I once signed up for a course in plant physiology. On the first day of class the professor explained to us his philosophy of teaching. We would not use a regular textbook, he said, because “a book can’t teach you any plant physiology - only a plant can teach you plant physiology.” Therefore, he further explained, the emphasis in the course would be on lab work. He would function only as a resource person (though I expect he would reject that particular term.) He would prepare no lectures and assign no reading. Instead he would set us going on laboratory experiments and the lecture sessions would be devoted to answering our questions on the experiments.

      Is this a good way to teach plant physiology?

      The use of laboratory exercises in the teaching of science is firmly entrenched, and I believe rightly so. However I also believe that the rationale of such exercises is seldom critically examined, and this leads to a considerable, and needless, amount of waste, inefficiency, confusion, and frustration. I speak from the perspective of both student and teacher. I also speak from the perspective of one who is intolerant of waste, inefficiency, confusion, and frustration. Most of the examples and illustrations I cite come from my experience as a student in college science courses, though I believe similar examples could be found at any level.

      There are a number of reasons for providing laboratory experiences in a science course. Among them are:

      - to demonstrate concepts and principles of science

      - to teach scientific methods and attitudes

      - to provide sensory input

      - to give the students a “feel” for the subject matter

      - to keep the subject grounded in reality

      All of these reasons have some validity, but some have more validity and importance than others. The first reason, demonstration of scientific principles, seems perfectly reasonable and probably accounts for the bulk of all laboratory exercises done by students. This goal is important, but there are problems. Laboratory work is not the only way to approach this goal, and laboratory work can be very expensive in terms of time and effort.

      As an example consider the principle that a body displaces its own volume when immersed in water. To help students understand and remember this principle it is desirable for them to see water overflow when an object is immersed in it, and to measure the volume of this overflow and compare it to the volume of the object. But how much time should be spent in this demonstration? Is it worthwhile to spend a full hour on it? Or a half hour? Or just fifteen minutes? Assume for a moment that the principle can be demonstrated in ten minutes - which I think is a very conservative estimate if each member of the class is expected to dunk an object and measure the overflow - now what should be done about the principle that a larger object weighs more than a small object of the same composition? Is this principle also worth ten minutes of class time? Both of these principles are very elementary, but they are important. Another very elementary principle is that two objects of the same volume but of different compositions may weigh different amounts. Is this principle also worth ten minutes of class time?

      My point, of course, is that there are too many principles for all of them to be demonstrated. Therefore the simple fact that an idea is a scientific principle is not in itself justification for spending time on its demonstration.

      Is a demonstration of a principle necessary for the learning of that principle? I don’t think it is as a general rule. I have never seen a formal demonstration of the principle that a fluid going through a narrow place in a tube increases its speed and decreases its pressure. I have never seen a formal demonstration that a muscle can only pull, it cannot push. I have never seen a formal demonstration that sound travels faster in water than in air. And I could name any number of other examples of principles which I have learned but which have not been formally demonstrated to me.

      If a principle had to be demonstrated to be learned we would be in bad shape, for some principles simply cannot be demonstrated. What should we do about Newton’s first law, the very basic principle that an object a rest will remain at rest until acted on by some force and an object that is moving will continue to move in a constant direction and speed until acted on by some other force? The first part of this principle presents no obstacles to students, but the second part can throw them into fits. “Everything’s got to slow down eventually,” they say, reasoning that everything they have ever encountered in their lives slows down of its own accord eventually.

      Faced with this reaction on the part of the students it would seem logical to demonstrate the principle for them. Unfortunately it is not that easy. I would argue that this principle is simply and completely impossible to demonstrate. One may speak of an ice skater slowing down slower than a roller skater slows down because of less friction, but that is not a really good example since the students know perfectly well that both will slow down eventually. One may speak of measuring the friction of rolling versus sliding objects, or talk about wind friction, or point out that the moon and planets don’t slow down, and many other things. But in the final analysis it is simply not possible to actually demonstrate the principle. Still we expect students to learn the principle, even though we cannot demonstrate it and even though it may seem to go against common sense.

      This example illustrates a very important principle about laboratory exercises: It is not the demonstration itself that is important - it is the interpretation of the demonstration. Students have seen ice skaters and ball bearings, and they’ve felt the wind while riding their bikes, and they’ve observed the moon going through its phases. Is the sum of these experience an adequate demonstration of Newton’s first law? It was for Isaac Newton, but for the rest of us it is not. The rest of us must have these experiences carefully interpreted to us, in plain simple English, if we are to extract the principle from them.

      With adequate interpretation demonstrations can be very simple. With adequate interpretation simply looking down the throat of a carburetor is a good demonstration of the venturi principle, or looking at a book lying on top of a desk is an adequate demonstration of the principle of inertia, or looking a frost on a window is an adequate demonstration that water can exist in the air even though it can’t be seen. Without an adequate interpretation, all these principles would be missed by even the most astute students.

      Because the interpretation of a demonstration is so much more important than the demonstration itself, then it follows that staging a demonstration is no guarantee that the principle will be understood and remembered. Truth will not speak for itself. A class of children dunking objects and measuring overflow may or may not be advancing their scientific education. It is up to the teacher to make sure that they know what they are doing and not just going through the motions. The teacher does this by giving careful explanation and by eliciting feedback from the students. If the teacher fails in this job then all the dunking of objects and measuring of overflow in the world is irrelevant. If the teacher succeeds, then the dunking and measuring may be relevant and well worthwhile, though of course it is still not essential. A good explanation of the principle is the only thing that is essential.

      Almost any demonstration or experiment is time consuming, some much more so than others, and many demonstrations and experiments require attention to a considerable amount of detail. This time and effort may be beneficial if it serves to focus attention to the principle being demonstrated. Or it may be detrimental. The students may get so wrapped up in the details that they lose sight of the purpose of the exercise, or the results may be so far removed in time from the start of the exercise that they become meaningless. For example in the dunking and measuring experiment there would be little benefit if the students get lost in the mathematical computations of the volumes they are measuring. The principle itself is simple and easily understood, but its demonstration may not be so simple.

      My worst experience in getting lost in the details of an experiment occurred in my first college chemistry course. Week after week I never seemed to know what was going on. I would work long hours getting down all the steps in the experiments but was doing little more than following a recipe. After a month or so I discovered what the trouble was. The course consisted of lectures by the professor and lab under a graduate assistant, and the lectures got about a week behind the lab work. When I took the initiative to read ahead things began to make a little more sense.

      Why did it take me a month or two to realize that the lectures were behind the lab? It would seem simple enough to figure out that a particular experiment illustrates a principle in Chapter Five, not Chapter Four. My explanation is that I was so tied up in details that I didn’t know what principle a particular experiment was designed to demonstrate.

      Even if the mass of details do not cause confusion, they may still become tedious and hence be frustrating and counterproductive. This occurred in a number of science courses I took in college. Once in a physiology course we spent a full three hour lab period showing that iodine is concentrated in the thyroid gland. We did this by injecting a rat with radioactive iodine and later cutting out segments of various organs and measuring their radioactivity with a Geiger counter. I use the word “we” in a rhetorical sense. There was one rat for each table of six students, which meant that I mostly sat around and watched while somebody else did the work. The principle that iodine is concentrated in the thyroid gland was familiar to me, from tenth grade biology if not earlier, so I wasn’t really interested in the results I would have gladly skipped the whole affair.

      In microbiology, as another example, out of fifteen laboratory sessions of three hours each, at least six sessions were devoted to a series of experiments designed to show the affect of different conditions on microbial growth. First we did an experiment growing bacteria at different temperatures. Then we did a very similar experiment to show how ultraviolet light affects microbial growth. Then we showed the effects of the presence or absence of oxygen on growth. Then we did experiments with disinfectants, antiseptics, antibiotics, and soon. Each of these experiments filled the three hour lab session and required more work both before and after the lab period. I never got confused by these experiments, but I surely got tired of them, for they were all so much the same. I think it would have been much better if we did only the first experiment in full and had the rest of the experiments condensed into ten minute demonstrations by the instructor.

      With all these problems that I have discussed one might then ask what principles, if any, should be demonstrated? My view is that only those principles that can be easily demonstrated should be demonstrated, and the demonstration should be simplified as much as possible. For example it takes about five seconds to demonstrate that a heavy object and a light object both fall at the same speed. Therefore it is well worthwhile to do so. However it is not worthwhile to set up an elaborate experiment in which the students carefully weigh different objects and use a stop watch to time their rates of descent. Similarly it takes only a moment to show that oil and water won’t mix, or that a convex lens can concentrate sunlight and start a fire, so these demonstrations may be well worthwhile. The interpretation of these demonstrations may take considerable time, but that is what teaching is all about. The important thing is not to get tangled up in an unproductive and nonessential web of details.

      To teach the scientific method is a rationale for laboratory work that many teachers consider the most important of all. Here again I take a different view. If one believes that the scientific method is not easily learned, that it can be learned only in a laboratory setting, and that failure to understand the scientific method makes it impossible to learn scientific facts and principles, then one is certainly justified in spending considerable time in laboratory work designed to demonstrate the scientific method. I will dispute all three of these ideas. The scientific method is not hard to learn and understand, it is not learned best in a laboratory setting, and an understanding of the scientific method is not essential for the learning of science.

      Children learn at an early age that the world is round, and that plants need sunlight to grow, and that carbohydrates are energy foods, and a lot of other things, with no thought of scientific methods. In a similar way one an learn that ATP is an energy carrier, or that Botrychium belongs to the order Ophioglossales, or that Jupiter has a gravity twelve times that of the earth, or any number of other things, without reference to the scientific method. Of course, I would not advocate deleting all reference to scientific methods in our teaching for it is only sensible to present it somewhere along the line. My point of contention is that not too much time should be spent in demonstrating it in every course. Once it is presented and discussed it need not be endlessly repeated at every opportunity. Requiring the students to verbalize the scientific method they use in every laboratory exercise can be aggravating. I think a class period or two devoted to explaining the scientific method in the eighth grade is quite sufficient. Thereafter only specialized scientific methods that pertain to specific topic need be discussed as they arise.

      I have always objected to speaking of “the scientific method” in the singular. Normally when we say “the scientific method” we are referring to contrived and controlled experimentation in a laboratory setting. This is not the only method of science. A second method is imaginative interpretation of common phenomena. Both astronomy and geology in their early development depended very heavily on this method. Copernicus could not experiment with the stars and planets, but by carefully analyzing the same observations that men had made for centuries he concluded that the earth went around the sun, not vice-versa, and astronomy began to be a science. James Hutton could not experiment with millions of tons of rock and millions of years of time, but by his imaginative interpretation of the same earth we all see he began the science of geology.

      A third method of science is simply massive and systematic observation. Linnaeus made good use of this method. A fourth method of science is the application of mathematics to physical phenomena. This method is very basic to physics. There are probably many other methods of science that could be identified.

      Contrived experimentation is only one scientific method out of many. In some areas of investigation it is the only suitable method. In other areas it is totally irrelevant. Even if one prefers to think of contrived experimentation as “the” scientific method, it does not follow that it can be demonstrated only in the laboratory, or even that it can best be demonstrated in the laboratory. All of the objections I made to demonstrating scientific principles also apply to the use of laboratory work for demonstrating “the scientific method.” Therefore I consider teaching of the scientific method as a very minor reason for laboratory exercises.

      I believe the third reason I listed, to provide sensory input, is the most important reason for laboratory work in most science courses, especially in biological fields. The structure of any subject consists of concepts and associations connecting these concepts. In the earliest learnings of an infant concepts must arise by associating sensory data. Certain sights, sounds, and feelings that are consistently experienced together come to be recognized by the infant. he begins to form concepts, in other words. Probably the first concept the infant learns would be “Mama”. Other sights, sounds, and feelings lead to the primitive concepts of “bottle”, “bath”, “potty”, and so on. As the child grows he begins to attach words to these concepts. He begins to acquire language. This enables him to begin to learn more abstractly. Instead of being dependent on direct sensory contact with the part of the world that he is learning about, he can be guided by others through the medium of language to manipulate and expand the concepts he already has. Primitive concepts give way to more advanced concepts. More advanced concepts eventually lead to the vast structures of knowledge that we call science, history, mathematics, and so on.

      Sensory input, other than as a channel of language, becomes less and less important the more one learns. However most subjects still require some direct sensory input. A few subjects, such as art and music, would be impossible to teach through words along. Other subjects, such as law or advanced mathematics, could be taught with no sensory input other than words. Most subjects are between these two extremes. Most subjects require, or at least can use, some direct sensory input. For example in history it is highly desirable to have pictures in the textbook. One could know something about what Washington looked like by reading a verbal description, but only the direct sensory input of a portrait can really convey the message. In arithmetic as another example, it may be quite possible to do problems without ever drawing diagrams, but it would not be sensible to do so.

      In science there are some things that can be learned only by direct sensory contact. When this is the case then laboratory work is indispensable. It is possible to read a verbal description of a stickleback fish, for example, and have some idea what it looks like, but if we are to do the job right we must have an actual specimen. It is possible to learn the characteristics that distinguish an elm leaf from a hackberry leaf, but one cannot be sure of telling them apart unless one can examine the real thing. It is possible to learn a tremendous amount about the human body from a textbook, but we wouldn’t want to train doctors without cadavers.

      The sensory data that a student is expected to make sense of can be very subtle. When this is the case then it is especially important to have direct sensory contact with the phenomenon being studied. I well remember trying to make sense out of microscope slides in my first college zoology course. I would be mystified at times when I could see no relation at all between the actual specimen under the microscope and the diagram in the book. The things that I was supposed to be seeing were there of course, but they were subtle. It took a while to learn to seem them.

      I remember this problem especially well because of an incident that occurred about a year later. Right before class in another zoology course I observed a student asking the professor for help. This student was apparently in the beginning zoology course and was having the same trouble I had had in making sense out of the microscope slides. He was explaining his difficulty to the professor and asking what he might be doing wrong. I was struck by the professor’s inability to be of any help. He could not understand the student’s problems. From my vantage point, however, I could clearly see both sides. After a few months of practice I had learned what to look for under the microscope. I had organized the sensory data, in other words, and no longer had any trouble. But it was only a year previously that I couldn’t make head nor tail out of anything so I could also understand the student’s problem. Only with considerable practice can one learn to interpret the very subtle sensory data that the microscope makes available. The professor did not know that the sensory data was subtle, for it had been many years since he had been a student himself.

      Organizing sensory data takes time and effort. In this way it is similar to demonstrating a principle or carrying out a contrived experiment. But unlike demonstrating a principle or doing an experiment this time and effort can not be abbreviated or dispensed with. Many a three hour experiment, such as the iodine-in-the-thyroid experiment I mentioned, could be profitably condensed into a ten minute demonstration. But a three hour laboratory session in which students are provided with new sensory data cannot be condensed into then minutes. One simply cannot spend ten minutes looking a series of microscope slides and get the benefit that three hours of careful study would bring, or spend ten minutes looking at an unfamiliar group of insects and learn as much as in three hours.

      My course in introductory zoology, like the microbiology course I mentioned, had fifteen laboratory sessions of three hours each. Like the microbiology lab sessions, the zoology lab sessions were tedious at times, but unlike the microbiology lab sessions they did not become frustrating. The difference is very simple. In microbiology lab we were demonstrating principles, but in the zoology lab we were getting sensory input. Each week we would have preserved specimens of some new and obscure group of animals to examine and dissect. The sessions were not redundant. They complemented, but did not repeat, what we had learned from the lectures and textbook. Provision of sensory input is a good reason to do laboratory work. Demonstration of principles is not such a good reason.

      The learning of specific skills and techniques is also a valid reason for laboratory work. If a skill must be learned in a subject then the appropriate laboratory setting must be provided. Learning to use a microscope, for example, or learning to use a separatory funnel in chemistry, or to use an oscilloscope in physics, can only be done with “hands on” experience.

      The learning of skills and techniques is very similar to the organizing of sensory data. In learning a skill the emphasis is on motor output, rather than on sensory input. However motor output is very much dependent on sensory input, particularly proprioceptive input. For example, learning to focus a microscope involves making sense out of the feel of the focusing knobs, and relating the visual results to the manipulation of these knobs. Thus learning to focus a microscope is largely a matter of organizing sensory data.

      The common denominator of organizing sensory data, on the one hand, and learning motor output on the other hand, is that neither is primarily a cognitive process. Neither can be put into words. Neither can be put into a textbook or a lecture. Thus both are good reasons for laboratory exercises.

      I tend to think of learning techniques as a less important reason for laboratory work than providing sensory data simply because most science courses are aimed at a cognitive understanding of the subject, not a proficiency in a technical area. But obviously there could be exceptions to this. In a medical technology course, for example, proficiency in techniques would be of utmost importance. However I cannot think of any situations in high school or introductory college science courses where techniques would be more important than understanding.

      The rationale that laboratory exercises should be provided to give the students a “feel” for the subject matter can be misleading I think. The goal is legitimate, but it s not something that can be approached directly. One gets a feel for a subject matter by learning the subject matter. The idea of giving a student a feel for a subject in science is similar to the idea of teaching appreciation in music. Having some experience in teaching music myself I have concluded that appreciation is not something that can be approached directly. The way to teach appreciation of music is simply to teach music. Hopefully appreciation will follow. If it does not then there is nothing to be done about it. Similarly one hopes that in learning biology one will get feel for the subject, but again if that feel does not materialize there is little to be done about it.

      Since it is not possible to approach this goal directly, there is always the temptation to approach it indirectly. More specifically it is tempting to think that if something is hard to understand out of the textbook then it will be easier to understand if one does something with his hands - lab work in other words. My viewpoint is that this doesn’t work. As I have already pointed out a demonstration or experiment can be misinterpreted, and interpretation is very important. My viewpoint is that one gets a feel for the concepts and principles of a subject from the textbook, but not in lab, and one gets a feel for laboratory work in the lab, not from the textbook.

      The final rationale for lab work that I have listed is to keep the subject grounded in reality. I think this is a very important idea. It’s very easy to build castles in the air, especially when one has a captive audience as teachers do. A prime example of a scientific reality gap is the idea that heavy objects fall faster than light objects. I believe I am correct in saying that this idea was blindly accepted for centuries before anyone bothered to check it out. Lab work eventually closed this gap.

      As important a reality grounding is, however, I don’t think it is a good rationale for lab work in a science course. Reality grounding is important in basic research. The reality testing that really counts is that done by practicing scientists, not by students. In a college chemistry course I once tested Avogadro’s number and found Avogadro was off by about a factor of a hundred. But I never put much faith in this particular bit of reality testing. Students cannot test every scientific idea any more than they can demonstrate every scientific principle. There just isn’t enough time.

      The term “reality grounding” can be interpreted a little differently. It can mean reconciling the diagram given in a textbook with the sight of the actual specimen. This is a very important bit of reality grounding. If a student is capable of correctly labeling every part of a diagram of a mammalian heart, but can’t tell a ventricle from an aorta on a real specimen, there is certainly a reality gap, and this reality gap can only be closed by laboratory work. However I think this situation is better interpreted in terms of organizing sensory data than in terms of reality grounding. Provision of sensory input is the third rationale I discussed, and as I have already pointed out, I consider it to be the most important rationale for laboratory work.

      There is a type of reality grounding that is very important. That is concerned with the teacher knowing what is going on in the minds of his students. If the teacher thinks the students know all about Chapter Three, when in fact they never really understood Chapter Three but managed to pass the chapter test by memorizing meaningless phrases, then there is a reality gap. But this is a matter of teaching, not a matter of science. More lab work will not close this reality gap, but may add to it.

      In conclusion I will summarize the main points I have made in this article. The most important reason for doing lab work in a science course is to provide sensory input. This goal is not just desirable, it is indispensable. The second most important reason is to learn specific techniques and skills. This again is not just desirable, it is indispensable. All of the other rationales are less important. Demonstrations of scientific principles and scientific methods are desirable only if they are not excessively expensive in terms of time and effort. A principle can be learned without a formal demonstration, and a demonstration is not nearly so important as the correct interpretation of the demonstration. The last two rationales I listed, to get a feel for the subject and to keep the subject grounded in reality, are either goals that cannot be approached directly, or are better interpreted as the other rationales, or simply don’t apply to lab work.

      With the perspective of these ideas I will return to the question I asked at the beginning of this article. Is laboratory work, and only laboratory work, a good way to teach plant physiology? My opinion now, after having spent many hours thinking it through and putting my ideas down on paper, is the same opinion I had at that first class session of plant physiology. I immediately dropped the course and took plant taxonomy instead.