The problem of determining the position, extent, degree of branching, and other root characters is a peculiarly difficult one. The complex network of extensive roots and delicate rootlets is laid bare in the soil or recovered from it only with great difficulty. Naturally, one's first thought might be to wash away the soil and thus uncover the roots. In fact, some of the earlier investigators of root habits followed this method.


   Hays 80 (1889), working in Minnesota, was able, by washing away the soil, to obtain useful information on the development of the roots of corn. Some time later (1892), King 121 of Wisconsin devised a method by means of which the roots were supported more or less in their natural position after the soil was washed away. This consisted in isolating a prism of soil several feet long and 1 foot thick by means of trenches dug around it to a depth equal to that of the root system (Fig. 114). The prism of soil was then fitted with a cage made of galvanized iron and poultry-wire netting. When this was in place, sharpened wires were pushed through the soil of the prism in parallel rows along the meshes of the netting. Next, the loose surface soil was taken off and replaced by a block of plaster of Paris which, when it hardened, held the bases of the plants in place. Finally, the soil was removed from the cage by a force pump with a stream of water 1/16 inch in diameter. The cross wires held the roots in position and they were photographed through the wire netting (Fig. 89).

   Fig. 114.--A modification of King's method of isolating root systems in the field. The bases of the plants are held id place by wiring them to narrow boards placed crosswise of the soil block and numerous cross-wires inserted through the prism support the roots when the soil is washed away. (After Edwin C. Miller.)

   Several investigators have used this method: Goff 71 (1897), in Wisconsin, in studying the root systems of strawberry, raspberry, grape, and apple; Ten Eyck 203 (1900) and Shepperd 186 (1905), in investigating the root systems of cultivated plants grown as farm crops in North Dakota, and Ten Eyck (1904) 204 in similar studies in Kansas.

   Various modifications based on this general plan have been used, in some cases the amount of soil being considerably reduced. In other instances, frames with supporting wire nettings in a horizontal position and 2 inches or more apart have been filled with sifted soil after the frames had been placed in holes in the soil just large enough to receive them. Obviously, leaving the soil undisturbed until the roots are grown affords better opportunity for normal root development.

   Miller, working in Kansas, has used rather extensively the method devised by King, although somewhat modified. He points out that it is open to criticism, first, because in order to use it with any degree of satisfaction, the prism of soil must be limited to about 18 inches in thickness, and on this account, only a section of the root system is obtained. Furthermore, the main roots of the plant may not be in the prism of soil which has been isolated; therefore, when the soil is washed away, only a poor representation of the root system is obtained. Finally, although the primary roots of the plants remain on the wires in the same position that they occupied in the soil, it is impossible to obtain all of the finer roots in their normal position. 140 The method is apparently exceedingly laborious, a large supply of water and. many hours of washing being required to remove the soil. The problem of removing the water from the trench is also ordinarily a difficult one. Except in very sandy soils, the pressure of water necessary to remove the soil breaks and carries away many of the finer root branches. When wet, the roots and branches cling together in such a way that it is very difficult to get a clear idea of their number, position, extent, etc., unless they are floated in large trays of water.


   The following methods, used by the writer and his coworkers in the excavation of hundreds of root systems during the past 12 years, have been found very satisfactory. It is for the purpose of aiding those interested in the study of root habits, a number that it is hoped will become increasingly great, that these methods are given in detail.

   Selecting the Plants.--In selecting plants for study, it is very necessary to keep in mind that competition has an effect upon the development of roots as well as upon tops. Hence, typical cultivated plants surrounded by others of their kind and planted at the usual rate should be selected rather than isolated individuals. It is also very necessary to choose areas free from weeds. In fact, it is almost impossible to trace with accuracy the root system of a cultivated plant in an area infested with weeds, such as the milkweed and dogbane (Fig. 28). Where crops are grown for the purpose of root-examination, the work may be greatly simplified by selecting fields which have been fallow and kept free from weeds the preceding season. A soil filled with living roots of previous crops, such as alfalfa and sweet clover, or with the roots of weeds makes root examinations very difficult indeed. In working with native species, it is the usual practice, where possible, to select representative areas where several individuals of two or more species may be examined by excavating a single long trench. Steep banks, old roadside cuts, etc., should be avoided, for it should be clear that plants growing in such situations do not have the normal root habit.

   Excavating the Trench.--When the site is selected, a trench is excavated 8 to 12 inches from the plants to be examined. A long trench (6 to 12 feet, depending upon the kind of plants to be studied) about 2.5 feet wide and 5 to 7 feet deep is most convenient. The trench is dug with vertical walls, and little or no attention is given to root habit during the process. Even if the plants are only 2 to 3 feet deep, a trench with a depth of 4 to 5 feet is most convenient, since it permits easy removal of the soil from the roots. In removing the soil in digging the trench, care is taken not to cover the plants. In many cases, it is convenient to pile the soil in such a manner that all of the walls of the trench may be used in excavating the underground parts, thus affording a larger number of plants for study. The size and shape of the trench varies somewhat with the soil. In very sandy soil it is necessary to have a wide one to lessen the danger of filling the trench by caving in of the walls. In fact, this is often a serious obstacle in recovering deeply rooted plants. On the other hand, trenches only 3 feet wide have been dug in loess soil to depths of 15 to over 20 feet. In such cases, a long trench with different levels is necessary, the soil being removed from one level to another until it is thrown out on the surface. Where the field is on a hillside, the trench is dug at right angles to the slope of the hill and the upper wall of the trench used for root examinations (Fig. 115). As one works into the hillside, the trench becomes deeper. In excavating species with very long, rather superficially placed laterals which extend 8 to 20 feet from the plant, such as pumpkin, yucca, etc., small trenches only 18 inches to 2 feet deep may be dug from the main one. The course of these roots must be determined by carefully picking away the soil, and the lateral trenches must be sunk at the side and behind the partially exposed roots.

   Fig. 115.--One, end of a trench used in excavating root systems.

   Excavating and Describing the Roots.--There is no easy method of uncovering the root system, and unless one is willing to spend considerable time and energy, and exercise a great deal of patience, it is better not to begin. But once started, the work, although difficult, is very interesting and in fact even fascinating. Only a few tools and other equipment are necessary--a small hand pick with a cutting edge on one end and a long narrow tapering point on the other, an ice pick, a tape measure and a meter stick, a pencil, a notebook, and drawing paper. The picks should be kept bright and free from rust so that the soil will not cling to them. This is quite as necessary in root excavations as a sharp knife in cutting microscopic sections.

   If the root habit of the plant is entirely unknown, it is best to begin removing the soil from the side of the trench near the surface. This should be done very carefully and the soil removed in small amounts at a time, holding the pick in a more or less vertical position with the point upward. Unless the surface soil is in good tilth, extreme care must be exercised or the delicate roots will be torn by the falling soil. Root branches, like stems, will resist a much greater pull if the strain is parallel to their direction of growth rather than at right angles to it. Information on the general root extent may be had by rather rapidly following the taproots, or main vertical roots, if any, to their extremities. This may require the removal of the loose soil and the deepening of the trench. Until some acquaintance with the structure of the soil and the general root direction is obtained, the root is very apt to be broken. In such cases the broken end may usually be recovered, with absolute certainty only in the deeper soil where other roots do not occur. It is better to begin again with another plant. After some practice, however, even very brittle roots with a tortuous course can be uncovered. It is never safe to follow a main root or lateral until it becomes very small in diameter and then estimate the remainder of its length. Frequently, such roots, after running for several inches, become larger in diameter, branch profusely, and sometimes continue their course for many feet. There is no relation between root diameter and depth of penetration. The only method insuring certainty is to recover the root ends with root caps. The growing tips are usually enlarged, light in color, and plainly visible.

   If, from studies of the plant in earlier stages of development, the general root habit is known, much time and. labor may be saved by examining the deeper portions of the root system first. Maximum depth, working level, and the details of branching of the deeper roots (which are usually relatively simple) may be obtained before the trench is half filled by removing the soil from the shallower portions. Moreover, in most cases, roots are much more easily excavated from the deeper soil than from the first 2 or 3 feet. This results, in part, from differences in soil texture as well as from the relatively fewer roots and their less extensive branching. Frequently, the deeper soil may be picked away in such a manner that the main root and all its branches are plainly in view throughout a distance of several feet. If the soil has a jointed structure, branching is very often entirely in one plane in the joints. This affords an excellent opportunity for exact counts as to the number, size, extent, degree of branching, etc., as well as for drawing. The moist soil keeps the roots from drying for a considerable period. In this connection, it may be said that where convenient, the trenches should be dug in such a manner that the sun's rays do not strike the roots directly. Not only do they remain fresh for a longer time, but the diffuse light, afforded for careful examination is much better than direct sunlight.

   Several of the main lateral branches should be followed in a similar manner, selecting Arst those that do not penetrate so far into the trench wall that other branches are destroyed in excavating them. With careful, painstaking study, especially after considerable experience has been obtained by practice, what at first may seem like a tangled root mass begins to unfold into a more or less definite pattern--that of the particular species or variety of plant concerned.

   When the depth of the roots of several plants has been determined and the general direction and greatest extent of the laterals are known, enough details will have been observed so that a mental picture of the root habit may be obtained. It remains now to depict this in a working description made in the trench with the exposed roots at hand. in the making of this description, numerous questions, calling for further examination, will arise. Such a description should include, among other things, the number of main roots (or branches from a taproot); their diameters; how rapidly they taper; the exact course they take through the soil; the average (as well as the minimum and maximum) number of branches per inch of main root at various places throughout their course; the diameter, length, and degree of branching, as well as the directions pursued by these laterals; the working -depth, i.e., average depth to which many of the roots penetrate and to which depth much absorption occurs; maximum depth of penetration; peculiarities of root characters; etc. When the description is completed, it should be tried on other plants as yet undisturbed in the trench walls, and any marked variations noted. Such methods are a great aid to exact observation and promote a high degree of accuracy in the work, for if any point regarding the root habit remains indefinite, opportunity is offered for further study. In following root branches, it is frequently necessary to complete the study and description of those near the trench wall before cutting further back to secure the remaining ones. The positions of laterals detached in this process should be plainly marked by means of wooden pegs or otherwise, so that they may be excavated after the root in hand has been followed to its end. Sometimes, it has been found advantageous, where the roots are especially numerous, to extract root portions and float them in trays of water lined with black paper for root counts and examination of minute branching (Fig. 116). In all cases, the trench wall should be at least 12 inches deeper than the root being excavated. The soil crumbles away and falls from the pick with far less danger of breaking the roots than. where the root is entirely surrounded with soil. Under the latter condition, satisfactory excavation is impossible.

   Fig. 116.--Portion of a root of corn floated in a tray of water, showing detail of branching.

   The ease or difficulty with which a root system may be recovered depends to a considerable extent upon the texture and water content of the soil and also upon the nature and length of the roots themselves. The amount of time necessary depends, of course, upon the extent and complexity of the root system as well as upon the experience of the investigator. Only a few hours may suffice for a thorough examination of a plant in its early state of development, but a very extensive root system may engage the attention of two or three persons for several days. In general, the roots of perennial native species are less fragile and are often more easily uncovered than those of cultivated plants, although great variation has been found in both groups. Under no conditions should root descriptions be made elsewhere than in the trench or the work concluded until several plants of the species have been carefully examined and a typical root system completely described and preferably drawn.

   Photographs and Drawings.--Working in soils of different textures, etc., throughout a wide range of habitats usually makes it possible to find conditions under which a root system may be excavated and removed from the soil practically in its entirety. Many such specimens arranged on appropriate backgrounds have been photographed or preserved (Figs. 20, 107). Aside from this procedure being very difficult and time consuming, certain other objections may be mentioned. In a root system that is at all extensive, details are lost in reducing it to the size of the photographic film or plate. Invariably, many of the finer branches and root ends are obscured. Moreover, all of the delicate rootlets and even some of considerable size dry and wither when the plant is taken from the moist soil. It is only rarely that a root can be photographed in place, conditions being most favorable when the branches are coarse and the color of the root contrasts more or less strikingly with that of the soil. Frequently, portions of root systems may thus be shown to considerable advantage (Figs. 37, 96).

   A more satisfactory method is to draw the roots on a large scale in the field. A scale of 4 inches to 1 foot, i.e., a reduction of two-thirds, is usually sufficient to. permit the picturing of even the finest branches. Such a drawing, retraced in India ink, can be greatly reduced in the engraving. Where the root system exceeds 4 to 5 feet in extent, a smaller scale is necessary. A usual one for mature plants is 2 inches to the foot. Penciled drawings are made in the field simultaneously with the excavation of roots and always to exact measurements. In the drawings, the root systems are arranged as nearly as possible in the natural position in a vertical plane; that is, each root is placed in its natural position with reference to the surface of the soil and a vertical plane from the base of the plant. In a few cases, roots have been so abundant that only half of the root system was shown. The drawing paper should be ruled in squares and mounted on a board of convenient size so that it can readily be taken into the trench when necessary. Two persons to excavate the roots, each with the advantage of having his findings checked by the other, a third to record the notes, and a fourth to draw the root system, greatly accelerate progress. In every case, it is sought to represent the typical root condition rather than the extreme. Such drawings, carefully executed, represent the extent, position, and minute branching of the root system even more fully and accurately than a photograph.

   In certain plants, like cacti, pumpkin, or even corn, where many of the roots run horizontally in the surface soil, drawings in the horizontal plane showing the appearance of the roots upon the removal of the surface soil are instructive (Fig. 44). Combinations of the two methods might be used advantageously for many species. Sometimes, the relation of roots to tops may be shown conveniently by carefully dissecting out all of the roots in a 4-inch vertical layer of soil in the wall of a trench and illustrating them and the tops by photographs or drawings. This bisect method shows, in a striking manner, the interrelations of underground and aboveground parts of the same or different species. 221