will have to use inexhaustible supplies, which means using energy from the sun.

A national energy policy then, must be a two-time strategy here. We have to do something about the current shortages yet take our long term needs into account. I propose that we deal with the current shortages by cutting down on demand because among other things we are going to have to be thrifty in the future in order to keep our demand within the limits of the renewable resources we will have to develop in the future.

Senator HOLLINGS. I don't want to interrupt your train of thought, but I said you were practical.

How do you practically cut down on demand? How does the Government get people to use less energy?

Dr. COMMONER. I am going to come to that.

Senator HOLLINGS. We can't get this crowd around here to set a limit on spending, much less cut down on demand for a heating stove in a cold spot in the middle of the country.

Dr. COMMONER. Well, let me get right to that and answer your question.

I want to say before I go any further that the value of S. 70 is that it takes into account very clearly the importance of cutting down on demand.

Well, now, how do we do it? Some of my colleagues and I have gone through an exercise to begin to answer that question. We have asked to what degree would it be possible to cut down on the use of energy without loss of goods and services in various economic sectors?

The outcome of our exercise tells us that it is possible to cut down on the use of fuel by raising the energy efficiency with which the goods are produced. We have come up with some examples that illustrate what kind of reductions in energy consumption would be possible.

Let's first take the area of transportation. Since transportation is responsible for about 25 percent of our use of energy in the country, it is a big target. It is worth cutting into and the results are striking. In our exercise, the first question we ask is: How do you measure the social value of transportation? Can you measure it by the number of gallons of gasoline consumed? No, because we don't travel to burn gas; we travel to go from one place to another. And so we measure the effectiveness of transportation by the degree to which it can move people and freight. The quality of this measure is indifferent to the mode of transportation and the kind of fuel burned.

We now move about twice as much freight by railroad as we do by intercity truck in the United States. But the rate of increase in trucking is much faster. All the increases in freight are being taken up by trucks. In effect we are replacing railroad freight with truck freight.

Now, trucks use six times as much fuel to carry a ton over a mile as railroads do. Therefore, going on through the exercise, we say we want to move the same tonnage of freight that we do now, and ask how to do it at minimum fuel costs. The answer is perfectly straightforward: Take the trucks off the road and put the freight on the railways. In table II we have gone through that exercise numerically. Senator HOLLINGS. Incidentally, is that practical?

Dr. COMMONER. Well, the railroads are there. I think they could use the business. You might say it is impractical because you would

be putting truckdrivers out of business, but in a moment, I will show you that we will need them all to drive buses.

At any rate, the exercise that we ran through-which essentially means taking trucks off the road and putting the freight on railroads, cutting down on the use of automobiles, and using buses, for example, for most of the trips that automobiles make, which are about 5 miles, replacing airliners with trains for 200-mile trips, which would have the added benefit of being faster-lops off 48 percent of the energy used in transportation, or 12 percent of the national energy budgetwithout any reduction in the number of passengers carried, the number of passenger miles travelled, or the number of freight miles carried. As our freight and passenger traffic rises, instead of meeting it by spending more fuel in our current wasteful way, we could relieve the crisis by shifting to more efficient ways of moving people and freight.

Now, you may say that is impractical because of the economic consequences, but you have to ask yourself: Where is the highest priority? Is it more important to avoid economic shifts or more important to avoid running out of fuel entirely in the next generation?

Now, let me give you another example. Take space heating, which space heating accounts for about 18 percent of our energy consumption. In our exercise reducing space heating uses of energy, we found the quickest way to do it would be to ask everybody to lower their thermostats by 5 degrees; that costs nothing. I suppose some people might have to go out and buy a sweater, but ley me make this point. Around the turn of the century, a careful study was made of comfort in relation to heat, wherein it was discovered that the most acceptable temperafor a room was 60 degrees fahrenheit. Today, if you were to do the same survey, it would probably be 75 degrees fahrenheit.

Has the biological requirement of the average American shifted between 1900 and now? It is unlikely. Rather, we have raised temperatures to avoid the use of, for example, the very simple way of keeping warm, wearing clothes. We propose, for example, a drop from 75, not to 60, but to 65 degrees. If you do that and convert to solar space heating in those parts of the country where it is feasible, you can achieve a 6.4 percent reduction of the national energy budget.

If you want to be really practical, in those areas of the Nation where the oil companies tell us they can't supply enough oil and gas over a period of a couple of months, I would suggest a very careful public education program explaining that there is no need to use that much oil and gas and encouraging people to simply lower their thermostats and put on sweaters. In a sense that will get us out of our, shall we say, enslavement to the oil companies, and you will be able to look them in the eye and talk back to them.

Let me give you another example. The manufacturing industries use 43 percent of our energy. Now, this is very complicated realm for our exercise, but I do want to mention one computation we have made on a particular good, fiber textiles.

In the United States, in the last 25 years, we used about the same poundage of fiber now as we did in 1946 but the kinds of fibers we use have changed. We now use synthetics rather than cotton and wool. The shift from cotton and wool to synthetics has put a burden on energy supplies. Why? Because building a fiber, which is a complex molecule, takes energy. The cotton plant does it by getting energy from the sun, taking it up in the leaves during the process of photo

synthesis, and converting that energy into a cotton fiber. You can think of a cotton fiber as congealed solar energy. A biologist like myself finds it very amusing when great fuss is made over the need for developing new ways of capturing solar energy because the green plants are doing it all the time without the benefit of any R. & D. They just do it. And in fact, they yield for us food, fiber, and fuel, as congealed solar energy.

And so cotton, and also wool, since the sheep eat the grass, which is doing what any green plant does, are socially useful goods delivered to us at no energy cost whatsoever.

Now, let me correct that, because we have managed, foolishly, to pour energy into the growing of cotton, for example, by using pesticides and fertilizers which take energy to make, as well as gasoline and so on; and in Arizona we use a good deal of energy to pump the water around for irrigation.

Still, taking all that into account, in other words, taking the more than energy wasteful way of growing cotton, and comparing it with the amount of energy used to make nylon, you get a very interesting result. When you make nylon, you have to ask yourself: That molecule has got to be put together, too; where does the energy come from? That energy comes from burning petroleum or natural gas-depletable resources. What's more the raw materials for nylon come from our stores of fuel resources, whereas the cotton plant uses carbon dioxide which is free in the air to make the fiber, the nylon process uses gas and oil as raw material inputs. So we use up part of our fuel reserves to make the fiber, use up more to burn to get the energy to do the chemistry; and, of course, unlike the cotton plant, we are polluting the environment.

Our computations show that even the most energy wasteful way of growing cotton uses 50 percent less energy to produce a pound of fiber than the production of a synthetic. If we want to save energy by making our fiber production more efficient, we should shift back away from plastics over to cotton and wool, away from plastics over to such materials as paper and lumber.

(The following information was subsequently received for the record:) ENERGY COSTS OF COTTON, COMPARED TO OTHER NATURAL AND SYNTHETIC ORGANIC HOUSEHOLD MATERIALS

(By Michael Corr, Executive Secretary, AAAS Committee on Environmental Alterations, Box 1126, Washington University, St. Louis Mo.)

The energy required for the manufacture of various materials has been a persistent theme in discussions of the environmental consequences of materials. policy, recycling and solid waste disposal. This memorandum develops an energy price for cotton for comparison with other natural and synthetic materials illustrating the often heard point that in terms of energy, it is advantageous to produce materials through biological processes which take advantage of solar energy via photosynthesis.

Considering the high degree of mechanization in agriculture, Michael Perelman has demonstrated that U.S. agriculture uses two calories for each calorie it produces. (Energy Use in Agriculture, Environment, October, 1972).

A detailed analysis of energy use in upland cotton farming yields supportive results. For the production of a pound of cotton (including seed as well as lint) 31,137 Btu's resource energy is required for fuel and power, including indirect use through chemicals, but not including indirect use via equipment manufacturing or building construction or for ginning. Growing cotton yields about one-fourth of the resource energy invested in the process (not counting solar energy.) If one allocates energy consumed according to the market value of the products, it takes 26,789 Btu's to grow a pound of cotton lint, an energy yield of 27% of the fuel and power inputs required (see Table 1).

Energy use for agricultural chemicals in 1964 exceeded energy use for machines on the cotton farm (not including ginning.) After chemicals and fuel for equipment, irrigation was the next largest user of resource energy. These figures, however, were computed on the basis of a national average for cotton growing regions. By approximately weighting the national average energy costs in Table 1 by the ratio of dollar costs in Table 2, we obtain a comparison of Southern Arizona energy costs for producing cotton to the national average.

It takes only 69% as much labor and 90% as much power and equipment to grow cotton with Arizona technology. However, more chemicals and irrigation expenditures are required, and the Arizona farmers lost money in 1964, while the national average cotton farmers made money. In addition, the Arizona cotton farmer used 35% more energy per pound of lint produced.

He contributed both to the water shortage and to the power shortage in the Southwest, and made no contribution to their economy, except by circulating his mortgage. In addition, contrary to popular belief, in spite of the sharp Arizona winters, he none the less felt obliged to broadcast 81% more pesticides and fungicides than the average cotton farmer.

TABLE 1.-RESOURCE ENERGY USED FOR COTTON LINT PRODUCTION: 1964

Scientific American, 1970. "Input Output Chart of the U.S. Economy," New York.

Starbird, I. R., 1966. "Costs of Producing Upland Cotton in the U.S., 1964." Economic Research Service, U.S.D.A. and personal communications.

U.S. Department of Agriculture, 1967. Agricultural Statistics.

U.S. Department of Commerce, 1971. Census of manufactures, fuels and electricity consumed, MC67(S)-4.

1 The above figures are for 1964. By 1969, there was a shift towards the use of diesel from gasoline and L-P fuels, and energy for trucks, tractors and mechanical equipment dropped from 9,900 Btu/lb to 8,200 Btu lb of lint produced. (Personal Communication, Irving Starbird, Economic Research Service, U.S.D.A., to Dabney Wellford. National Cotton Council of America, Memphis, June 28, 1972.)

2 In 1964 irrigation costs were 0.713 cents per pound of lint on the average. (Personal Communication, Irving Starbird to Michael Corr, Aug. 16, 1971.) In 1964 the average farm price of electricity was 2.31 cents per Kwhr. Thus 0.713 cents per lb -2.31 cents/kwhr 0.38 kwhr/lb. 0.308 kwhr x 3412 Btu's/kwhr=1,050 Btu's/lb., or 3560 Btu's/lb lint for irrigation assuming the worst case of all-electric irrigation, with 29.5 percent efficiency in the generation and transmission of electric power. Note that this figure is in the nature of an average. For arid areas such as Pinal County, Arizona, electric powered well pumps demand 3.9 kwhrs delivered power, or 45,050 Btu's of rescurce energy per pound of cotton lint produced for irrigation alone (personal communication, John Hay (Tucson) to Michael Corr, Aug. 11, 1971). Thus, under these arid conditions, irrigation requires 13 times as much energy as in the average case.

In 196, non-irrigation direct power costs were 0.076 cents per pound of cotton lint (Starbird, August 1971) or 380 Btu's resource energy per pound of lint (see note 2).

4 Starbird's telephone service cost is 0.084 cents per pound of cotton lint produced. Costs for the communications industry include about 3% for energy, or in this case, 0.003 cents per pound of cotton lint produced. The 1962 all industry average price for energy was 16,400 Btu/cents. Thus the farmer is using 49 Btu's indirectly for telephone service in producing one pound of cotton lint (Scientific American, 1970. "Input Output Chart of the U.S. Economy," New York, and U.S. Departent of Commerce, June 1971, 1967 Census of Manufactures, Fuels and Electric Energy Consu ned, table 3. Fron I. R. Starbird and F. K. Hines, E.R.S., USDA, "Costs of Producing Upland Cotton in the United States," 1964, Agricultural Economics Report No 99, Septen.ber 1966, we find the following expenditures per bale of cotton:

Fertilizer.
Herbicide.

Insecticides and fungicides.

Defoliants.

Other chemicals.

Total materials.

Dollars

per bale

10.18

1.41

5.07

.89

27

17.82

Direct and indirect purchase costs of chemicals for production of agricultural products a nounts to about $130 per $1,000 worth of chemicals produced. Thus the production of $0.036 worth of agricultural che nicals requires

Since 0 47 cents in chemicals is bought f:o.n SIC 28 to produce one pound of cott on lint, for direct and indirect energy through chemicals we have a usage of

0.47 centsX27,500 BTU's/cents = 12,900 Btu's (RE)

Clearly, the environment sustains heavy damages for such marginal farming. Acknowledging the broad categories of environmental costs associated with cotton farming, cotton compares favorably with all plastics for which energy prices are known (see table 4). Paper does also, but might not if the production energy were amortized over the life of the material.

Note that the high density synthetic materials in table 2, such as nylon and polyethylene, require in the neighborhood of 54,000 Btu's per pound of production, or about twice as much as cotton (27,000, on the national average). While the energy cost of cotton in some regions of the Southeast, favored by damp weather, and lower consumption of agricultural chemicals is considerably lower (about 10 percent), conservatively using the national figure, we can compute the effects on national energy consumption of producing the fiber we needed in 1971 according to the mix of synthetics and naturals which existed in 1947 (see table 5). In 1947 the average energy usage for fiber production was about 31,450 Btu/lb., while by 1971 it had risen to 43,500 Btu/lb. an increase of 41 percent. If in 1971, we had produced the same poundage of fiber as in 1947, but had used the 1947 mix of fibers, in this sector alone, we would have experienced an energy saving of 128.6 X 1012 Btu, or 0.63 percent of national industrial energy consumption.

TABLE 2.-DOLLAR COSTS PER BALE OF COTTON LINT AND SEED: 1964

TABLE 3. RELATIVE ENERGY COSTS FOR PRODUCING A POUND OF COTTON LINT; NATIONAL AVERAGE, AND SOUTHERN ARIZONA

This energy saving, of course, is one of many that could be realized through using materials with low energy costs. For instance, if wood sashes were to completely replace metal sashes, a total energy saving of about 41% less energy expenditure for sashes is technically possible. Similarly, if hardwood flooring were substituted for ceramic, vinyl, and carpet (91% synthetic) an energy saving of 47% would be technically possible. 1

The import of such case studies is that it would be highly desirable to do careful studies of the energy, and more generally the environmental costs associated with the production of natural, synthetic, and metal products so that in applications for which different materials compete, the energy and environmental costs as well as the immediate market costs could be taken into consideration in any public policy which would affect the mix of materials produced by agriculture and the process industries. While in special applications exotic materials may be highly desirable, it should be remembered that the 1947 mix of technologies, if it were utilized to produce the 1967 GNP, would require 35% less energy per dollar of value added as was actually the case in 1967.2

1 Boksenbaum, Howard, 1971. Personal Communication, Cnter for the Biology of Natural Systems, Washington University, St. Louis, Mo.

2 Commoner, Barry and Michael Corr, 1971. Power Consumption and Human Welfare. Paper prepared for the Symposium on the Energy Crisis, AAAS Annual Meeting. Philadelphia, December 28.