Composition of Corn in the United States, 1946-1947 (1953)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

COMPOSITION OF CORN IN THE UNITED STATES 1946-1947* INTRODUCTION The chemical composition of crops varies in relation to a number of factors, many of which are not clearly understood. Climate, species or variety, soil type, management prac- tices including fertilization, and the storage or processing of the crop comprise the most im- portant considerations. In launching a survey of the nutrient contents of corn in the United States, there were limitations to the number of factors that could be studied. For example, it was unpractical in a nation-wide survey to attempt a detailed study of all of the varieties of corn, because several hundred varieties may be grown in any one state. The decision was made to ignore variety in the selection of samples but to record it at the time of sam- pling. Similar problems were encountered with regard to soil type and management practices, but no attempt was made to record data on these factors. The number of samples collected in this survey was limited to the facilities that were available for analyses. It was estimated for 1946 that a maximum of 200 samples could be processed through the commercial, institu- tional, and state feed control chemical labora- tories that might cooperate in the study. It was judged that this number of samples would give an average for the nation accurate to 1 per cent for proximate nutrients and certain min- eral elements and 3 per cent for most of the micro-elements and vitamins. PROCEDURE Selection of Samples The sampling plan was based primarily on climatic regions and on corn production data. * Initial funds for this work were provided by Swift and Company. Later grants were made by the American Feed Manufacturers Association, the Gould Research Foundation, Merck and Company, Inc. and the Ralston Purina Company. The United States was divided into ten regions. These were so delineated that the climatic conditions of rainfall and temperature were somewhat similar throughout each region, and the effect of these factors on composition was assumed to be fairly uniform. A map showing the regions is given in Figure I. The percentage of the nation's total corn production contrib- uted by each region, state, and county was computed and used as a basis for allocating samples. Within a state the counties were stratified according to production. The number of strata was equal to the number of samples to be taken in the state and the total produc- tion in all strata was equal. One county was then selected at random from each stratum. Two hundred counties were designated to be sampled in each year, and one farm was sam- pled in each county in 1946. In 1947 an extra farm was designated in each of 50 counties, dis- tributed among states and regions at a rate proportional to production. Sixteen states were not granted any samples because they did not produce enough corn. Two states that were allotted samples did not cooperate. In regions 9 and 10, no single state could be selected for a sample, but one state was chosen at random to represent each of these regions. Within states, the county corn production was the basis for allotting the samples. Where one county alone did not rate a sample by virtue of its own pro- duction, a county was chosen at random from a group of counties with sufficient corn produc- tion. Selection of the sampling site within a county: County agents and extension personnel in the various states were enlisted to procure the samples. In those states from which five sam- ples or less were taken, the Extension Agrono- mist was requested to collect all samples. In the more heavily sampled states, however, the help of the County Agents was enlisted. In the selection of actual farms in the coun-

COMPOSITION OF CORN IN THE UNITED STATES JB i M I ? i

COMPOSITION OF CORN IN THE UNITED STATES lies listed, the Extension Agronomist solicited the aid of a statistician. In sections where statisticians were not available, the following plan was executed. The map of the county to be sampled was divided approximately in half by a single straight line. By the toss of a coin it was decided which of the two areas should be further divided. The second area chosen was divided approximately in half by a third line perpendicular to the second, and a choice made again by tossing a coin. This procedure was re- peated eight times. Instead of dividing the area chosen on the eighth toss, however, the center point of that area was taken to locate the farm or delivery point to be sampled as indicated under (a) or (b) below. (a) The sample was taken from the farm which was nearest to the center point of the area designated by the eighth coin toss. (b) Instead of being taken from the farm nearest the center point prescribed under (a), the sample might be taken at the elevator, grinding mill, or any such point and in the same county. When this was done, the sample was taken from the first load of corn brought to the delivery point and arriving after a cer- tain hour set by the sampler on a certain day. It was stressed that, regardless of the method used, the sampler should place no restrictions on the sample other than (a) that it be unground corn, and (b) that it be of corn harvested (or to be har- vested) as grain for feeding purposes. The variety, color, type, or condition of the corn should in no way determine whether or not a particular field, crib, or load be sampled. The sampler was warned against making any discrimination against what might be deemed "nonrepresentative" samples. If the corn satis- fied the two simple restrictions given above, it should be sampled. The sampler was encour- aged to send any additional descriptive infor- mation about the sample along with the sample when forwarding it to Washington, B.C. Sample size: It was decided that five to ten pounds of shelled corn or its equivalent in ear corn would be adequate for the final sample. If ear corn samples were taken, it was suggested that they should be shelled before sending them to Washington. Regardless of the size of sam- ple, it was to be taken from the field, crib, or load in 15 to 20 approximately equal portions from well-distributed points in the field, crib, or load. If the final sample so taken was too unwieldy, it was to be mixed and quartered down to desirable size. The year 1947 was an abnormal one for corn, and a large proportion of the harvested crop was soft. Since the corn harvested in 1946 was predominately well matured, the 1947 sampling was set up in such a way that an accurate con- trast of the normal and abnormal crops might be obtained. It was decided to sample the same 178 farms that provided samples in 1946. In order that the data might be more valu- able in certain respects, it was planned to sample additional counties in certain states. Selection and sampling of the additional coun- ties were the same as described above. Selection of sample at the farm: In 1947 it was anticipated that there would be wide varia- tion, even within a field, in maturity of corn at frost date. Many farms might have only soft corn. It was believed that some corn might not be cribbed at all, being fed from the field in- stead. Since it was desired to sample the corn crop that would be fed, it was important that a representative sample be obtained, not a sam- ple of mature corn only. In view of this, it was highly desirable that samples be taken from the field. As far as possible, the time of sampling was between the date of the first killing frost and the husking date. A large enough sample was taken to yield 5 pounds of shelled corn. It was desired that the selection of the sam- ple on the farm be as objective as possible. Directions for sampling the field were there- fore given as follows: "Proceed to the southeast corner of the corn field and walk west along the south side a dis- tance of 10 stalks or 10 hills. Walk north into the field 10 stalks or 10 hills, taking all ears

COMPOSITION OF CORN IN THE UNITED STATES from the 10th stalk or 10th hill. Now walk north 5 more stalks or hills and take all ears from the corresponding stalk or hill. Continue the sampling of every 5th stalk or hill, as indi- cated by the pattern in the diagram shown below. Proceed along the path indicated until enough ears have been collected to yield 5 pounds of shelled corn." N W X X— — X X X 1 1 1 1 X X X X X 1 1 1 1 X X X X X 1 1 1 1 X X X X > < Field Diagram In the event the corn raised on the farm des- ignated for sampling had already been stored, the sample was taken from the cribs. In sam- pling from the cribs, an attempt was made to choose the sample in such a manner that the various grades of corn harvested were properly represented. For example, if the farmer's crop was 30 per cent soft corn and 70 per cent ma- ture, then 30 per cent of the sample was taken from soft-corn cribs and 70 per cent from mature cribs. In the sampling operation, no restriction was made on grade, color, variety, type, or condition of corn. For the two years combined, only 20 samples were white corn, and in 1946 very few samples graded other than U.S. No. 2. Data Secured at Time of Sampling Date Name of owner and address (including county) of the farm from which the corn came Type of corn: white or yellow flint or dent hybrid or open pollinated Name of variety or hybrid Stage of maturity (mature, soft, chaffy, etc.) Yield per acre: the 1946 yield for field sampled in 1946 the estimated 1947 yield for field sam- pled in 1947 Name and position of sampler Chemical Analyses Following collection, samples were shipped to the National Research Council offices in Washington, D.C., where they were dried and subsampled for vitamin, mineral, and proxi- mate analyses. These subsamples were then shipped to the cooperating chemical labora- tories. No attempt will be made to describe the methods of chemical analysis which were used. In most instances, they were standard accepted methods. Sufficient duplicate samples were sent out to different laboratories to verify the results obtained. The proximate analyses and calcium and phosphorus analyses are be- lieved, on the basis of duplicate samples, to be very reliable. There has been no reason to question the other mineral and vitamin analy- ses, although in some instances, as, for example, the sodium determinations, the methods used lacked sensitivity. For all practical purposes, however, the analyses were satisfactory. With copper, the results for the two years were in disagreement with regard to both mean and variability. It is believed, however, that the discrepancies in the copper data are of no prac- tical concern. Two hundred samples were requested and 178 received in 1946. In 1947, 251 samples were requested and 212 received. Proximate chemi- cal composition data were obtained for 169 samples in 1946 and 197 samples in 1947. The distribution of the missing samples was such that the original allocation distribution was not greatly disturbed. Analyses were obtained for proximate nutrients on almost all samples re- ceived and for calcium and phosphorus on the majority of the samples. The number of sam-