The Fly Farm

by GRANT CANNON

1

THE breeding of plants and animals during the past century has been based on the hope that, through the mating of the best individuals, like would not only create like but do a little better than that, and so improve the race. And, in view of the high-producing cows, hens, pigs, and crops we have today, it must be admitted that careful selection has paid off—though improved care and feeding of the livestock and the use of high-analysis fertilizers on the crops have played a part.

Livestock men and plant breeders, however, have noticed that there is a limit to the improvement they can achieve by the old methods of selection. At some point in their program, usually around the tenth generation, the upswinging records begin to level off and finally reach a plateau. Some geneticists felt that at this point we were reaching the upper limit of the species. But along came the creators of hybrid corn and showed that there was at least one more production boost which could be squeezed out by approaching the genetic problem in an entirely different way.

Instead of breeding the highest-producing plants to get even higher production, G. H. Shull and E. M. East, who developed this new system, began a program of intensive inbreeding. And with corn, inbreeding can be very intense indeed for, with care, the pollen which the plant produces in the tassel can be collected and dusted over the silk that pushes out of the ear to fertilize the seed of the same plant. After a number of inbred lines had been developed through several years of selffertilization, these scientists wound up with just what the armchair philosophers had always predicted-spindly, anemic plants; some plants which showed lethal characteristics and died in the sprout; and plants which yielded gnarled, thinly seeded ears. But when one emaciated inbred corn plant was crossed with another unrelated but equally scrawny inbred, they produced seed with almost explosive vitality. This was the hybrid; and suddenly the word hybrid ceased to mean mongrel or wanton and became synonymous with productivity. Shull, in describing this burst of vitality, called it heterosis, but the more common name for the phenomenon is hybrid vigor.

It was soon discovered, however, that the seed which the hybrid itself grew so abundantly on its longer, heavier ears was fine for feeding hogs or making flour for corn pone, but was a dud when planted as a seed for a new crop. It seemed to have used up all of the hybrid vigor in its own growth.

For years the geneticists were tantalized with the knowledge that they could create this wonderful seed, but their parent stock, weakened by generations of the necessary inbreeding, could produce so little of it that it was not possible for commercial seedmen to grow and sell it at a price farmers could afford. At this point D. F. Jones discovered that two unrelated, high-producing hybrids could be mated, by detasseling one of them and allowing the other to fertilize it, and still retain the hybrid vigor in the seed produced. This double-cross hybrid, as it is called, is the seed which is planted in 98 per cent of our cornfields and has boosted our national corn yield by 20 per cent. Last year, that 20 per cent increase in corn production amounted to some 680 million bushels of corn — enough to feed out 29 million hogs to market weight.

Even though the scientists are not agreed on just what rearrangement of the thousands of genes in the chromosomes causes heterosis, breeders of plants and animals have been trying to produce hybrid pigs and chickens to eat the hybrid corn, and hybrid bees to suck the nectar from hybrid clover.

Perhaps the most astounding discovery made by these investigators was that a virgin queen bee could be made to lay fertile eggs simply by giving her a whiff of carbon dioxide gas. With this technique, the bee, like the self-fertilizing corn, could be highly inbred over a relatively few generations, but with most plants and animals it was soon found that years and years of painstaking care were necessary to create closely inbred lines. A partial solution to this problem has been developed by the artificial insemination laboratories of the dairymen where bull semen, frozen and stored at 110 degrees below zero Fahrenheit, can be held indefinitely. This is a rather expensive process which has been worked out only with cattle and still leaves the geneticist with the problem of the female line, which can only be accomplished by maintaining the living animals.

But while the cost and the problems and the fine results were all too apparent, geneticists were far from sure that the techniques used by the hybrid corn growers were the best methods for achieving heterosis.

2

IT WAS at this point that it was decided to set up the pilot plant for testing breeding theories. For livestock, the geneticists chose their old friend the common fruit fly — Drosophila melanogaster — the tiny insect which has probably contributed more to our genetic knowledge than any other living thing since Gregor Mendel planted his wrinkled peas, and which is capable of producing more generations in a single year, twenty-five or more, than a cattle breeder will see in his lifetime. In addition to the ability to produce a new generation every two weeks, the fruit fly has the advantage of small size and appetite. Thousands of families of Drosophila can be kept in their breeding pens, quarter-pint milk bottles, in an eight by ten foot laboratory which is about the size of a single breeding cage for chickens; and one stalk of bananas provides enough food for several months.

The sponsor of this unusual project, Purdue University, with financial support from the Rockefeller Foundation, selected Dr. D. C. Warren to direct the work, assisted by Dr. A. E. Bell, who was loaned by Purdue, and Dr. C. H. Moore, who, like Dr. Warren, is an employee of the Bureau of Animal Industry of the U.S. Department of Agriculture.

These three scientists set up their tiny, closetlike breeding laboratory in the basement of the Poultry Building on the Purdue campus, and installed equipment in the room to create as near as possible an absolutely controlled environment. Air conditioning, heating, and humidity systems keep the fly farm at a constant 74 degrees with a 50 per cent humidity; and fluorescent lights keep the room brightly lit for 12 hours a day. In an outer room, they set up a battery of low-powered binocular microscopes for measuring and counting the fruit fly eggs and for examining the livestock. Recently they have added to their equipment a microscales so sensitive that it is capable of recording the weight of a single fly (35,000 of them weigh an ounce) or even the egg of a fly.

Every few months, a stalk of bananas is brought to the laboratory; the fruit is peeled, ground up, mixed with acetic acid (to give it the fine overripe smell the flies love), and blended with enough charcoal to blacken the whole mess. The mixture is then put in the deep freeze to supply uniform food for generation after generation of flies. When a mating is made, a virgin female and a male are selected and put into a bottle. On the bottle cap is a gob of the black banana paste. Within 12 to 24 hours the female begins laying her eggs on the food. Each day the laboratory assistants remove the bottle cap, replace it with another blackdabbed cap, and count and measure the white eggs against the black food. To date, the highest producer laid 197 eggs in a single day —the lifetime record is 3500. The flies usually live 40 days, though one old female died a grent-great-greatgreat-great-grandmother at 86 days of age.

As the eggs hatch into larvae and the larvae go into the pupal stage, the scientists evaluate the results of the matings and decide on the next crosses to be made. They insure that they are starting with virgin females (since the female fly need mate but once to fertilize her lifetime production of eggs) by clearing the bottles of all the flies which have broken out of the pupal cases and then taking their breeders from those which next emerge. These young flies are anesthetized with ether, and the dark-bodied males separated from the females. When the female fly wakes from her drugged sleep she finds herself in a new bottle with her groggy mate, and the cycle begins again. Her life is one of monotonous productivity. Though her eggs are carefully counted, the record she makes, even if it is an outstanding one, is considered to be a test of the mating of her parents rather than an indication that her young should be chosen to carry on the race.

“The purpose of this project is not to find breeding systems which will produce outstanding individuals,” Dr. Warren has said. “We are interested in genetic methods which can be used to improve entire populations of plants or animals.”

In developing a high-producing population, the parent stock is judged from an entirely different point of view than in usual livestock breeding. In a way it is a division of labor; the parents produce the young and the young produce the crop, but don’t go on to become parents of the next generation of young. Different systems are used to breed the parent stock than are used for the creation of the high-producing generation. This approach would seem like madness to cattlemen who have paid as high as $200,000 for a champion beef bull in the hope that he will sire future champions; but the hybrid corn breeder, whose inbred parent plants seem hardly worth reproducing, understands perfectly.

3

So FAR, the genetic pilot plant has worked on four different systems of breeding. All of these systems are being tested at the same time with different sets of flies. In each system the results have been measured in terms of the size of the eggs laid by the female flies and the number of eggs laid. This has been a limiting factor in the work, since only the performance of the female could be tested directly. With the new microscales the body weight of the offspring is being recorded and a performance test on both males and females will be possible.

The first of the four breeding systems is the same as is used by practically every livestock breeder today. The daughters of the highestproducing female flies are bred to the sons of other top egg-layers. This has resulted in a general increase in production and then a leveling out on a higher plane, just as it has with poultry, cattle, and plants when applied to them. The next step is to try to determine why this happens.

In tackling this problem the scientists are trying to solve one of the great mysteries of genetics. They feel that the answer lies in finding out what arrangement of the genes causes the leveling out of production. But to find their answer they must account for thousands of genes in each chromosome — and each mating produces a recombination of these genes in the offspring. Their job is a little like breaking two strands of tiny beads into a box, shaking them up, and then trying to figure out how each strand was originally threaded — and doing it all in the dark.

One of the tools they are using to help them to find the answer is a man-made fly whose chromosomes have been bombarded with X rays. The bombardment knocks out genes, rearranges their order so that in a mating there is no matching of corresponding genes. The effect of mating a test fly with this man-made fly is similar to that of putting the chromosomes of the tested fly through a sieve. The geneticists can study the flies resulting from this cross with the knowledge that certain of their genes come only from the fly being tested. Since the gene arrangement of the man-made fly is well known, the geneticists know which of the genes in the offspring come only from the tested fly.

With a second group of flies, Dr. Warren and his associates are testing new breeding techniques in producing single-cross hybrids. Generations of brother-sister matings have produced a number of inbred lines. When two of these inbred lines are crossed, they produce a hybrid which shows the same bounce in vitality that makes the fastgrowing, high-yielding corn and the heavy-laying chicken hybrids so valuable to farmers. But where corn and chicken breeders select only the highestproducing inbreds to cross, Warren, Bell, and Moore are checking to see if this system, which seems so logical, is actually the best one to follow. They divide each inbred line into three equal groups based on their egg-laying ability. They then make all of the possible crosses between high medium, and low producers and check the matings by the production records of their offspring. Their curious results have added one more question to the unanswered riddle of heterosis, for they have found that when they cross high producers with medium and even low producers, the offspring often lay more eggs than do those resulting from a mating of two high producers. These unexpected results are causing corn and poultry breeders to wonder if they may not be throwing away valuable breeding stock.

The third experiment, devised by F. H. Hull of the University of Florida, is a radical departure from the breeding systems commonly used today. In this test, the scientists are making use of the fact that the female fly can be a monogamistnot just by inclination, but by her whole physical mechanism. One mating can fertilize all of the eggs she will ever lay. The male, on the other hand, has no such physical restrictions. He can be, and usually tries to be, a polygamist.

In trying to develop a rugged openbred population that will release hybrid vigor in the young when crossed with the frailer inbred line, the workers at Purdue think of the openbred male as their key and the inbred female as the lock. For this experiment each of 42 males is mated with four females to discover how vigorous and productive his children will be. In order to build better keys, he is then taken from his first wives and put in another bottle with four unrelated females from his own openbred population. Here he spends his declining days producing new families. At this point, the males have produced 168 families with their inbred mates and an equal number with their second set of wives.

When all of the first group of young have been tested for egg-laying ability, egg size, and body weight, the results show which of the males were the best keys for unlocking hybrid vigor. The scientists then pick six sons from the openbred families of each of the best seven fathers. The other 35 families are given the flit gun treatment. These 42 males are mated to another swarm of inbred females and the process is repeated.

Generation after generation of this breeding system has created two parent groups which are producing young with a remarkable amount of hybrid vigor. And one of the parents is still a rugged, openbred individual.

The fourth experiment is testing the theoretical breeding system suggested by R. F. Comstock of the University of North Carolina. This system is the most exciting of all, for it attempts to create vigorous hybrids without using inbred parents on cither side.

The pilot plant is employing the same techniques they developed for the third system — 42 males are mated to 168 females of the other parent stock and mated again to 168 females of their own stock. This time, however, all of the parent families are rugged openbred groups.

Since both parent groups are being modified by selection, the key and lock analogy doesn’t apply here. This is more like fashioning two gears that will mesh. Or like loading a pair of dice so that they will always roll a 5 and a 2.

Because both sides of the cross are being tested, they pick their males from one group and their females from the other in the first crossing and in the next generation pick the females from the first group and the males from the second.

The results have been spectacular. By selecting parents in each line on the basis of their combinability with the other line, and completely disregarding the ability of the parents to set records of production, they are creating a second generation that can out-lay or out-grow anything they have ever been able to produce by mating the best with the best. In other words, they are getting hybrid vigor from a mating of two openbred lines.

4

To THE farmer, this breeding system could be far more revolutionary than was hybrid corn’s. Hybrid poultry breeders could give up the expensive tedious process of working with inbred lines. Cattle and hog breeders could begin working toward combinability of lines, instead of depending on individual production records. Plant breeders as well as livestock men could produce the fabulous hybrid at greatly reduced cost — if the promising results which the pilot plant is getting with the fruit fly can be applied to other things.

Warren, Bell, and Moore know that most genetic principles apply to all living things, to moss and pine trees, to ants and giraffes; but, being scientists, they do not simply assume that their results will be universally applicable. In the beginning, they were afraid that if they started with flies which had been bred for generations in laboratories they might from the very start be working with livestock which had become a special case even among flies. Their first flies were collected from eight different college laboratories and mated at random. They haunted grocery stores and fruit stands, collecting wild specimens flitting around the ripe fruit, and added these to the collections. When, at this point, the three of them were invited to attend an international conference of geneticists in Canada, they took along their banana-baited milk bottles and made a cross-country collection of wild flies. With these wild flies added to their flock, they feel sure that their results have not been brought about by any special quality in their original stock. Whether these findings are applicable to other livestock, however, is a different problem.

Each of the breeding systems is being duplicated with chickens or pigs at the various experiment stations scattered around the country. But while the flies have produced over a hundred generations, the swine and poultry have produced five, which is too few to show any important results. “We have always felt very reluctant to release our findings as recommendations until we could test the experiments on at least one other type of organism,” Dr. Warren said. “In keeping with the rapid, pilot plant idea, that second organism should also be able to produce many generations every year. I think we have found our tester in the flour beetle, which produces twelve generations a year.”

Now, in a separate small laboratory down the hall, thousands of beetles burrow through the flour in their breeding-pen bottles, mate, reproduce, and are studied as carefully as are the flies. After each mating the flour is sifted through a large mesh screen to capture the adults, then through a smaller screen to sift out the pupae, and finally through a silk-like screen to capture the eggs. The results of each of the same four mating systems are analyzed, new crosses are made, and the breeding systems go on toward a more critical check.

There is a slight possibility that all of this work may only produce a fruit fly which is a hardier, more irritating pest than the one that now drowns itself in every open vinegar bottle. On the other hand, scientists may come up with genetic principles which will revolutionize the farmer’s job of feeding the world. After all, Mendel’s peas revealed hereditary laws which are just as applicable to the elephant as they are to quack grass or the kangaroo. And, curiously enough, the heritability of the size of the egg is the same in the fly as it is in the chicken, just as the numbers of eggs laid are strongly influenced by environment in each of them.

They may even discover the secret of heterosis. “I can’t say that we will be fortunate enough to uncover the answer to the problem of hybrid vigor,'’ Dr. Bell has said, “but I do feel that the solution to this puzzle will only come by hitting at it from many different angles as we are doing.”

At least they have had enough generations to study. In terms of human life the number of generations they have analyzed would stretch from Aristotle to Einstein.