The Scientific Method

Agricultural science also must be reasonably balanced among the different fields. Think of the technical problems involved in developing more livestock in the South: High-yielding, disease-resistant pasture and feed crops, their adaptability to soil types and management requirements, suitable breeds of livestock, animal nutrition, control of pests, marketing and storage facilities, farm organization, efficient machinery, and so on. Failure to develop any one of them may keep us from reaping benefits from the other lines of research.

The scientific method of working and thinking that distinguishes a scientist includes observation, often supplemented with planned experiments, classification and correlation of the facts, and the development of ideas. Scientific inquiry is carried out with thousands of special devices and kinds of apparatus. These are constantly being improved so that more facts and more kinds of facts may be gathered. But it is the method, not apparatus, that makes a scientist.

A scientist observes carefully. He records what he sees. For this, he needs training. He must know what to look for and not let significant things escape him. His likes and dislikes do not influence his measurements. His records are complete—not just made up of the facts that are "important," according to some unproved notion, or that are "typical examples," according to some preconceived theory. His observations are as nearly quantitative as possible. A thing is not just long or heavy, but is so long and weighs so much. He exercises judgment on how far to go into detail, based upon experience and the nature of the subject under inquiry. He follows an orderly procedure. Where samples are taken, he uses some definite plan to avoid accidental sorting, even through some unconscious bias of his own. The purpose of observation is to learn—to test old ideas or develop new ones; but the scientist doesn't go about observing at random; some general notion of the end result is necessary.

Ideas and principles do not emerge from a miscellaneous collection of facts. These must be organized into a system of classification. Each unit has to be defined and named. The system must be precise to fulfill its purpose, so we can remember the facts, see relationships, and develop principles.

The accuracy of any classification and its usefulness depend upon how much is known about the subject, and the better the classification, the easier it is to discover new knowledge. Thus the classification is always being changed and improved. Classification itself is a tool. Just because one knows the name of a thing, it doesn't follow that he possesses knowledge about it or can discover new knowledge. In fact, mere knowing of names is useless unless we remember the characteristics, or can learn them by reading or from someone else. But we do need to know names in order to find out what is already known. Much of what a farmer wants to know about his crops, animals, and soils is classified under the names of the things, like Ranger alfalfa, Jersey cows, and Miami silt loam.

Simple observation often does not tell us much about causes or results, so the scientist must set up experiments. Suppose we want to know how to increase yields. Think of all the factors that influence yield: The kind of soil, and how it is managed, weather, insects, diseases, kind of seed, rotations, and many another. All operate at once and, unless we can control the varying influences, or at least account for their separate effects, we cannot tell how any one operates. Thus the problem is broken down and special conditions set up so that all possible factors can be controlled or accounted for. In doing so, the scientist is cautious, careful; his experiments are part of a general plan. On a uniform area of known soil, small plots are laid of the several varieties of soybeans, for example. These all get the same treatment. Careful records are taken of the weather conditions, and the trials are repeated enough to see how the varieties behave with extremes of weather and with normal conditions.

Such would be a simple experiment but not an easy one. Diseases or insects could spoil it. Yields must be measured accurately. Even more refined experiments will be needed to test other variations. On other soil types the same varieties may react quite differently. The various varieties may not each yield the best with the same time of seeding, depth of planting, and fertilizer treatment. They may vary in quality for hay, for food, or for industrial uses. After classifying the results of his experiments, the scientist can draw conclusions about the yield and quality of soybeans that may be expected according to variety, soil type, and the several cultural practices.

The scientist's experiments must be distinguished from ordinary experience. We all make "experiments," in the sense of trying this and that. Some people even publish such experiences as scientific experiments. One can find almost anything "proved" by them—all sorts of new practices hailed as panaceas for solving the farmer's problems or for avoiding ruinous hazards. But before accepting such conclusions, one must insist upon seeing the facts that support the idea and also a clear description of the experiments. Were they so controlled that the results could only have been due to the factors claimed? Or could they have been caused by something else or perhaps come about by pure chance?

Many matters that must come under scientific study cannot be subjected to ordinary experiment. One cannot hold all the stars still and move just one, for example; mountains and volcanoes cannot be moved or controlled; even true soils cannot be gathered into a laboratory or in one place. We can get samples of rocks or soils and subject these small parts to experiment, but that is far from the real mountains or soils as they exist in nature. Indeed, the experimental method is often confused with the scientific method, when, in fact, it is only one part or one method in science. With his instruments for measuring, the scientist often puts himself in a particular position to take advantage of natural phenomena. Astronomers go to places where they can expect a particular view of an eclipse, let us say. Or a soil scientist will take measurements of the temperature, moisture content, and other characteristics of a natural soil as it freezes and thaws. These are sometimes called "natural" experiments, in contrast to the more usual kind that may be set up in the laboratory or greenhouse.

A scientist may study plants in a wild, untouched landscape to see what conditions of soil, climate, light, and so on favor one plant, or one group of plants, over another. Then he will find other places, like the first one, except that the natural vegetation has been burned, or severely grazed, or cleared away and abandoned again. Through a combination of such natural experiments he can develop principles for predicting the kinds of plants and their growth that may be expected under various conditions—principles of immense practical importance in forestry and range management.