''Our research leaves no doubt in my mind that today's genetic engineering threshold can only be compared to the splitting of the atom,'' proclaims market analyst James Murray.
They think big in Chicago, capital of the US agricultural heartland, where Enrico Fermi got the first atomic reactor running 39 years ago.
Last May, Policy Research Corporation, the consulting firm that Dr. Murray heads, joined with another such firm, The Chicago Group Inc., to publish what they modestly called ''An Assessment of the Global Potential of Genetic Engineering in the Agribusiness Sector.'' It sells for the equally modest price of $1,250 a copy.
At the time, Murray told the press it was myopic to emphasize medical and pharmaceutical applications in reporting on genetic engineering. He said he is convinced that ''the dollar value of the agribusiness applications . . . could be 10 times that of medical applications. Agribusiness applications have a market potential of $50 billion to $100 billion, medical applications only $5 billion to $10 billion by the year 2000.''
Other prophets of agrigenetics are somewhat less sanguine.
''The potential tantalizes the imagination,'' says H. O. Kunkel, dean of the College of Agriculture at Texas A&M University. ''But,'' he adds, ''remember that our agricultural system has been evolving for hundreds of years and it will be a while before genetic technology can be incorporated.'' His colleague Page Morgan compares current speculation about genetically engineered crop plants to thinking about going to the moon 20 years ago. ''The engineers said they had the capability, but at the time they hadn't developed the equipment to make the trip possible. That's about where we are today (with genetic engineering on the farm) ,'' he says.
As was the case when President Kennedy set the national goal of putting astronauts on the moon, some critics are downright skeptical. Thomas N. Urban, president of Pioneer Hi-Bred, a major seed supplier, has said: ''Genetic engineering techniques cannot simultaneously work with large numbers of genes, which is a prerequisite for most hybrid and variety improvements. Plants have some 10,000 genes, and very few of their characteristics are controlled by a single gene. The new techniques will be helpful in speeding up our work. . . . But they won't change conventional breeding methods.''
Somewhere along this spectrum, between the extremes of euphoria and doubt, lies a realistic prospect for at least a minor agricultural revolution. But to what extent it will be propelled by the exotic techniques of genetic engineering rather than by more conventional breeding and how fast it will develop are at present unclear.
''The complexity of plants and animals presents a greater challenge to advances in applied genetics than that posed by microorganisms,'' says the congressional Office of Technology Assessment (OTA) in a major study of the new genetics released last April. ''Nevertheless,'' it adds, ''the successful genetic manipulation of microbes has encouraged researchers in the agricultural sciences. The new tools will be used to complement, but not replace, the well-established practices of plant and animal breeding.''
Winston J. Brill of the University of Wisconsin at Madison, who has been a leader in research to give grain crops such as corn the ability to fix nitrogen, is equally cautious. Reviewing the prospects for the new genetics recently in Scientific American, he notes that, while genetic engineering may eventually lead to crop improvements or entirely new crops, so far ''its most important effect has probably been on industrial laboratories, which have been alerted to the possibility of applying biological methods to agriculture.'' He adds, ''Perhaps the most significant indicator of the revolutionary nature of the work is that almost all these laboratories have been established within the past two years.''
Indeed, Brill has enough faith in that ''revolutionary'' work to take on a second job as scientific director for the Madison laboratory of Cetus Corporation, a major genetic engineering company. Brill will split his time between industry and university, giving up formal teaching but, he says, also giving priority to maintaining contact with students.
A hint of the possible long-term payoff of this arrangement is given by the investment firm L. E. Rothschild, Unterberg, Towbin in a recent report on Cetus. It says: ''Cetus Plant Genetics is one of the new venture subsidaries . . . it is being built around Dr. Winston Brill.'' Noting that ''the advanced techniques of genetic engineering are just being introduced to the agricultural field,'' the report says, ''nitrogen fixation, disease resistance, extended plant tolerance to adverse temperature and soil conditions, and new plant varieties are logical areas of research.''
The work on nitrogen fixation (NIF) of Brill and others illustrates the kind of challenge agrigenetics involves. If plants could pick up nitrogen from the air, it would reduce the need for chemical nitrogen fertilizer. One way to give corn, for example, the NIF chararcteristic, is to induce microbes that have the capacity, either naturally or artifically, to cuddle up to corn the way they do to legumes. But making microbes work in the field, if they don't already do it naturally, is far different from using them in the chemical or pharmaceutical industries where they are held in carefully controlled vats. In the field, they have to live rough and get along with their neighbors.
Brill calls this ''subtle question of the interaction of microorganisms with one another and with the biosphere as a whole'' one of the major stumbling blocks in agrigenetics. He recalls an early experiment in which he innoculated a soybean field with a strain of bacterium developed in the laboratory to have higher NIF ability. It failed to perform, unable to compete with the bacteria naturally present.
Brill and his coworkers have gone on to develop strains of bacteria that can work in the soil. He is trying to develop strains of NIF bacteria that will bind to corn plants. One such bacterium, Azotobacter vinelandii looks promising. Corn now grown in the US won't support this bacterium. Through selective breeding of varieties from other countries, Brill's associate, Stephen W. Ela, has begun to overcome this limitation. They now have corn that can get about 1 percent of its nitrogen from the air. Brill says they are ''sufficiently encouraged by our results to try to improve that percentage.''
Another way to give grain crops NIF capacity might be to introduce the necessary genes to the plants themselves. This is especially difficult. There are 17 different genes that have to be transferred - and made to work together. This involves overcoming a major biological barrier. Bacteria are what are known as ''prokaryotes,'' - their cells have no nuclei. All higher organisms, from yeasts to humans, are ''eukaryotes'' - having cells with nuclei. While single genes, including human genes, have been transferred from eukaryotes to bacteria and made to work, it is uncertain that a complex of bacterial genes could be transferred across this great biological gap and be made to work in eukaryotes.
Researchers at Cornell University, the Pasteur Institute in Paris, and the University of Paris have managed to transfer all 17 NIF genes from a bacterium to a yeast. But, as of this writing, that gene complex had yet to express itself. Commenting on this in a recent review of plant genetics, E. C. Cocking and colleagues at the University of Nottingham, England, called the possibility of these NIF genes being expressed ''remote.''
Even if the NIF genes were able to work in yeast this is no guarantee that they could be transferred to corn or wheat and be expressed in those plants. Again, if this were accomplished in the laboratory, there still would be the challenge of turning the laboratory tissue culture into a viable crop plant.
Broadly speaking, agricultural genetic engineers have two types of technology to work with. They can fuse cells of different species to produce hybrids with a mixture of genetic characteristics. Or they can insert copies of genes directly into the genetic instructions of a target species. That is how the ''sunbean,'' which made headlines last summer, was formed. It was an experiment in which a French bean protein gene was transferred to a sunflower.
One way to introduce such a foreign gene is to use a natural carrier. Geneticists have been studying the bacterium Agrobacterium tumefaciens as such an agent. It is able to infect plant cells. Biologists at the Max Planck Institute in Cologne, West Germany, last year reported using A. tumefaciens to implant specific genes. But they couldn't propagate whole new plants carrying the gene. Last spring, Marc Van Montagu of the Free University of Brussels announced that his laboratory had cleared that hurdle. Whole new plants carrying an inserted gene were grown.
Then on June 29, US Secretary of Agriculture John R. Block announced similar work being done by a Department of Agriculture-University of Wisconsin team led by USDA biologist John Kemp. They used A. tumefaciens to insert the gene that codes for a specific protein of the French bean into sunflower cells. However, noting that the payoff from such research would likely come in the next century, Secretary Block also reported that the researchers had only a mass of sunbean tissue culture. They had yet to induce this to form viable plants.
Turning plant cells into mature plants is an uncertain business. This is another major barrier to agrigenetic engineering. ''In microorganisms, the changes made on the cellular level are the goals of the manipulation. With crops , changes made on the cellular level are meaningless unless they can be reproduced in the entire plant,'' OTA explains. It adds, ''Therefore, unless single cells in culture can be grown to mature plants that have the new, desired characteristics - a procedure which, at this time, has had limited success - the benefits of genetic engineering will not be widespread.''
Such are the technical difficulties that make experts cautious in predicting the impact of genetic engineering on agriculture. In animal husbandry, there is little foreseeable benefit to be gained from the exotic new genetic techniques at this time, except for new pharmaceuticals for veterinarians.
Even with crop plants, where long-term possibilities are more evident, traditional selective breeding remains the key technology. Genetic engineers will have to cooperate closely with the plant breeders. ''The new technologies may provide potentially useful tools, but they must be used in combination with classical plant breeding techniques to be effective,'' says OTA.
That is why the International Plant Research Institute (IPRI), a genetic engineering firm, has gotten together with Davy McKee Corporation, a major breeder, to explore the commercial use of plants not now grown as crops and commercial crops that have been improved by genetic engineering. Among other projects, they will try to double the yield of the cassava (tapioca) plant without having to use fertilizers.
IPRI hopes to improve the plant's genetics, perhaps by inserting foreign genes for disease resistance, by improving protein content, or by genetically eliminating the toxic substances the plant produces and which now must be removed by special processing. Davy McKee, for its part, will design a processing plant to turn cassava starch into the sugar fructose or into ethanol.
In this venture geneticists, plant breeders, and process engineers will be working together to plan the best way to harvest some of the fruits of the new biotechnology.
Meanwhile, if classical plant breeding is so important, some biologists are concerned that its basis is being eroded as the spreading extinction of wild plants narrows Earth's genetic resources. Winston Brill points out that according to some estimates, the extinction rate is running as high as a thousand species a year. Such losses to the gene pool are not likely to be made up by genetic engineering. Moreover, mankind could be losing valuable plant traits of which it is unaware, as well as narrowing the genetic reserves to maintain the health of present crop species.
Not all experts agree with OTA's statement that ''it is uncertain how much genetic variation for improvement exists'' in crop species. Thomas Urban, for one, says, ''We have yet to discover any legitimate research data confirming the fears that genetic potential has plateaued.'' Nevertheless, a wide range of genetic resources are needed. OTA notes: ''A wild melon collected in India . . . was the source of resistance to powdery mildew and prevented the destruction of California melons. A seemingly useless wheat strain from Turkey . . . was the source of genetic resistance to stripe rust when it became a problem in the Pacific Northwest. Similarly, a Peruvian species contributed 'ripe rot' resistance to American pepper plants, while a Korean cucumber strain provided high-yield production of hybrid cucumber seed for US farmers.'' Genetic engineering is not yet ready to match the natural wealth of genetic diversity, and thus, to meet such challenges.
In spite of some efforts to build seed banks, such as the US National Seed Storage Laboratory at Fort Collins, Colo., experts believe that too little is being done to preserve even the gene pool of established crop plants.
Genetic engineering has opened the prospect of a dramatic new approach to crop improvement. Yet as OTA points out, Earth's ''lost genetic diversity is irreplaceable.'' This could turn out to be the biggest genetic challenge of all