Chapter 31

EXPERIMENT STATION WORK

Our entire scheme of agricultural education and experimentation is new. The poultry work at experiment stations is very new. Ten years will about cover everything worthy of a permanent record in the poultry experiment station files.

Stations Leading in Poultry Work.

Among the earliest stations to begin poultry work in this country were Rhode Island, Massachusetts, Connecticut and Maine. Rhode Island conducted the first school of poultry culture. The two stations of New York State were also early in the work, and Cornell now has the leading school of poultry culture in this country.

West Virginia has always maintained a considerable poultry plant. Outside of the states east of the Appalachians, the first poultry work to be heard of was that of Prof. Dryden at the Experiment Station of Utah. Prof. Dryden's work was of a demonstrative nature. His early bulletins were forceful and well illustrated, and did much to call attention to poultry work.

In all this early work the great Mississippi Valley, where four-fifths of the nation's poultry is produced, entirely ignored the hen. The writer began his work with poultry at the Kansas Station in 1902, but his chickens were housed in a discarded hog house, and no funds being available, little was accomplished. In the last three or four years these experiment stations are rapidly falling into line and a number of poultry bulletins have recently been issued from these younger schools.

A few of the early landmarks in experiment station work was as follows:

The Utah Station clearly found that hens laid about 65 per cent. as many eggs in the second as in the first year, and that to keep hens for egg production beyond the second year, was unprofitable.

Massachusetts proved that corn was a better food for layers than wheat, and that the prejudice against it was founded on a misapplied theory.

The New York Station at Geneva demonstrated that poultry generally, and ducks in particular, are not vegetarians, and must have meat to thrive and that vegetable protein will not make good the deficiency.

The Maine Station was chiefly instrumental in introducing trap-nests, curtain front houses and dry feeding. The breeding work at Maine will be discussed at length in the last section of this chapter.

The United States Department of Agriculture did not take up poultry work until 1906. The publications issued by the department before that time were written by outsiders and printed by the Government.

The following is the list of the addresses of the experiment stations who have taken a leading interest in poultry work. It is not worth while giving a list of poultry bulletins, as many of them are out of print and can only be consulted in a library.

Maine--Orono. Mass.--Amherst. Conn.--Storrs. Rhode Is.--Kingston. New York--Ithaca. New York--Geneva. Maryland--College Park. West. Va.--Morgantown. Iowa--Ames. Kansas--Manhattan. Utah--Logan. Calif.--Berkeley. Oregon--Corvalis. U.S. Gov.--Washington, D.C. Ontario--Guelph (Canada).

Many foreign governments have us out-distanced in the encouragement of the poultry industry. Our Canadian neighbors have done much more practical work in getting out among the farmers and improving the stock and methods along commercial lines. As a result the Canadians have built up a nice British trade with which we have thus far not been able to compete. The work by the Ontario Station on the subject of incubation is discussed in the Chapter on Incubation.

Australia, like Canada, has given much practical assistance in marketing the poultry products, the government maintaining packing stations, where the poultry is packed for export. The Australian laying contests are quoted in the present volume. They outclass anything else in the world along that line.

In England, Ireland and especially in Denmark, the government, or societies encouraged by the Government, have done a great deal to develop the poultry industry. Depots for marketing and grading are maintained and the stock of the farmers is improved by fowls from the government breeding farms.

The Story of the "Big Coon."

With apologies to Joel Chandler Harris, I will tell a little story.

Uncle Remus was telling the little boy about the "big coon." It seems that the "big coon" had been seen on numerous occasions, but all efforts at his capture had failed. One night they saw the "big coon" up in the 'simmon tree, in the middle of the ten-acre lot. All hands and the dogs were summoned. To be sure of bagging the game, the tree was cut down. The dogs rushed in but there was no coon.

"But, Uncle Remus," said the little boy, "I thought you said you saw the big coon in the tree."

"Laws, chile," replied Uncle Remus, "doesn't youse know dat it am mighty easy for folks to see something dat ain't dar, when dey are lookin' fer it?"

When scientific experimenters entered the poultry field about fifteen years ago, they found it swarming with old ladies' notions. For everything a reason was given, but these reasons were derived from the kind of dreams where that which pleases the human mind is seized upon and search is made to find ideas to back it, not because it is true, but because it "listens good" to the dreamer. The first duty of the scientist was to banish these will-o'-the-wisp ideas that lead to no practical results.

For illustration Round eggs were supposed to hatch pullets and long ones cockerels. Eggs will not hatch if it thunders. Shipped eggs must be allowed to rest before hatching, the drug store was the universal source of relief when the chickens became sick, and red pepper and patent foods were the egg foods par excellence. These things, thanks to the scientist, are no longer believed or regarded by well read poultrymen, and instead his attention has been turned to matters having a more happy relation to his bank account.

In clearing away the useless popular notions, the scientists themselves have not been free from their influence, especially when they seemed to agree with accepted scientific theory. Many, indeed, are the 'coons in poultry science that have been seen because they were being looked for.

As a partial explanation it should be said that men available for scientific poultry work are very scarce. Poultry keepers schooled in the University of the Poultry Yard have no conception of scientific methods, and would explain experimental results by a theory that would fail to fit elsewhere. The available scientists on the other hand are seldom poultrymen.

Among the first men to take up animal husbandry work of all kinds, were the veterinarians. For years the only poultry publications put out by the U.S. Government were by veterinarians. These dust covered volumes with their five color plates of the fifty-seven varieties of tapeworms, still rest on the shelves of public libraries, a monument to the time when the practical poultryman knew only things that weren't so, and the scientific poultryman knew only things that were useless.

The first general law that all experimenters should know and the ignorance of which has caused and still causes the waste of the major portion of experimental brains and money, we will call the "Law of Chance." Let the reader who is not familiar with such things take two pennies and toss them upon the table. They are both heads up. He tosses them again, one comes heads, the other tails. The third time repeats the second. The fourth both come tails. The law of chance says this is correct. Heads should appear 25 per cent., tails 25 per cent., and mixed 50 per cent. of the time. Now let the reader try this in a lot of twelve tosses. Does it prove the law? Try it again. Are all lots alike? Now pitch a hundred times, then pitch pennies all day. By night the law will be so near proven that the experimenter will be willing to concede its validity.

Now suppose the lots of twelve tosses, each were lots of twelve hens, one Plymouth Rocks, the other Wyandottes, or one fed corn and the other wheat. The law of chance clearly proves that the larger number of unites, the nearer the theoretical truths will be the experimental results. Note, however, that small lots may by chance be as near the truth as large lots.

In practice two grave errors are made: First, conclusions are drawn from small lots compared with each other; second, conclusions are drawn from large lots compared with small lots. In the first case both may be off; in the latter case the small one may be off. Examples of the first error are to be found in the scores of contradicting breed and feed tests, that were published in the early days of poultry research. The second error is exemplified in the Ontario experiments in incubation, to which reference has already been made.

Here is a further example of this error. From the fifth egg laying competition at the Hawkesbury Agricultural College in Australia, I copy the following:

No. of Hens. Variety. Ave. Egg Yield.

6 Cuckoo Leghorn 190.16 30 S.C. Brown Leghorn 177.00 138 S.C. White Leghorn 174.93 12 R.C. Brown Leghorn 173.50 12 R.C. White Leghorn 172.66 18 Buff Leghorn 160.55 6 Black Leghorn 138.33

The ranking of Cuckoo Leghorns as first is a chance happening due to the small number; likewise the Black Leghorns had a streak of bad luck and received lowest place. To one not familiar with such work, the real significance of the table is that the S.C.W. Leghorns did the best work. A totaling of all other varieties gives 84 fowls with an average egg production of 170.5, which bears out the conclusion. As these birds were all kept in pens of six, we would expect to find the highest single pen to be White Leghorns, because, when compared with all other Leghorns, they have both the highest average and the greatest number. This accords with the fact that as the highest single pen is found to be White Leghorns with an egg yield of 239 eggs.

The above illustrates another important phase of the laws of chance, which says that not only is the average likely to be nearer the theoretical average sought when the number is increased, but that the individual extremes will be more removed.

Important Experimental Results at the Illinois Station.

From an Illinois Experiment Station report, the following is quoted:

"The stock used was Barred Plymouth Rock pullets. These pullets were a very uniform Barred Rock stock that had been bred as an individual strain for many years. They were practically the same age, and except for the factors mentioned were treated as uniformly as possible.

First Year's Results.

No. Hens. Diet. Ave. Egg Yield.

10 Nitrogenous Diet 132.9 10 Carbonaceous Diet 128.4 10 Wet Wash 155.8 10 Dry Wash 111.4

"The results of the first test are somewhat surprising for it is generally believed that the nitrogenous diet is best for laying hens. The difference indicated in the first year's results was so light that it was decided to repeat the experiment the second year.

"As the wet wash is clearly proven to be superior, these hens were used the second year to compare meat meal with fresh cut bone.

Second Year's Result.

No. Hens. Diet. Ave. Egg Yield.

10 Nitrogenous 142.2 10 Carbonaceous 134.5

10 Meat Meal 102.2 10 Green Cut Bone 128.9

"The results of the second year clearly indicate the great superiority of green cut bone as compared with the dry unpalatable meat meal. The comparison of a highly nitrogenous ration with that of a ration consisting largely of corn, while showing the advantages of the nitrogenous rations, does not show the contrast expected.

"Some visiting poultrymen expressed the opinion that corn is a better poultry food than commonly supposed. Considering this fact and the great fundamental importance of the question at issue, it was decided to repeat the experiment a third year, and feed a large number of birds on each ration.

No. Hens. Diet. Ave. Egg Yield.

100 Nitrogenous 126.9 100 Carbonaceous 127.2"

I will leave the last without comment, for the whole thing is a hoax. The Illinois Experiment Station has never owned a chicken. These "Illinois" experiments were planned and executed in a few minutes of the writer's spare time. The basis of the experiments was a pack of cards containing the individual records of the Maine Experiment Station hens, shuffling the cards and averaging the desired number of records as they come in the pack, made the distinction between the various diets.

Experimental Bias.

Pet ideas consciously or unconsciously mold practice. A bias toward an idea may show itself in the planning and conducting of an experiment, or it may come out in the later interpretation.

An illustration of the first kind is found in the early work of the West Virginia Station (Bulletin 60). With the preconceived notion that hens should have a nitrogenous diet an experiment was planned and conducted as follows:

One lot of hens was fed corn, potatoes, oats and corn meal. A contrasted lot reveled in corn, potatoes, hominy feed, oat meal, corn meal and fresh cut bone. The results were in favor of the latter ration by a doubled egg yield.

To any experienced poultryman the reason is evident. The variety of the diet and the meat food are what made the showing.

About the same time the Massachusetts Station planned a similar experiment. The bias was the same, but it took a fairer form. The hens were both given a decent variety of food and some form of meat. The bulk of the grain was corn in the carbonaceous, and wheat in the nitrogenous ration. The results were in favor of the corn. This astonished the experimenter. He tried it again and again tests came out in favor of corn. At last the old theory was revoked, and the fallacy of wheat being essential to egg production was exploded. If by an irony of fate in the shuffling of the hens, the wheat pen had the first time showed an advantage, the experimenter might have been satisfied and the waste of feeding high priced feed when a better and a cheaper is at hand, might have gone on indefinitely.

Of bias in the interpretation of results all publications are more or less saturated. A reading of the Chapter on Incubation will illustrate this. A common error of this kind is the omission of facts necessary to fully explain results. Items of costs are invariably omitted or minimized. Food cost alone is usually mentioned in figuring experimenting station poultry profits, which statement will undoubtedly cause a sad smile to creep over the face of many a "has-been" poultryman.

The writer remembers an incident from his college days which illustrates the point in hand. Let it first be remarked that this was on the new lands of the trans-Missouri Country, where manure had no more commercial value than soil, and is freely given to those who will haul it away.

The professor at the blackboard had been figuring up handsome profits on a type of dairying towards which he wits very partial. The figures showed a goodly profit, but the biggest expense item--that of labor--was omitted. One of the students held up his hand and inquired after the labor bill.

"Oh," said the smiling professor, "The manure will pay for the labor."

When the class adjourned, the student remarked: "They say figures won't lie, but a liar will figure."

The third way in which experiments are made worthless is by the introduction of factors other than the one being tested. This may be done by chance, and the conductor not realize the presence of the other factor, or the varying factors may be introduced intentionally under the belief that they are negligible. Of the first case an instance may be cited of the placing of two flocks in a house, one end of which is damper than the other, the accidental introduction into one flock of a contagious disease, or one flock being thrown off feed by an excessive feed of greens, etc., etc. These factors that influence pens of birds greatly add to the error of the law of chance. In fact it amounts to the same thing on a larger scale. For this reason not only are many individuals, but many flocks, many locations, and many years needed to prove the superiority of the contrasted methods.

The criticisms in the following section will amply illustrate the case of foreign factors being unwisely introduced into an experiment.

The Egg Breeding Work at the Maine Station.

As is well known the Maine Station was for years considered by all poultrymen to be doing a great and beneficial work in breeding for increased egg production. Up until the fall of 1907, the poultrymen of the country were of the opinion that this work was in every way successful, and a large number of private breeders had taken up the use of trap-nests in an effort to build up the egg production of their fowls.

When early in 1908 Bulletin 157 of the Maine Experiment Station was published, it showed by averages as given in the table on page 202 that the egg yield at the station was for the entire period on the decline. In Bulletin 157, the statement was made that "arithmetical mistakes" and "faulty statistical methods" accounted for the discrepancies between the former publications and the criticised data. The further explanation that "the experiment was a success as an experiment," etc., only appeared to the public mind as a graceful way of explaining what was, to the practical man, an utter failure of the entire work.

The unfortunate death of Professor Gowell, together with the fact that he had equipped a private poultry farm with station stock, added to the confusion, and the result of the bulletin was the precipitation of a general "pow-wow" in which the poultry editors were about equally divided between those who were casting insinuations upon the personnel of the station, and those who decried the whole effort toward improving the egg yield.

After going over the publications of Professor Gowell, visiting the station and meeting the present force, I came to the following conclusions regarding the matter:

Professor Gowell's work is open to severe criticism. Errors have been made in conducting the work at Maine which have made it possible for a mathematical biologist to take the data and seemingly prove that selection, as practiced by Professor Gowell, actually resulted in lowering inherent egg capacity of the strain of Plymouth Rock hens under experimentation. Had Professor Gowell's successor been a practical poultryman, it is my candid opinion that the public would have been given a radically different explanation of the results.

Professor Gowell is the author of the following statement: "The small chicken grower is earnestly urged to use an incubator for hatching." This opinion is not in accord with that of the majority of breeders and the more progressive experiment station workers. The opinion has been expressed by Professor Graham and others, that the particular results at the Maine Station may have been due to the decrease of vitality caused by continued artificial hatching. This view may be wholly without foundation. Nevertheless, as the common type of incubator is under heavy criticism, and it is pretty well proven that chicks so hatched have not the vitality of naturally hatched chicks, surely a series of breeding experiments would carry more weight if the replenishing of the flock had been accomplished by natural means.

For the first few years of the breeding work the house used was the old-fashioned double walled and warmed pattern. The last few years of this work were conducted in curtain front houses. That the cool house is an improvement over the warm house is generally conceded, but there are many poultrymen who are still of the opinion that the warm house will give a larger egg yield, though at a greater expense and less profit.

In the early years of the work the method of feeding was also a time-honored one, and included a warm mash. About the middle of the experimental period Professor Gowell brought out the system of feeding dry mash from hoppers. This custom became a great fad and Professor Gowell and Director Woods have preached it far and wide. Perhaps it is an improvement, but it is to-day much more popular with novices than with established egg farms. Many old line poultrymen have tried dry mash only to go back to wet mash, by which method the hens can be induced to eat more which is conducive to high egg yields. Whether these changes in housing and feeding have been improvements as claimed by those who introduced them, or whether their popularity may be explained in part at least by the psychology of fads, is a point in question, but certainly the marring of a breeding experiment by introducing radical changes in the factors of production is at best unfortunate.

A much more serious criticism than any of the foregoing is to be found in a change of the size of flocks and amount of floor space per fowl. I have gone over carefully the published records of Professor Gowell, and the review of Dr. Pearl, and the following table represents, as near as I can determine, these factors for the series of years. In the year 1903 I find no clear statement as to the manner in which the birds were housed, and I may be in error in this case. Otherwise the table gives the facts.

Year Hens in Flock Per Hen Egg Yield 1900 20 8. sq. ft. 136.36 1901 20 8. sq. ft. 143.44 1902 20 8. sq. ft. 155.58 1903 20 8. sq. ft. 135.42 1904 50 4.4 sq. ft. 117.90 1905 50 4.4 sq. ft. 134.07 1906 50 4.4 sq. ft. 140.14 1907 50 4.4 sq. ft. 113.24

Certainly this oversight is a serious one, and one especially remarkable considering the fact that the comparison of different size flocks formed a prominent part of the Maine Station work during the last three years of the breeding test. The results of the work at the Maine Station on testing flock size, conducted without relation to the breeding work, gave the following results:

No. of Hens Sq. ft. per Hen Egg Yield 150 3.2 111.68 100 4.8 123.21 50 4.8 129.69

No comparisons of 50 and 20 bird flocks in the same year are available, but by extending the comparisons of the 50, 100 and 150 flocks into the 20 flock size, we can get some idea of the error that has been here introduced. The result of the Australian egg laying contest in which the flocks were composed of six hens, shows a yield of about one and one-half times as heavy as the Maine records, which certainly seems to substantiate the ideas here brought out.

It is a well established fact in poultry circles that many men who succeed with a few hundred hens, fail when the number is increased to as many thousands. When the breeding experiments under discussion were started, Professor Gowell had under his supervision about three hundred hens. When the work was closed the experiment station plant had been increased to four or five times its capacity, and Professor Gowell had a large private poultry plant of his own in addition.

It is interesting to note in this connection that the last four years of the records are explained by Professor Gowell as being low, due to various "accidents" (?) It is unreasonable to suppose the true explanation of these "accidents" would be found in connection with the increased responsibility and size of the plant.

The breeding stock sent out by Professor Gowell has given general satisfaction, and was found by Professor Graham of the Ontario Station, as well as by a number of private individuals, to be of superior laying quality to that of the average Barred Rock.

Clearly there is only one way to prove whether Professor Gowell's work has been a wasted effort, and that is for flocks of his strain to be tested at other experiment stations against birds of miscellaneous origin.

That much has been lost to the poultrymen of the country by the recent upheaval at the Maine Station, I believe to be the case, but that does not mean that the men now in charge will not in the future be of great value to the poultry interests. They are, however, in the class of pure scientists rather than applied scientists, but if let alone they will dig out something sooner or later which they or others can apply to the benefit of the industry.

Upon the whole, I think that the present case of the trap-nest method of increasing egg production stands very much as it is has always stood, being a commendable thing for small breeders who could afford the time, but not practical in a large way, except at experiment stations. On a large commercial scale the system of selecting sires by the collective work of his first year's offspring would probably get the quickest results.

The best use of the funds of the people in the promotion of agricultural industries is in the permanent endorsement on the one hand of a few high grade research stations where the deeper theories may be worked out, and on the other the teaching of such good principles and practices as are already known.

The greatest opportunity for Government effort lies in the development demonstration farm work in poultry Just as it is doing with the corn and cotton in the South.