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 change materially the composition of the nor- mal plant. Relatively small nation-wide sur- veys, such as those reported here, appear to be adequate for such periodic checks. It was hoped that the correlation study would show sufficiently significant relation- ships between the contents of various nutrients to be of practical use. A possible use envisaged was the estimation of the content of difficult- to-determine nutrients from the content of easily determined nutrients for samples on which complete analysis is not feasible. It ap- pears, however, that the correlations are in general too low to be of value for this purpose. SUMMARY In 1946, 169 samples, and in 1947, 197 sam- ples of corn were collected at harvest time from 30 states in 10 climatic regions in the United States. Chemical determinations were made of proximate nutrients, vitamins, and minerals. These data were used to calculate means, coefficients of variation, and other sta- tistical information. The variation between regions was signifi- cant (19 to 1) for fat in both years, and for a number of nutrients in one year but not the other. The large within-state variation for most nutrients casts doubt on the practical impor- tance of regional differences. Regional varia- tions in fat are possibly confounded with varietal and color differences. Color of the corn could not be significantly related to any constituent except carotene and fat. Between varieties of yellow corn no sig- nificant differences were found except for fat. No sufficiently high correlation was found between any nutrients to indicate a possible relationship useful in estimating the amounts of nutrients that are difficult to determine. The mean protein content of No. 2 corn as found in this study is lower than that reported by Morrison in 1936. No direct explanation for these lower values was found as a result of this study. ACKNOWLEDGMENT The Committee is indebted to Miss Hazel Orcutt, who kept records and compiled the data, and to Mr. M. L. Richards, who was re- sponsible for the statistical analyses. Great credit is due to the extension and research personnel of the colleges of agriculture, to the state feed control laboratories, and to the lab- oratories of the feed and associated industries whose help in collecting and analyzing samples made this survey possible. REFERENCES 1. Committee on Feed Composition, National Re- search Council. 1947. Composition of Feeds. Report No. I. 2. Committee on Feed Composition, National Re- search Council. 1947. Composition of Feeds. Supplement to Report No. 1. 3. Barley, E. B., and E. E. De Turk. 1948. Corn Pro- tein and Soil Fertility. What's New in the Pro- duction, Storage, and Utilization of Hybrid Seed Corn. American Seed Trade Assoc., Hybrid Corn Division, Annual Report 3: 84-95. 4. Morrison, F. B. 1936. Feeds and Feeding. 20th Ed., Morrison Publishing Co., Ithaca, New York. 1050 pp. Illus. 5. Schneider, B. II., H. L. Lucas and K. C. Beeson. 1953. The Nutrient Composition of Corn in the United States. Agricultural and Food Chemistry 1: 172-177,1953.
8 COMPOSITION OF CORN IN THE UNITED STATES small and often did not agree in sign. Of 434 correlation coefficients computed for the two years, only 34 equalled or exceeded 0.5. It may therefore be concluded from this investigation that the correlations between nutrients are not very useful, and a knowledge of the composi- tion of corn with respect to a few nutrients does not aid materially in assessing its composition hi other nutrients. Thus, in feeding practice one must either use average data from tables to apply to a given lot of corn or, if greater pre- cision is desired, determine the nutrient content for that lot. From the standpoint of sampling, it appears that each nutrient should be con- sidered independently. The negative correlations of nitrogen-free extract with protein, fat, and ash would be ex- pected because the former is determined by subtracting these three nutrients and fiber from the total dry matter. No attempt is made to explain the other correlations shown. The rela- tively large number of nutrients with which fluorine and carotene are appreciably corre- lated should be noted. The more significant correlation results which, after being averaged for the two years, equaled or exceeded 0.3 are presented in Table VI. This figure was selected because correla- tions of this size or less mean that 9 per cent or less of the variance in one factor is associ- ated with variance in the other factor. The simple coefficients of correlation between all nutrients are given in Table VII. DISCUSSION The observation of the feed industry that the protein content of corn declined appreci- ably in the years prior to 1946 seems to have been verified by this study. Morrison (4) gives the mean protein content of No. 2 corn as 9.4 per cent. In contrast, a value of 8.7 per cent was obtained in the 1946 survey, and one of 9.1 per cent in the 1947 survey. Barley and De Turk (3) concluded that "the decline in protein reported by the feed industry is not due to inherently low protein hybrids" and that "the decrease in corn protein is be- lieved to be caused by both a decreased soil nitrogen at regular rates of planting and by an increased rate of planting on soils relatively k>w in nitrogen." These views were not contra- dicted by the results of the present work. At least, the heavier corn producing areas tended to yield corn of low protein content, and varie- tal differences in protein content accounted for a very small portion of the variance. Nothing conclusive on these points is offered by the present studies, however, because their design did not permit arriving at such conclusions. Regional, state, and varietal differences were appreciable and statistically significant for some nutrients. For most of the nutrients, however, it appears that these factors per sc have little to do with the nutrient content of commercial corn. To account for the major variability in nutrient composition of corn, one must look to other factors. Soil type and local- ized climatic effects are indicated, with atten- tion directed particularly to some of the things, such as the micronutrient element content, which might vary within a given soil type. Other factors of possible importance are those associated with cultural practices such as fer- tilization. These factors seem especially worth consideration for those minerals and vitamins which have high coefficients of variation. In surveys designed to inquire into the causes of locational variations in composition, it appears that special emphasis would have to be placed on the above local factors in allocating samples, and intense sampling rates would have to be employed. Controlled experiments would pos- sibly be helpful. The change in protein content of corn through the years indicates the desirability of periodic checks on nutritive composition of feeds on a national scale. New varieties of crops and greater use of fertilizers, particularly the micronutrient elements, permit the production of many crops under new and different ecologi- cal conditions. Thus, different soil and climatic factors combined with the new varieties may
