Research methods in ecology

by Frederic E. Clements

48. Depth of samples. The general rule is that the depth of soil samples is determined by the layer to which the roots penetrate. The practice is to remove the air-dried surface in which no roots are found, and to take a sample to the proper depth. This method is open to some objection, as the actively absorbing root surfaces are often localized. There is no practical way of taking account of this as yet, except in the case of deep-rooted xerophytes and woody plants. It is practicable to determine the location of the active root area of a particular plant and hence the water-content of the soil layer, but in most formations, roots penetrate to such different depths that a sample which includes the greater part of the distance concerned is satisfactory. Some knowledge of the soil of a formation is also necessary, since shallow soils do not require as deep samples as others. The same is true of shaded soils without reference to their depth, and, in large measure, of soils supplied with telluric water. In all cases, it is highly desirable to have numerous control-samples at different depths. The normal cores are 12 or 15 inches; control-samples are taken every 5 inches to the depth desired, and in some cases 3–inch sections are made. It has been found a great saving of time to combine these methods. A 5–inch sample is taken and placed in one can, then a second one, and a third in like manner. In this way the water-content of each 5–inch layer is determined, and from the combined weight the total content is readily ascertained.

49. Check and control instruments. A number of instruments throw much light upon the general relations of soil water. The rain-gauge, or ombrometer, measures the periodical replenishment of the water supply, and has a direct bearing upon seasonal variation. The atmometer affords a clue to the daily decrease of water by evaporation, and thus supplements the rain-gauge. The run-off gauge enables one to establish a direct connection between water-content and the slope and character of the surface. The amount and rapidity of absorption are determined by means of a simple instrument termed a rhoptometer. The gravitation water of a soil is ascertained by a hizometer, and some clue to the hygroscopic and capillary water may be obtained by an artificial osmotic cell. All of these are of importance because they serve to explain the water-content of a particular soil with especial reference to the other factors of the habitat. It is evident that none of them can actually be used in exact determinations of the amount of water, and they will be considered under the factors with which, they are more immediately concerned.

Physical and Physiological Water

50. The availability of soil water. The amount of water present in a soil is no real index to the influence of water-content as a factor of the habitat. All soils contain more water than can be absorbed by the plants which grow in them. This residual water, which is not available for use, varies for different soils. It is greatest in the compact soils, such as clay and loam, and least in the loose ones, as sand and gravel. It differs, but to a much less degree, from one species to another. A plant of xerophytic tendency is naturally able to remove more water from the same soil than one of mesophytic or hydrophytic character. As the species of a particular formation owe their association chiefly to their common relation to the water-content of the habitat, this difference is of little importance in the field. In comparing the structure of formations, and especially that of the plants which are found in them, the need to distinguish the available water from the total amount is imperative. Thus, water-contents of 15 per cent in gravel and in clay are in no wise comparable. A coarse gravel containing 15 per cent of water is practically saturated. The plants which grow upon it are mesophytes of a strong hydrophytic tendency, and they are able to use 14½ or all but .5 out of the 15 per cent of water. In a compact clay, only 3½ of the 15 per cent are available, and the plants growing in it are marked xerophytes. It is evident that a knowledge merely of the physical water-content is actually misleading in such cases, and this holds true of comparisons of any soils which differ considerably in texture. After one has determined the physiological water for the great groups of soils, it is more or less possible to estimate the amounts in the various types of each. As an analysis is necessary to show how close soils are in texture, this is little better than a guess, and for accurate work it is indispensable that the available water be determined for each habitat. Within the same formation, however, after this has once been carefully ascertained, it is perfectly satisfactory to convert physical water-content into available by subtracting the non-available water, which under normal conditions in the field remains practically the same.

The importance of knowing the available water is even greater in those habitats in which salts, acids, cold, or other factors than the molecular attraction of soil particles increase the amount of water which the plant can not absorb. Few careful investigations of such soils have yet been made, and the relation of available to non-available water in them is almost entirely unknown. It is probable that the phenomena in some of these will be found to be produced by other factors.

51. Terms. The terms, physiological water-content, and physical water-content, are awkward and not altogether clear in their application. It is here proposed to replace them by short words which will refer directly to the availability of the soil water for absorption by the plant. Accordingly, the total amount of water in the soil is divided into the available and the non-available water-content. The terms suggested for these are respectively, holard (ὅλos, whole, ἅpδov, water), chresard (χοῆςις, use), and echard (ἕχω, to withhold).

52. Chresard determinations under control. The determination of the chresard in the field is attended with peculiar difficulties. In consequence, the method of obtaining it under control will first be described. The inquiry may be made with reference to soils in general or to the soil of a particular formation. In the last case, if the plants used are from the same formation, the results will have almost the value of a field determination. When no definite habitat is the subject of investigation, an actual soil, and not an artificial mixture, should be used, and the plants employed should be mesophytes. The individual plants are grown from seeds in the proper soils, and are repotted sufficiently often to keep the roots away from the surface. The last transfer is made to a pot large enough to permit the plant to become full-grown without crowding the roots. The pot should be glazed inside and out in order to prevent the escape of moisture. This interferes slightly with the aeration of the soil, but it will not cause any real difficulty. The plant is watered in such a way as to make the growth as normal as possible. After it has become well established, three soil samples are taken in such a manner that they will give the variation in different parts of the pot. One is taken near the plant, the second midway between the plant and the edge of the pot, and the third near the edge. The depth is determined by the size of the pot and the position of the roots. The holard is determined for these in the usual way, but the result is expressed with reference to 100 grams of dry soil; the average is taken as representative. The soil is then allowed to dry out slowly, as sudden drouth will sometimes impair the power of absorption and a plant will wilt although considerable available water remains. Plants often wilt in the field daily for several successive hot dry days, and become completely turgid again during the night. If the drying out takes place slowly, the plant will not recover after it has once begun to wilt. The proper time to make the second reading is indicated by the pronounced wilting of the leaves and shoots. Complete wilting occurs, as a rule, only after the younger parts have drawn for some time upon the watery tissues of the stem and root, by which time evaporation has considerably deceased the water in the soil. It is a well-known fact that young leaves do not wilt easily, especially in watery or succulent plants. Three samples are again taken and the average water-content determined as above. This is the non-available water or the echard. The latter is then computed on the basis of 100 grams of dry soil, and this result is subtracted from the holard to give the chresard in grams for each 100 grams of dry weight. The chresard may also be expressed with respect to 100 grams of moist soil. As a final precaution in basal work, it is advisable to determine the chresard for six individuals of the same species under as nearly the same conditions as possible. When it is desired, however, to find the average chresard for a particular soil, it is necessary to employ various species representing diverse phyads and ecads. Such an investigation is necessarily very complicated, and must be made the subject of special inquiry.

53. Chresard readings in the field. The especial difficulties which must be overcome in the field are the exclusion of rain and dew and the cutting off of the capillary water. It is evident, of course, that experiments of this sort must also be entirely free from outside disturbance. The choice of an area depends upon the scope of the study. If the chresard is sought for a particular consocies, the block of soil to be studied should show several species which are fairly representative. In case the chresard of a certain species is to be obtained, this species alone need be present, but it should be represented by several individuals. Check plots are desirable in either event, and at least two or three which are as nearly uniform as possible should be chosen. The size and depth of the soil block depends upon the plants concerned. It must be large enough that the roots of the particular individuals under investigation are not disturbed. There is a limit to the size of the mass that can be handled readily, and in consequence the test plants must not be too large or too deeply rooted. The task of cutting out the soil block requires a spade with a long sharp blade. After ascertaining the spread and depth of the roots, the block is cut so that a margin of several inches free from the roots concerned is left on the sides and bottom. If the block is to be lifted out of place, so that the sides are exposed to evaporation, this allowance should be greater. In some cases, it may be found more convenient to dig the plant up, place it in a large pot, and put the latter back in the hole. As a general practice, however, this is much less satisfactory.

After the block has been cut, it may be moved if the soil is sufficiently compact, and then allowed to dry out in its own formation or elsewhere. The results are most valuable in the first case, though it is often an advantage to remove blocks cut from shade or wet formations to dry, sunny stations where they will dry more rapidly. The most satisfactory and natural method, however, is to leave the block in place, and to prevent the reestablishment of capillary action by enclosing it within plates. This is accomplished by slipping thin sheet-iron plates into position along the cut surfaces. The plate for the bottom should be somewhat wider than the block, and is slipped into place by raising the block if the soil is not too loose; in the latter event, it is carefully driven in. The side plates are then pushed down to meet the former. The size of the plates depends upon the block; in general, plates of 1, 2, and 3 feet square, with the bottom plates a trifle larger, are the most serviceable. Access of rain and dew is prevented by an awning of heavy canvas which projects far enough beyond each side of the block to prevent wetting. The height will depend of course upon the size of the plants. The awning must be used only when rain or heavy dew is threatened, as the shade which it produces changes the power of the plant to draw water from the soil.

The time necessary to cause wilting varies with the habitat and the weather. When the block is large and in position, two or three weeks are required. This period of drying incidentally furnishes an excellent opportunity for determining the rate at which the particular soil loses water. The holard sample is taken daily for several days before the block is cut out, in order to obtain an average, care being taken of course to avoid a period of extreme weather. The echard samples are taken as soon as the wilting is sufficient to indicate that the limit of available water is reached. The air-dry soil above the roots is first removed. The treatment of the samples and the computation of the chresard are as previously indicated.