to be able to make a contribution which would make more specific my approval of the Bionics approach by giving an illustration of how the study of biological organisms might contribute to the solution of a particular engineering problem. To do this in more than vague terms calls for more knowledge of both physiology and engineering than I possess. However, I have been struck by the fact that the steady character of vision seems to have some relation to the fact that the eye vibrates. There are problems in the field of aircraft instrumentation involved in detecting the slow drift of a hovering helicopter or the small changes in altitude of a landing aircraft. It has been suggested that a movement or vibration imported to the instrument might make it a more effective sensor of any new changes in its environment.

I can't pursue this suggestion at length, not being an expert, but it does suggest that perhaps the oscillation of lens and pupil of the eye about which we have also heard at this conference might have a function which we miss and tend to characterize as noise if we start from the mechanical end of the spectrum and mistakenly consider biological organisms as "servomechanisms."

WHERE ARE WE NOW AND WHERE ARE WE GOING?

Otto H. Schmitt
University of Minnesota

Having been assigned the sweeping title "Where are we now and where are we going?" I shall try to accept this broad canvas and shall sketch-in with wide strokes and sweeping generalizations the situation as I see it and as I extrapolate it. In our subsequent discussion I hope you will fill in the picture.

First, I would like to establish the position of this field of interest for which the name, bionics, has been coined among the Physical and Biological areas of science and technology. It is not an especially new field, for if we will allow the subject matter of the present meeting to define it operationally, it has been the principle line of interest in several laboratories for over a decade under such names as biological engineering or applied biophysics.

You are probably aware that biophysics as a science is now becoming firmly established as a regular discipline among the basic sciences with the usual trappings of any respectable science--a Biophysical Society, university departments of biophysics, regular offices in government granting agencies,

etc. This science takes as its basis the concept that important new advances both theoretical and technological will arise through application of the traditional tools of the physical sciences to biological problems and materials. Especially important among these tools is the physicist's trick of mathematical model building and his laboratory practice of using apparatus as technically complicated as may be necessary to provide quantitative documentation of his experimental measurements.

Biophysics now has three major branches, one dealing with structure, one with function, and one with systems organization. At the present time the structure field of biophysics is the one most productive with brilliant work in molecular biology, virus structure, chemical genetics, collagen, and muscle structure. Function biophysics, dealing with problems such as nerve conduction and excitation, muscle contraction, hormone secretion, and bioreceptor functions, has been very active and productive but not so spectacular in its results, perhaps because some of the first cream in this field has been skimmed by physiologists who concentrate in much the same area. The third field dealing with organization, and almost identical in interests with those of this group, has been lagging in comparison with the other two. This is mainly because it is the least familiar of the three and perhaps the most demanding in its interdisciplinary training requirements. You are doing a signal service in drawing attention to this latest growing area with perhaps the greatest potential contribution of all three. A new name offers a new home but I am not sure we are not offering our Russian friends a chance to claim this field, too, if, having accepted sputniks, we now call our field "bio-niks."

I have pointed out that bionics is almost exactly coextensive with the third area of applied biophysics for three reasons. First, because we must not neglect in our enthusiasm for discussing the organization of biological control systems and applying these ideas to engineering, that biological structure and function also offer fruitful fields for theft of design ideas. Naturally, and rightly, we are most immediately concerned about brain function and computer ideas, but let us not fail to put out a little seed in the other gardens.

Second, I would have you look carefully at the structure of scientific and technological societies where these biophysical researchers present their results and their speculations and meet with their fellow workers, in person and through technical publications.

It is very important that there be enough of these organizations to represent adequately our several kinds of interests. It is equally important that we avoid splintering into innumerable special groups at the behest of each enthusiastic opportunist who sees his own narrow area of science or technology as the central focus of all. As a working rule we should have societies about as wide and as narrow as a man's general scientific area of interest, not his special interest. For example, we need both the American Physical Society and the Institute of Radio Engineers. They have quite a few members in common but most of the members of each would feel lost in the other. Similarly we need the Biophysical Society and a corresponding applied function which the PGBME JECMB - IFME complex is trying to serve, but we must not build 10 separate Biophysical Societies and Applied Biophysical Societies.

Third, we must attend to training applied biophysicists. Ask yourself whether your formal education was well designed to qualify you for this field. You are the ones who have the initiative and imagination to plunge into and pioneer in an interdisciplinary field. When this field becomes a recognized discipline instead of interdisciplinary and the easily reached cream is skimmed, then we will need to train people specially to work in it. Those smart enough for this difficult field are also smart enough to know that they will fare better economically in a field for which they are well trained, other considerations being equal. The National Institutes of Health with the cooperation and assistance of several other government agencies and the advice of scientific societies are now launching an interim effort in this direction with several experiments already completed or under way. We are now using the team approach to cover interdisciplinary fields like ours; in the future we will expect basic competence in each member of our teams.

Now I would like to turn to some specific technical problems in our own area which I see as neglected and on which I would like to provoke you into offering advice and consideration.

In making our mathematical models and our mechanismic conceptual models, which are somewhat different, we frequently fall into the comfortable habit of substituting a model with which we can conveniently work for one which embodies the experimentally known properties of the real systems. This is related to the familiar mathematician's trick of making a few changes in a problem which convert it from a real one which is awkward into a pseudo-real problem that is fun mathematically, but may be useless or even misleading scientifically.

I am unsympathetic with the familiar plaint that biology is too complicated and dependent on too many variables for quantitative study. It is not too complicated. We have just been stupid in choosing variables to study and have therewith restricted ourselves to prefabricated mathematics from the physical fields instead of developing new basic forms to meet our needs.

Probably we must fumble a bit before we get our mathematical and conceptual tools sharpened to the point where they can go incisively into problems of neuronal organization but we should soon pass to this second stage. Remember that we begin to believe our models if we talk about them enough even as our ancestors believed in phlogiston. Any competent neurophysiologist or neuroanatomist would groan to learn of the relay logic elements we have been calling neurons and the telephone switchboard networks we have been calling neural nets. At least some of us should start with the now, rapidly accumulating knowledge of the real neuron and its satellite cells, its real electrophysiology, its real archelectronics and its real internal structure, especially its protonic solid state electrical structure through which its biochemical specificity may operate, and we should evolve a calculus appropriate to the real neuron individually and in its normal social environment.

In doing this we should pay attention to the electrical field patterns of coherently excited neuron groups, for electrically this kind of zero frequency directional antenna theory seems a subject quite unexplored and worthy of attention. I mention this topic especially because it bears on the missing area of electrophysiological research between the intracellular recording of single element and the gross recording of huge numbers of cells as in the EEG and ECG. For probing the central nervous system, for selectively stimulating it, and later for coupling the living organism more tightly to its machine auxiliary, we need this kind of access to the organism.

We have rather cavalierly neglected the relationship between the newly discovered facts of DNA action in protein synthesis and coding and the electrical read-in-read-out transducers, probably because almost none of us are simultaneous good biochemists, high polymer physical chemists, neurophysiologists and computer theoreticians, that is to say, adequate biophysicists. Indicentally we almost never look at the rich subject areas of plants in biology.

For some reason our bandwagon has failed to run down certain very rich looking streets of research. In our search for reliable systems using unreliable components, we have failed to exploit adequately many well known biological tricks and to explore other most intriguing facts.

Biology uses freely the nonreplicate redundance of hierarchical overlapping control designs in its structures and in its communication codes which are replete with expendable adjectival modulations of primary symbols. It is interesting, for example, to consider building code languages for the ECG or the EEG which in a very few bits allows the presentation of a record which is not necessarily the same as the original but which is clinically indistinguishable from it diagnostically.

We must have more of these closed-loop tests of our results in bionic systems. In our telemetry work for bioastronautics we should be looking at