in less than one second. There is also the potential of receiving similarly rich information in a few seconds without asking a question and having to wait thousands of years for the answer. Such richness is in our future, although we probably will have to build huge radio systems to achieve that capability. But we know how to do such things now; there is no technology that we don't already have, there will just be a lot to build-billions of dollars worth. Although we could build the system on the earth, it might be better in space-large dishes shielded from the earth by huge screens that keep manmade transmissions out of the system (Fig. 18). With a diameter of 5 kilometers, the system could be one of our most idealistic and grandest projects, perhaps, in the long run, one of the best things we could do with our space transportation system. Whether or not we do this depends on how much wisdom and idealism there is on this planet, and that, of course, is one of the other great questions of life. How good are we? ■

Questions and Answers Question: Does the unit of time that is peculiar to the earth-our year-affect the results of the equation for the number of intelligent civilizations? Drake: It would if we did things in terms of years but the number N is unitless. The equation is a rate of production in things per year times L in years. Thus years cancels out and the unit of time we use doesn't matter.

Question: Why have you chosen inefficient rockets for your examples? Drake: You are getting into the sophistication of rockets. It's true that what is really important to the rocket is momentum ejected rather than energy, and so there are optimized versions of the antimatter-matter rocket. For example, rather than using the energy to expel

gamma rays it can be used to expel hydrogen atoms that serve essentially as propelling pellets. In that way, you can increase the efficiency but only by factors of two, three, or four. Qualitatively there is no difference. With regard to the interstellar ram jet, that, of course, is a nice way to go if you can. But scoops that are hundreds of kilometers across are required to collect the hydrogen atoms, which, in turn, must be funneled to a central point and used efficiently in a fusion reactor. Whether all that technology is possible we do not know. If it were possible you could achieve pretty high speeds.

Question: Is the intent of our listening effort to receive messages or just to

expand technology in this area?

Drake: We are listening in the radio spectrum for a variety of signals but signals that would all be intentionally transmitted. We are looking for continuous wave signals, we are looking for pulse trains, we are looking for drifting pulse trains, we are looking for polarization-modulated waves-all the various things that Maxwell's equations allow in electromagnetic radiation. It's this aspect that's special about the NASA search over previous searches. Previous ones have searched only for continuously transmitted signals at a fixed frequency. The NASA project looks for all varieties of signals, and that's what costs a lot and requires a big computer capacity.

be far away if this ratio is greater than 1. This will be the case if

> 2,

or

In 2 a+

(3)

where N is the integrated number of detectable civilizations in the specified range, P1 is a breakpoint power level, and YP is the maximum power a civilization might radiate. In general, the maximum power will be orders of magnitude larger than P1, and y≫ 1.

Now what does this equation tell us about the bright, detectable civilizations? Are they near or far from us? If the ratio represented by Eq. 2 is greater than 1, the number of bright civilizations detectable despite large distances from the earth will be larger than the number of dim civilizations detectable only when they are close. In other words, the brightest civilizations as seen from the earth are more likely to

In y

In general, the brightest power levels are orders of magnitude larger than the threshold power P1, so the In 2/ In y term in Eq. 3 will be negligible. We can thus simply say that if a < 5/2, the brightest civilizations as detected at the earth are far from us. As the exponent in the power law approaches 5/2, we move to the other extreme: The ratio in Eq. 2 goes to 0. In other words, the dim stars dominate, and we will most likely find our civilizations among the close

stars.

=

The figure above represents plots of the space density of objects emitting at power P versus that power in arbitrary units. Thus the a 5/2 line represents the situation of dim, near objects completely dominating as the type of object detectable at the earth. However, we see that both the plot for cosmic radio sources and, even more so, the plot for stars deviate considerably from the a = 5/2 line, implying that it is the distant, bright civilizations that are more likely to be detected at the earth.■

Frank Drake earned his Bachelor of Engineering, Physics, with honors at Cornell University and his M.S. and Ph.D. in astronomy at Harvard University. While a professor at Cornell he was director of the Arecibo Observatory in Puerto Rico. From 1971 to 1981 he was the director of the National Astronomy and Ionospheric Center, and about three years ago he moved from the east to the west coast to become Dean of Natural Sciences at the University of California, Santa Cruz. Although he has done a variety of work in astrophysics, including research on pulsars and the radio noise from Jupiter, he is most widely known for his belief that intelligent life exists elsewhere in the universe. Beginning in 1960 with pioneering efforts on Project Ozma, he became a leading authority on methods to detect signals emitted by extraterrestrial life. He and Carl Sagan helped design the messages that have left our solar system inscribed on plaques and records aboard the Pioneer and Voyager spacecraft.

cipital cortex. John Sepkoski convinced us that extinction, like speciation, must be regarded as an integral part of evolution, playing the critical role of "making place" for newly evolving species. And Frank Drake projected a cosmos full of life and intelligence and with marvelous humor described efforts to communicate with that intelligent life.

I have consulted with our guests, and they have to a man agreed to a full and free-flowing discussion. I request only that questions and comments be clear and brief. Let us begin.

Audience: I have a question for Frank Drake. What countries are searching for extraterrestrial beings?

Drake: Two countries are making major efforts the United States and the Soviet Union. The Soviets have been searching now for twenty years. In fact, for a long time they were the only people searching. One of their projects, which is based at the Lebedev Physical Institute in Moscow, uses an array of about five radio-frequency receivers placed across the Soviet Union. A similar network is operated from the Gorky Research Radiophysical Institute. Both institutions have, until recently, been looking for short but powerful radio-frequency pulses, a type of signal very different from what we Americans are looking for. They recognize, as we do, that one of the really difficult aspects of a search is selecting the search frequency. Their way of finessing that problem is to look for short pulses, which appear on all frequencies. Their hope is that the extraterrestrials are thinking the same way and are transmitting short pulses.

Now the problem with short pulses is that human activities-operating cars and motorcycles, for instance-produce lots of them. So the Soviets look for short pulses that are coincident in an array of widely separated telescopes. If a pulse is cosmic, it will appear at all stations, but if it is interference, it appears only at one.

So far the Soviets have detected two interesting sources of coincident short pulses. One is the sun, and nobody had known before that the sun emits short radio-frequency pulses. The other was an American reconnaissance satellite that transmits information in the form of big, short radio-frequency bursts over a broad and variable band of frequencies to hinder reception by unfriendly receivers. But the Soviets did pick the signal up, and it got them very excited until they were told what the source was.

One of the problems with the Soviet program is that their small antennas can detect only very strong signals. In fact, to be detected by their system, a source at a reasonable distance of 1000 light years must have a luminosity equal to that of the sun. So the Soviet search will detect only those civilizations with capabilities well beyond those of earthlings, and for that reason the Americans don't think it is very effective.

The Soviets are also building a 70meter steerable, parabolic radio telescope on a mountain near Samarkand, which is to be used not only for conventional radio astronomy but also in a program similar to that of the Ameri

cans.

I should note that Canada, France, the Netherlands, and Australia have also carried out searches, but theirs have been less extensive than the Soviet and American efforts.

Audience: I have a question for Professor Hubel. What chemicals are involved in visual perception, and are the transport mechanisms electronic or ionic? Hubel: Your question has major subheadings. One concerns how nerve impulses are transported along nerve fibers, or axons. There is a certain electric potential-about a tenth of a voltacross the membrane of the axon of a nerve at rest. But when some stimulus reaches the beginning of the axon, ion channels in the membrane there open briefly, positive ions flow into the axon, and the membrane potential changes. The potential change at the next region along the axon is somewhat less, but if it is still great enough to cause ion channels there to open, it is augmented by another influx of positive ions. Because of that positive feedback, the change in potential travels unattenuated along the length of the axon. The impulse travels along the axon like the snap of a rope at one end travels to the other end. Information, rather than any

thing physical, is conducted. But the transport is ionic in the sense that it involves the flow of ions rather than electrons.

When the impulse gets to the specialized structures, the terminals, at the end of the axon, the change in potential there causes release of a substance called a neurotransmitter. The transmitter diffuses to the next nerve and, by changing its permeability to ions, makes that nerve either more or less likely to fire. Between twenty-five and fifty neurotransmitters are known, although as short a time ago as about twelve years only four were known. New ones are being discovered every year. All the known neurotransmitters are very small molecules. Many, like gamma-aminobutyric acid, are amino acids. The enzyme acetylcholine and the hormones epinephrine, or adrenaline, and norepinephrine are among the most common. Why so many neurotransmitters exist is not known.

Audience: But if the transport of a nerve impulse is ionic, how can the impulse travel so fast?

Hubel: The speed of transmission, which ranges from about 1 meter per second to about 100 meters per second depending on the type of axon, is entirely predictable from such factors as the capacitance across the axon membrane and the permeability of the membrane to ions. You apply an equation not much more sophisticated than Ohm's law and out comes the transmission speed. One of the reasons nerve impulses travel so fast is the fact that axons are encased, everywhere except at particular points called nodes of Ranvier, in an insulating sheath of myelin. The flow of ions through the membrane occurs almost exclusively at those nodes, which are about a millimeter apart.

Audience: Is the same mechanism involved in the transport of audio signals? Hubel: There are no basic differences

between the transport of auditory and visual signals. Each nerve system has some very specialized cells, but essentially the same transport mechanism is involved.

Bitensky: Are the neurotransmitters small so they can diffuse rapidly, and does their variety support subtle dialogues among nerves?

Hubel: Well, yes to the first question. The smallness of the molecules probably reflects an evolutionary drive for faster diffusion and easier release and uptake. Concerning the second question, the terminals of certain axons contain many different transmitters, so the opportunity for a much more complex dialogue exists. But I don't know of any cases in which more than two are released. Usually one is a so-called modulator, and the other is really doing the job. The modulator may change certain things, but in fact usually it is not known why more than one is released. It can be shown that one is enough to do the job. Bitensky: Do neurons react to a variety of transmitters?

Hubel: Usually to at least two-one excitatory and one inhibitory.

Audience: My question is addressed to anyone who wishes to respond. In view of the complexities of the human nervous system, do you think computerbased artificial intelligence makes any sense?

Hubel: That is something I think about a lot because I have quite a bit of dialogue with a number of friends who work on artificial intelligence. I think that the majority of people in artificial intelligence are not trying to produce a thinking brain, or anything like one, but to build intelligent machines for image translation, robotics, and so on. Those are very worthwhile goals, so one can't object to them any more than one can object to the goals of, say, electronic engineers. On the other hand, a certain number of people in artificial intelli

gence are trying to learn how the brain works by developing computer programs to solve problems the brain is known to have to solve. They then ask whether the brain solves the problem the same way. Their efforts are very useful because the more people who are thinking about how the brain might work, the more guidance we have as to the type of experiments that we might do. I'm not sure whether that is the answer you

want.

David H. Hubel

Bitensky: The differences between brain and computer are very striking. The brain is terribly slow compared with the computer, but the richness of its interconnections- about 1015 synapses-is far, far greater. Many scientists in artificial intelligence say vehemently that it is just as absurd to try to emulate the brain as it is to try to fly like a bird. Fixed-wing airplanes are quite different from birds. Certainly, many fascinating things may emerge from understanding how the brain solves various problems.

Audience: Would any of the panel care to comment on whether extrasensory perception-ESP-is an unsolved problem in the science of life?

Drake: I'll be glad to answer that one. About once a week I get a letter from someone who tells me I am wasting my time because he or she is already in contact with the extraterrestrials through ESP. My response is always to ask the person to tell me something the extraterrestrials know that we don't know already. So far I've gotten no response. Adding to my skepticism is the large number of experiments conducted daily that very conclusively refute ESP. Those experiments take place primarily in two places-Reno and Las Vegas. The odds of winning some of the games of chance played there, say blackjack or roulette, are about 1 percent lower than the odds of losing. So if even a very few people had enough ESP to foresee or influence what is going to happen even 1 percent of the time, they could become regular winners and run the casinos out of business. The entire gambling industry would collapse. As far as I'm concerned, the fact that the casinos continue raking in the money day by day proves conslusively that ESP does not exist.

Audience: My question is addressed to George Wald. Although Wilder Penfield may have been unable to locate mind as a thing in the cerebral cortex, he very definitely showed that mind as a process is located in specific hardwired structures in the brain. So can't we say the the mind is located totally in the cerebral cortex and in the reticular formation?

Wald: I can only comment. I spend a great deal of time trying to sort out the obviously sloppy ways in which the words mind and consciousness are used. Yes, indeed, we can determine to a degree the pieces of machinery that are involved in the workings of the mind or consciousness. But where does that get us? Some great physicists have essentially said that all matter has an accompaniment of mind. What do they mean by that? They don't mean