambium layer forms an unbroken cone running the full length of the stem excepting at the nodes, where the leaf scars interrupt its continuity. The area that is next disintegrated by the bacteria is that occu- pied by the phloem parenchyma, - which lies between the fiber bundles and cambium layer (tissue 2, in fig. 2). The inside surfaces of fiber bundles, when well retted, are “nearly always more lustrous than the outside surfaces, indicating that the inside surfaces, lying 8 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. next to the phloem parenchyma, have been subjected to the action of the bacteria for a longer time and that the traces of the phloem parenchyma have been more completely destroyed than those of the outer parenchyma that lies next to the outside of the fiber bundles. The phloem parenchyma is more accessible than the outer paren- chyma, since the former lies next to the cambium layer and the bacteria soon assimilate the small quantity of food material in the cambium layer. It does not take so long to destroy the phloem parenchyma as the outer parenchyma, because of the thinner layers of pectin present in the phloem parenchyma. 7 x 27 ue Fic. 3.—Fiber bundles of flax taken from the same stem, showing variations in the size, the shape, and the number of cells. The figures show the number of cells counted in each bundle. The outside edges of the bundles are uppermost. The crosses at the points of grooves indicate where separations would occur during retting. (Outline camera-lucida drawing, enlarged 140 times.) After the phloem parenchyma is disintegrated the bacteria and the enzym pectase which they secrete are found next to the inside surfaces of the fiber bundles and the parenchyma that hes between them. The fiber bundles and the strips of parenchyma between them may be likened to a picket fence in which the .“ pickets” or fiber bundles are about five times as wide as the spaces. The comparison holds good from the standpoint of bacterial activity; the fiber bundles form obstructions to the progress of the bacteria both be- cause the quantity of pectin present in the fiber bundles is many times as great as that in the areas between them and because the cementing substances between the fiber cells are partly woody and more or less impervious to the attacks of bacteria.’ HExamina- 7 Tadokoro, T. Studies on flax retting. Jn Jour. Col. Agr. Tohoku Imp. Univy., v. 5, p. 84. 1913. FLAX-STEM ANATOMY IN RELATION TO RETTING. 9 tions with the microscope of cross sections made at the end of this stage in retting show that the fiber bundles, outer parenchyma, and epidermis are intact and that the parenchyma between the fiber bun- dles has been destroyed. The relation of the thickness of the pectin layer to the retting pe- riod is indicated by the shorter retting period of small stems. Stems 1.5 to 2 millimeters in diameter take 12 to 36 hours longer to ret than stems varying from 1 to 1.5 millimeters in diameter. The larger stems have thicker layers of pectin. The panicle end, which is the small end of the stem, rets before the root end of the stem, where the pectin layer is relatively thick. As soon as the parenchyma between the fiber bundles is destroved the retting liquid and bacteria surround the bundles on all sides excepting the outer sur- faces that le next to the outer parenchyma. According to Tadokoro the cell walls of the epidermis, excepting the cuticle, and the cell walls of the paren- chyma and the middle lamelle of the fiber cells are composed al- most entirely of pectin compounds. Hence the relative thickness*® of the walls of these tis- sues as compared with each other and with the middle lamelle of the fiber cells give some idea as to the time it should take to break down these tissues in Fic. 4.—Changes in the appearance of the cortex of the retting process. A va opens the last BIE of eee. The cortex : 3 : ‘ is here shown separate rom the wooden core, but study of Figure 2 still almost intact. 10. shows: that the pectin layer is thickest in the fiber bundles, followed in order by the epider- mis and outer parenchyma. The phloem parenchyma and that lying between the fiber bundles have pectin layers of about equal thickness but somewhat thinner than that of the outer parenchyma. Of all the tissues the cambium has the thinnest layer of pectin. Since the layer of pectin in the cell walls of the epidermis is thicker than that in the outer parenchyma it would naturally be expected that the epidermis would not be destroyed until after the outer parenchyma. ‘The path of bacterial activity is directed against the outer parenchyma, and microscopic examination shows that those portions of the outer parenchyma lying next to the spaces between the bundles are destroyed next (tissue 4, in fig. 2), leaving only the strips of outer parenchyma that are found in the middle of the outer surfaces of the bundles intact. The reason for this is that the bac- ® Tadokoro, T. Op. cit., p. 37. 55D159—23 2 10 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. teria have to work four or five times as far to meet at the middle of the outer surface of a bundle as to reach the epidermis. The retting fluid now incloses almost nine-tenths of the circum- ferences of the fiber bundles, and the fiber bundles are held together only by the epidermis and the small strips of outer parenchyma lying at the middle of the outer surfaces of the bundles and bridg- ing the space between the fiber bundles and epidermis. The retting of these two nonfibrous elements of the cortex that remain appears to occur almost at the same time as the separation of the fiber bundles into several smaller units that are known as commercial fibers, and when this last step 1s accomplished retting is completed. It is well known by flax retters that the disintegration of the epidermis occurs as about the last change in the retting process, and Tadokoro states that the cuticle is detached mechanically during the last stage of retting.? Whether these changes occur si- multaneously depends somewhat upon the stem diameter; in very coarse stems the fiber bundles are not sepa- rated into smaller units until some time after the loss of the epider- mis, owing probably to the extreme thickness of the pectin layers in the bundles. The fiber bundles are Fic. 5.—Changes in the EDD eget OF ee ee Rs composed of several flax during the last stages of retting. The cortex ; is here shown separated into a number of smaller thousand fibe1 cells, girine, Natuly aver, mera oh mereuervan. ae © gind acres ecctionam | one bundle will aver- age about 25 cells in number. The fiber bundles vary in size and shape and in the number of cells. Figure 3 shows camera-lucida drawings of the outlines of fiber bundles taken from the cross section made at the center of a single flax stem. Some of the bundles are three or four times as large as others. Some have 60 cells, some have only 25, and some in extreme cases only 8 or 10 cells. Flax fiber bundles are not, as is commonly supposed, cylindrical or elliptical in cross section. On — the contrary, they are irregular, being two or three to eight times as- broad, measured parallel to tangents drawn on the circumference of the stem, as the thickness of the bundles measured radially. They are somewhat rounding to slightly zigzag in outline on the outer »' Tadokoro, I. Op: cit.; p. 37. ~ cent and show the bun- FLAX-STEM ANATOMY IN RELATION TO RETTING. Et surfaces and concave and grooved on the surfaces that face toward the center of the stem. From the standpoint of retting the degree of concavity is of interest, as the deep grooves furnish natural places for the separation of the fiber bundles during the retting process as well as during the subsequent processes of breaking and scutching. Frequently there is anastomosis of several fiber. bun- dles coming up from the base of the stem, so as to form one large bundle, and it is quite common to find deep indentations marking the spots where union took place. These grooves make it easier for the large bundles to be retted and separated into smaller units. The irregular shapes and sizes of the fiber bundles help to explain the difficulty of obtaining a uniform ret and also explain why linen threads and cloth do not have the same uni- formity as cotton of the same quality, since all cotton fibers are single cells of uniform diameter. Figures 4 to 8, in- clusive, show a series of external longitudi- nal views of the cortex and fiber during the last stages in retting. In Figure 4 retting has separated the cortex from the wooden core, but the cortex is still almost intact and un- divided. The leaf scars have become translu- Fic. 6.—Changes in the appearance of the cortex of dle traces. In Figure flax dupZig = last freee of petting: Only a few : pieces of epidermis and parenchyma here remain 5 the cortex has sepa- clinging to the fiber pane which are not yet sepa- 4 1 2 - rated into the small widths necessary for industrial rated into a number of eee ee CAO. smaller strips, the bac- | teria have destroyed most of the parenchyma and epidermis that held the fiber bundles together, but the epidermis and parenchyma are still clinging to the bundles in fairly large pieces. In Figure 6 only very small pieces of epidermis and parenchyma are found clinging to the fiber bundles, but the leaf scars persist, and the fiber bundles are not yet separated into the small widths of fiber necessary for industrial use in spinning. In Figure 7 retting is completed, the leaf scars have disappeared, the fiber bundles are entirely tree from impurities, and are split lengthwise into fibers of suitable widths. The fibers in Figure 7 may be compared with those in Figure 8, in which a repre- sentative sample of uncombed water-retted Dutch fiber is shown. The microscopical study carried along with the retting process confirms the work of other students of retting in that the outside of 12 BULLETIN 1185, U. S. DEPARUMENT OF AGRICULTURE. the stem is the last to be retted. Very close to the time when retting is complete the fiber bundles separate into several smaller units, and the cuticle is detached from the outside of the stem. This study also points out that retting proceeds from the cambium layer toward the outside of the stem and that the waterproof union of the fiber bundles and wood branches formed by the leaf scars at the nodes, make retting at the nodes more difficult than in the internodes. These conclusions indicated that the search for a completion-of-retting test should be focused on the outside surface of the stem rather than on the area between the wooden core and the cortex occupied by the green inner bark or cam- bium layer. SOME MACROSCOPIC AND EXTERNAL CHANGES IN THE FLAX STEM DURING RETTING. Some retters claim that sliminess of the stem 1s closely associated with the end point of retting, and others claim that stem brittleness is thus associated? In order to eliminate any chance of missing a change that is at all as- sociated with the end point of retting, a list ve . . . Oo lic. 7.—Changes in the appearance of the cortex of Nias made of the changes flax during ‘the last stages of retting. The fiber mentioned in the litera- bundles here shown are entirely free from impurities SAR a: , pace and are split lengthwise into fibers of suitable width, ture and those that have indicating the completion of the retting process. heen noted in this studv. x 10. : The approximate — se- quence of these changes was determined by studying the retting proc- ess repeatedly. Observations were made every six hours and records were taken of the time when each change was first noted. Such changes as sinking of the straw, brittleness of stem, and maximum sliminess of stem are of little value, as they take place too far in advance of the completion of retting. Brittleness of the stem merely indicates that retting is proceeding satisfactorily and that stagnation has not oc- curred. The loosening of the cortex from the wooden core, as determined by the Belgian or loose-core test, also occurs at times several days before the completion of retting ina 14-day retting period. Toward the end of retting the stems became gradually lighter in color (fig. 9), changing from a yellow-brown to a very hght yellow or almost 10 Huntley, F. A. Fiber flax investigations. Wash. Agr. Exp. Sta. Bul. 33, p. 18. 898. FLAX-STEM ANATOMY IN RELATION TO RETTING. 13 _ white color. This change in color is not dependable, as flax stems _ that are well cured may be practically white in color when first _ placed in the retting tank. The change in color is closely associated with the last stages in retting, although the eye may not be depended - on to decide between the different degrees of lightness in color. This _ change may be noted as a color change or as a disintegration process _in the exterior portion of the cortex. Figure 9 shows a gradual in- _ crease in lightness of color as the bacteria disintegrate the epidermis and the outer parenchyma. The stem on the left is distinctly cloudy and murky in color, owing to the loosened cell contents in the tissue _ between the cuticle and the fiber bundles, although the normal color _ of the unretted stem is light yellow-brown. In the center is shown a stem that is murky only in spots; the stem is light colored, and the cuticle is loosened excepting at these spots. On the right a stem is shown that is ~ much lighter in color - and whose retting is very nearly completed ; the cloudy areas are very small, and the tissue between cuticle and fiber bundles ap- pears to have been de- stroyed excepting in those areas. The maxi- mum lightness in stem color is reached when the cuticle sloughs off. If the fibers are suffi- Fic. 8.—Water-retted Dutch fiber of good quality. Note ciently washed at this the fineness of the strands and compare with those in point the naked fibers 78°" %* 1° may be seen, glistening and transparent against the white back- _ ground of the wooden core. The leaf scars resisted the retting process of disintegration for a long time, and their disappearance was one of the last changes in the external appearance of the stem. Fiber bundles and the wooden core branch out into each leaf in the growing plant. When the leaves drop off at maturity the scars left by them are covered by a waterproof substance known as suberin that cements together the _ tip ends of the branches of the wooden core and fiber bundles that formerly fed each leaf. These areas are known as leaf scars and show up as dark-brown heart-shaped spots about 0.5 to 1 millimeter in diameter at intervals about every half inch or nearly every centi- meter along the stem. This means that in a stem 75 centimeters (30 inches) long there will be close to 100 leaf scars forming water- _ proof unions between the fiber bundles and the wooden core. The 55159—23—_3 ee 4 \ q : | 14 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. | i tT | leaf scars also overlap the adjoining epidermis at_each node, so that the Hh waterproof union includes epidermis, fiber bundles, and w ooden core The leaf scars cling persistently until toward the very last stage in retting. Figure 10 shows on the left a piece of unretted cortex spread out flat; note the distribu- tion of leaf scars, represented by small black, triangular to heart- shaped spots. In the center the partly retted portion of a flax stem shows how, after breaking, the shives cling to the cortex at the nodes where the leaf scars occur. On the right is shown a piece of cortex that is almost through ret- ting; note that the broad or coarse places in the fiber center about the leaf scars. Examinations were made of samples of good to medium grade fiber from Belgium, Holland, Japan, Chile, Ireland, and Canada. In the best grades of fiber which had been well retted no leaf scars could be found. In the underretted and coarse grades of fiber leaf scars were found where there were coarse places in the fiber. Figures 11, 12, and 13 give some idea as to the nature and the per- sistence of the leaf scars. Figure 11 shows a piece of unretted cortex enlarged ten times. The fiber bun- dles show only as very dark strips. The structure of the epidermis is obscured by the tissues underlying it. The bundle traces are not visible in the leaf scar. An undeveloped ~ bud is shown in the axil above the leaf scar. In Figure 12 retting has proceeded far enough to separate the wooden core from the cortex, but the cortex is almost intact. The cel- lular structure of the epidermis shows, as part of the underlying tissue has been destroyed. The | re. 9._Chanbes an dhe color of fax 2eal scar has been partly: dissecnem I stems during the process of retting. A away from the rest of the cortex. i Seas colnn de, (asia again as the Note how persistently the branches | last stages of retting. The light-colored of the fiber bundles cling to the leaf HII stem on the right is almost retted. : =) scar. Figure 138 shows a _ leaf scar and the surrounding tissue as it appears when retting is almost complete. The retting has removed nearly all the pa- FLAX-STEM ANATOMY IN RELATION TO RETTING. 15 renchyma, and very little other than the fiber bundles and the epidermis remains. The tissue near the leaf scar has been de- stroyed and the leaf scar partly disintegrated, exposing the bun- dle traces. Almost retted. _ (Fiber fromm portion |ifh 17 of asingle stem), ~\i i Partly retted. : | iil Piece of unretted cortex. E BD / fi 4 =2cm. Fie. 10—Leaf scars in retting flax. On the left a section of unretted flax cortex is _ Spread out flat to show the distribution of the leaf scars. In the center is shown a piece of a partly retted flax stem and on the right a piece of almost retted cortex. Note that the broad places in the fiber center about the leaf scar. The study so far indicates that no dependence can be placed upon the time of sinking of the straw or the external appearance of the stem as to color or sliminess and that all but three changes that 16 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. may be of use in testing wet stems may be eliminated. These changes are (1) the fineness of the fiber, which includes the complete- 4 ! d He (Wie : one ‘k Aapal ee a TAA Path Wy 48 mall; j J i —~. Slog = a a Y, M Hd 1 ' LNB D raat = = ee nO es EE So, So, role a x Re eS ee >> —————————— = A CO Se ee ene ap — Fj 7 r - = ——a — a Sp Bes wae S re enna, tr = 20cm. Fic. 11.—A piece of unretted flax cortex flattened out to show the exterior view. The structure of the epidermis (e) is obscured by the under- lying tissue. The bundle traces are not visible in the leaf scar (l). The fiber bundles (f) show as dark strips. u==Undeveloped bud. ness with which the cor- tex is separated into the fiber bundles and also the final separation of the fiber bundles, particularly the larger ones into sev- eral smaller units; (2) the disintegration of the ~ epidermis, which includes the digestion of the pectin in the inner cell walls and the mechanical sepa- ration of the cuticle; and (3) the disappearance of the leaf scars as an indi- cation that the water- proof unions of fiber and wood branches at the nodes are broken down. MECHANICAL METHODS OF TESTING WET FLAX STEMS FOR COMPLE- TION OF RETTING. The problem that arose at this point in the study was how to devise a test or combination of tests for completion of retting that would include two or if possible all three of the changes most closely associated with comple- tion of rettine, namely, the disappearance of the leaf scars, the fineness of fiber separation, and the disintegration of the epi- dermis. The various steps in the Belgian or loose- core test were analyzed in order to determine why it 1s undependable and whether any part of it might be of use. In mak- ing the loose-core test the stem is broken about one- fourth of its length above the. base end and the short piece dis- carded. The cortex or cone of retted fiber is then pushed back several centimeters from the broken end (fig. 14), so as to expose FLAX-STEM ANATOMY IN RELATION TO RETTING. 17 the wooden core entirely free from any fibers. The wooden core is then broken 6 to 9 inches above the first break (fig. 15), the worker being careful not to leave any jagged pieces of wood that would catch on the fibers when the test is made. In the third step one hand grasps the stem above the second break and the other pulls on the exposed lower end of the wooden core (fig. 16). In a positive loose-core test (fig. 17) the wooden core slips out from the cortex with- out any clinging on the part of the fibers. The cortex or fibrous cone pulls away from the wooden core as a single unit, and there- fore the loose -core test does not give any idea as to the fine- ness of division of the fibers. For the same reason it does not give any indica- tion as to the degree of disintegration of the epidermis, for even though the cuti- cle were already-sepa- rated and ready to drop away from the fibers at the shght- est dislodement at the time the loose - core test is made, it re- mains clinging to the fibers in an undis- turbed condition as the core is withdrawn. It is therefore neces- sary to devise a test in which the cortex is not handled as a single unit and in Fic. 12.—Retting flax, showing the persistence of the leaf scars. The process of retting has here reached one of the last stages. The cortex has separated from the wooden core, but is still almost intact. The structure of the epidermis shown as part of the un- derlying tissue has been destroyed. Explanation: which the stages of retting in the differ- ent parts of the cortex b = Fiber branches, e = epidermis, f = fiber bundles, l= ieaf scar, p = point of attachment. Note how the fiber branches cling to the leaf scar at points of at- tachment. may be observed. A difficulty is experienced in testing wet flax stems for completion of retting regardless of the method used; when the stems are with- drawn, one at a time, from the retting tanks which are tightly packed with flax straw, all or part of the cortex may be rubbed off by contact with the adjacent stems. The stem, being in the nature of a wedge, slips out more readily if it is withdrawn from the butt end of the bundle. This precaution is not sufficient to prevent injury to the cortex from contact with the other stems. The diffi- 18 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. culty may be entirely overcome, however, through the use of test cylinders placed at intervals in each tank. The test cylinders are packed loosely, so that the stems may be withdrawn one at a time without rubbing against each other. Test cylinders may be made of three thin boards nailed together so as to form a three-sided prism, with a half-inch crack along one edge to permit the free circulation of water and with a small triangular piece of wood nailed across one end for the bottom. Withdrawal of the stems from the test cylinder is facilitated by cutting off the pani- cles of the stems and placing them in the cylinder with the root ends up. A test eylin- der may also be made of a section of bamboo about 24 inches in di- ameter by knocking out the nodal parti- tions in a piece long enough to hold the stems and boring many holes about 1 centi- meter in diameter in the side for water cir- culation. In the loose-core test the wooden core is moved in the direction of the root end of the stem, and the cortex is held stationary. There are a number of other = 2 i ways in which the cor- ® ery ae eee tex and the wooden Fic. 13.—Retting flax, showing the persistence of the core may be separated, leaf scars. Retting has disintegrated nearly all the , P ome tissues in the cortex except the fiber bundles and the and these are divided cp dermis that overlaps them and holds them together. into three oroups. In he leaf scar is partly disintegrated, exposing the OTP. ‘bundle traces. Explanation: ) = Bundle traces, ¢ = group 1 are included se r is, 3st I 1 tates v3 sear, = = = f . eed Bs ber bundles, 7 = leaf scar, wu = unde two w ays of making the loose-core test, in which the total resistance to separation of cortex and core may be felt at once. In group 2 are included various ways of testing the dissolution of the leaf scars, and these tests are therefore called leaf-scar tests; in the leaf-scar tests the resistance to separation of cortex and core may be observed at each node, one at a time. In group 3 are included the epidermis tests, in which it is intended to show the fineness of division of the fibers and the extent of the disin- tegration of the epidermis as indicated by the quantity of cuticle that has separated from the stem. In testing the efficiency of these various methods a double objective may be kept in mind: (1) What test or combination of tests is the most dependable for deciding when to terminate the retting process, and (2) what test is easiest P ‘ : z 3 FLAX-STEM ANATOMY IN RELATION TO RETTING. 19 made by ie ned hands? In other words: (1) What test which may be confirmed by examination of the fiber in the dry state _ gives positive indication that retting is complete? (2) what test from the viewpoint of the inexperienced worker gives the strongest - indication that retting is complete? The various ways that have been tried in this study for mechani- cally separating the cortex from the wooden core are listed as fol- - lows: Group 1.—Loose-core tests: The cortex is held stationary. (1) The loose core test made toward the root end of the stem. (2) The loose-core test made toward the panicle end of the stem. MG; a So \ - po ® ic. 14.—Step 1 in the loose-core test of retting flax fibers. The cortex is shoved back from the wooden core so as to leave several ‘centimeters entirely free from fibers. Group 2.—Leaf-scar tests: The wooden core is held stationary. (3) The entire girth of cortex is peeled away from the wooden core, beginning at the basal end of the stem and keeping the cortex parallel to the cere. (4) The same as test 3 except that the peeling is started at the panicle end of the stem. (5) Peeling the cortex from the base end of the stem toward the panicle and keeping it at a 30° to 45° angle to the wooden core. (6) The same as test 5 except that the peeling is started at the panicle end of the stem. (7) Pulling the entire cortex away from the wooden core, starting at the base end of the stem and keeping the cortex at right angles to the core. (8) The same as test 7 except that the test is started at the panicle end of the stem. As an indication that retting is incomplete, the resistance to separa- tion of the cortex from the wooden core is almost invariably strongest at the nodes. Fiber branches and wood branches cling to the Teaf 20 BULLETIN 1185, U. S. DEPARTMENT OF AGRICULTURE. scars at the nodes long after the cortex is separated from the core along the internodes. Figure 18 shows on the left a loose-core test made toward the root end of a stem in which retting is incomplete, the fibers being gathered in tangled masses at the nodes. On the right is another stem removed duri ing y the same stage in retting. In this the cortex has been peeled away from the internodes but the fibers cling to the leaf scars. The efficiency of tests 1 and 2 of the loose-core group is about the same, although test 1 may be depended upon to ‘offer resistance to separation more nearly up to the end poimt of retting. Test 2 has another objection in that the projections of wood at the nodes under the leaf scars may catch the fibers and cause them to tangle even after retting is completed. The tests grouped under the heading “ Leaf-scar tests” are char- acterized by a distinct tugging that may be felt at each node as the entire cortex is ZO peeled away from the wooden core. The entire cortex is used, rather than ‘a strip, as the entire cortex affords addi- tional resistance to separation, which it is necessary to over- come in_ splitting the cortex on the op- posite side of the stem from which it is pulled. The core is held stationary and in a horizontal | ’ position while the cortex 1s pulled ver- tically. The great- est resistance to sep- &) aration was found in leaf-scar test 7 FIG. Step 2 in the loose-core test of retting flax ( fig. 19), in which the first Deenk swhere fae cortex has. heen shoved hack,