THE UNIVERSITY OF ILLINOIS LIBRARY ; Return or renew all Library Materials! The Minimum Fee for each Lost Book is $50.00. The person charging this material is responsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft mutilation, and underlining of books are reasons for discipli- nary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN LI6I O-1096 The Yield Complex of Sweet Corn I. Effect of Advancing Maturity II. Relations Between Yield Components By W. A. HUELSEN and W. H. MICHAELS UNIVERSITY OF ILLINOIS AGRICULTURAL EXPERIMENT^ STATION Bulletin No. 432 CONTENTS PACK PART I: EFFECT OF MATURITY ON THE YIELD COMPLEX OF SWEET CORN 511 PLAN OF THE EXPERIMENTS 511 TECHNIC OF HARVESTING 513 CALCULATION OF BIOMETRICAL CONSTANTS 514 CLIMATOLOGICAL DATA 515 MATURITY STUDIES BASED ON EXAMINATION OF INDIVIDUAL EARS 515 Relation Among Ear Components 520 Relation Between Maturity and Toughness of Pericarp 528 Relation Between Moisture and Chemical Composition 532 Relation of Time of Silk Emergence to Ear Development 539 MATURITY AS RELATED TO THE YIELD COMPONENTS IN A MASS POPULATION 546 Number of Sorted Unhusked Ears per Acre 547 Weight of Sorted Unhusked Ears per Acre 561 Number of Prime Husked Ears per Acre 565 Weight of Prime Husked Ears per Acre 567 Percentage of Prime Husked Ears 568 Total Weight of Prime Plus Dented Husked Ears 568 Percentage of Prime Plus Dented Husked Ears 569 Weight of Kernels Cut From Prime Ears 569 Percentage of Kernels Cut From Prime Ears 569 Weight of Kernels Cut From Prime Plus Dented Husked Ears 570 Percentage of Kernels Cut From Prime Plus Dented Husked Ears 570 Time of Appearance of Dented Ears 571 Yields of Culls, Green Ears, and Husks 571 COMPARISON OF Two STYLES OF CUT AND THE EFFICIENCIES OF CUTTERS. . 572 MOISTURE CONTENT OF THE COBS 574 SILK EMERGENCE AS A MEASURE OF MATURATION 574 GENERAL DISCUSSION AND ECONOMIC ASPECTS OF FINDINGS IN PART I 577 PART II: CORRELATION ANALYSIS OF THE YIELD COMPLEX IN SWEET CORN 581 EXPERIMENTAL METHODS 581 STATISTICAL METHODS 583 INTERPRETING THE CORRELATION COEFFICIENTS 583 EXPERIMENTAL RESULTS 584 Correlation Coefficients of Yield Components 584 Regression Coefficients of Yield Components 5% DISCUSSION OF CORRELATION ANALYSIS 599 SUMMARY AND GENERAL CONCLUSIONS, PARTS I AND II.... 601 LITERATURE CITED 607 LIST OF TABLES AND CHARTS 608 Urbana, Illinois April, 1937 Publications in the Bulletin series report the results of investigations made by or sponsored by the Experiment Station. The Yield Complex of Sweet Corn D I. Effect of Advancing Maturity II. Relations Between Yield Components By W. A. HUELSEN and W. H. MICHAELS* URING the very brief period when sweet corn is edible and suitable for harvest, rapid and highly complex changes occur in the physical and chemical composition of the kernels. During the latter part of the harvest period these changes become very obvious in the form of progressively greater reductions in quality until, sooner or later, the kernels reach the point of being entirely inedible. The rapid changing of quality in sweet corn as maturity advances is known to all growers and canners ; and because of its importance the subject has received a good deal of attention from investigators, especially with reference to the chemical composition and the progres- sive increases in toughness of the kernels. Very little attention has been paid by investigators, however, to the related phenomenon of total yield, which is a function of maturity and closely associated with it. In fact, altho total yield is the climax of all the farmer's efforts and is the most important aspect of the crop from his viewpoint, sur- prisingly little attention has been given to the systematic analysis of yields. Numerous fragmentary studies are available, but they do not thoroly analyze total yields. The yield of soybeans has received con- siderable attention, as shown by the work of Borst and Thatcher, 2 * Uhland, 21 * and Woodworth. 23 * Small grains have been studied by Burnett and Bakke 5 * among others, and garden peas have received attention from Sayre, Willarnan, and Kertesz, 19 * and others. Sweet-corn growers long since observed that total yields increase very materially within certain limits the longer the ears remain in the field, but that quality, on the other hand, decreases very rapidly after the kernels reach the milk stage. This distinctly inverse relationship between total yields and quality as maturity advances has been the cause of endless misunderstanding between the grower, who wants maximum total yields per acre, and the canner, who is concerned pri- marily with high quality. One purpose of the experiments described in "W. A. HUELSEN, Associate Chief in Olericulture; W. H. MICHAELS, form- erly Associate in Olericulture. "These numbers refer to literature citations, page 607. 509 510 BULLETIN No. 432 [April, this bulletin was to attempt to clear up this difference of opinion regarding the proper time to harvest the sweet-corn crop. In order to do this an exhaustive study was made of the yield and quality of green sweet corn from a number of different angles and with respect to time of maturity. As a result certain general principles are advanced which may be applied in the harvesting of and the contracting for the sweet-corn crop. The experiments were undertaken for the purpose also of studying the whole concept of yield in sweet corn. In a crop harvested, as is sweet corn, before it matures entirely in the botanical sense, yield must be considered as a moving average changing from day to day, even from hour to hour, in certain definite directions. Yield must not, how- ever, be conceived as the moving average of any single character or component, but as the collective expression of a great many different components. Thus the yield of unhusked sweet corn is actually the yield of a system of components consisting of the husks, the shanks, the silks, the kernels, and the cobs. It may be quite immaterial for ordinary purposes if the relationship between the various components and their proportional importance with respect to the whole is entirely unknown ; but for those interested in canning corn this relationship becomes sig- nificant, for a single component the yield of kernels is then more important than the yield of whole unhusked ears. In order, therefore, to understand yield of sweet corn as a whole it is necessary to under- stand the yield of the various parts, not only at any given time, but at all times within the edibility period, because, as already mentioned, yield changes constantly with respect to time. The discussions and the comments in this bulletin are made pri- marily with the cannery-contract grower in mind. a The market grower also is of course interested in problems of quality in relation to advancing maturity, but he is not especially concerned with the so- called "cream-style" maturity, which is really the early dough stage, because the corn is then well past the best stage for eating from the cob. He is more concerned with the late milk stage of maturity, which is equivalent to "whole-grain-style" maturity. Market growers should read the bulletin with this difference in mind. *In a forthcoming circular of the Illinois Station the different types of growers' contracts and the bases upon which canners in Illinois buy sweet corn, are discussed in relation to the data obtained in the experiments reported in this bulletin. 1937} YIELD COMPLEX OF SWEET CORN 511 PART I: EFFECT OF MATURITY ON THE YIELD COMPLEX OF SWEET CORN The investigations reported in Part I consisted, first of all, of studying under carefully controlled conditions the development of a field of sweet corn thruout the edible stage. The effects of advancing maturity on the yield and quality of the various yield components were measured. From this part of the study was obtained mean values, or averages, of a mass population. PLAN OF THE EXPERIMENTS The experiments covered a two-year period (1931 and 1932) and three varieties of sweet corn Country Gentleman, Narrow Grain Evergreen, and Golden Bantam. Because mean values may not lead to the same conclusions as observations taken from single ears, each mass-population experiment was accompanied by a parallel individual- ear experiment, and the connection between the two was analyzed carefully. Certain additional problems which arose during the course of these investigations were studied in a series of secondary experiments deal- ing with the relation between toughness of pericarp and moisture con- tent of kernels, the value of moisture content as a measure of maturity, and the relation between the date of silk emergence and individual ear weights. The mass-population experiments with ears of mixed maturity were regarded as the primary experiments, for the conditions of operation were truly representative of those under which crops are regularly harvested by growers. For the 1931 primary experiment a uniform area of dark silt loam was blocked out on the University farm at Urbana, Illinois. Selected strains of Country Gentleman and Narrow Grain Evergreen* were planted by means of a two-row corn planter set to deliver 4 to 5 kernels per hill, the hills spaced 38 by 40 inches apart. A fertilizer application of 100 pounds 0-16-6 analysis per acre was hill-dropped by means of a fertilizer attachment on the planter. Half of the field was planted 'The varietal name "Narrow Grain Evergreen" is used in preference to the more unwieldy Narrow Grained Evergreen. The former name is used almost universally in the canning trade, but the latter is often found in seed catalogs. As a matter of fact, in trade literature the name "Narrow Grain" is very popular and properly so, because the suffix "Evergreen," implying that the variety is really a strain of Stowell's Evergreen, is probably superfluous. 512 BULLETIN No. 432 [April, with Narrow Grain Evergreen and half with Country Gentleman. Dur- ing the early seedling stage the corn was thinned to two plants per hill. Cultivation and subsequent handling were in all respects identical for the two varieties. In 1932 the experiment was repeated with several variations on a different block of dark silt loam. The Golden Bantam variety was added ; the seed was planted by hand in hills 38 by 38 inches apart ; Country Gentleman and Narrow Grain Evergreen were thinned to two plants per hill, and Golden Bantam to three plants per hill. No ferti- lizer was used. Three separate plantings from the same lot of seed of each variety were made in 1932 as compared with only one in 1931, the dates being as follows: 1931 Tune 10 Country Gentleman Series I Narrow Grain Evergreen Series I 1932 May 11 Country Gentleman Series II Narrow Grain Evergreen Series II Golden Bantam [Crop abandoned] May 26 Country Gentleman Series III Narrow Grain Evergreen Series III Golden Bantam Series I June 10 Country Gentleman Series IV and V Narrow Grain Evergreen Series IV and V Golden Bantam Series II During both years Country Gentleman and Narrow Grain Ever- green made a normal growth. The first planting of Golden Bantam suc- cumbed entirely to Stewart's disease and was eliminated on that ac- count. Series I and II of Golden Bantam also were attacked, but produced a fair crop. Precautions were taken to eliminate such experimental errors as those resulting from border effects and missing hills. Each plot con- sisted of 20 hills except in Country Gentleman Series II and III and Narrow Grain Evergreen Series III, where, because of missing hills, 10-hill plots were used. Ten replications were used thruout, and only the averages of the mean yields are presented in the tabulated data. As Culpepper and Magoon 6 * have shown that the rate at which a field of sweet corn comes into silk is a fairly accurate basis of pre- dicting maturity, and as Myers 17 * reached similar conclusions with respect to field corn, a complete record of silking was taken in all of the experiments considered here. In these, as well as in other sweet-corn experiments under way at Urbana, it has been found that no advantage accrues by counting 1937] YIELD COMPLEX OF SWEET CORN 513 silks after 75 percent of the theoretical total has been reached. The 75-percent point is therefore used in this paper as the basis for deter- mining maturity. The validity of this practice is discussed on pages 539 to 546 and 574 to 577 of this bulletin. Because of the rapid silking in hot weather, counting silks on each of the 780 plots involved proved to be out of the question. In 1931 the field was divided evenly, and silk counts were made on 11 plots of each variety distributed at uniform intervals thruout the field. Altho a slight variation in time of maturity was indicated by the silk counts, the variation did not exceed one day. Consequently the date on which the majority of the plots reached 75 percent of the theoretical total stand was considered the silking date. In 1932 a similar procedure for each series was followed. TECHNIC OF HARVESTING Snapping. In order to eliminate the effects of personal selection 8 during snapping, all the shoots showing silks on the plots selected were harvested, brought indoors and sorted very carefully into usable ears and culls, and the usable ears husked. Husking was done by hand in 1931, the shanks being broken off as closely as possible to the butts, but in 1932 Peerless huskers were used. Husked ears were sorted into three classes prime ears, dented ears, and culls. In 1932 culls were further subdivided into green b (immature) ears and culls. Grades. The official U. S. cannery grades were followed as closely as possible in sorting the corn. Prime ears are equivalent to the U. S. No. 1 grade in all respects. Dented ears are No. 1 ears showing more than one dented kernel per ear. Green ears are ears the cobs of which are filled over a space not less than 4 inches in length but which are too immature to be used ordinarily in canning. Culls are ears which are unfit for human consumption. Cutting. The kernels were cut from the cobs in 1931 with a hand- operated Burpee cutter. This machine cut the kernels off in whole- grain style right down to the cob, but it had a tendency to gouge the *In harvesting commercial plantings, selection is partly a personal factor, and two pickers may differ by as much as 15 percent in the amount of corn harvested from a given piece of ground. b Some canners use green or immature ears. 'Final U. S. cannery grades for sweet corn were established only after these experiments had been completed. Mr. W. E. Lewis of the U. S. Bureau of Agricultural Economics, who established the grades, discussed the question of grades on several occasions with the senior author. Accordingly the sorting was done on the basis of the grades which were later officially promulgated. 514 BULLETIN No. 432 [April, cobs. In 1932 the latest model Peerless whole-grain cutter was used. As it could be regulated by means of a micrometer screw adjustment for any depth of cut desired, its work was unusually accurate. When the kernels were cut "Maine" or cream style, a Sprague No. 5 cutter was used and the depth of cut was set at i/& inch. CALCULATION OF BIOMETRICAL CONSTANTS Practically all data in these experiments represent means of several observations. For example, the data on plot results involve means of 10 replications. Wherever possible, biometrical constants have been computed ac- cording to the following formulae: i y f VZ\ r* y (~Y\ "l2 ; D = / _ (where JT equals the \ n n observed variables) .6745 S.D. P.E. m = - Assuming that the correlation equals zero and a and b are the probable errors of the means, the probable error of a difference would then be P.E.d = Where a, b, c, etc. are the probable errors of the means, and N the number of separate averages, the probable error of a mean of means, where each mean represents an equal number of variates, would be P.E. = N Where n is the number of variates used in computing the respective means, where a, b, c, . . . x represent the respective probable errors of the means, and k equals the sum of the variates, the probable error of a mean of means where the mean is weighted would be V wV + w 2 6 2 + n 2 c 2 n 2 * 2 P.E. = k Certain other constants which are used will be stated in their proper places. 1937} YIELD COMPLEX OF SWEET CORN 515 CLIMATOLOGICAL DATA Records on seasonal fluctuations of temperature, rainfall, and sun- shine during the periods of 1931 and 1932 covered by these experiments (Figs. 1 and 2) were obtained from the official weather station at Urbana, Illinois. In the crop year of 1931 temperatures were high, especially in September. The mean monthly temperatures were uniformly higher than the normal mean. Rainfall during July was less than half of normal but was in excess of normal during June and the first half of September. In the crop year of 1932, also, mean temperatures were higher than normal, except during September. Rainfall was less than normal thruout the entire crop period. During the three years 1930 to 1932 the mean annual rainfall (October 1 to September 30) accumu- lated a deficiency of 10.89 inches under the normal 33.82 inches (29.79 inches in 1930; 33.02 inches, 1931; and 2776 inches, 1932). This deficiency, accompanied by three years of excessive temperatures, created an unusual set of conditions under which to grow a crop. The yields of white sweet corn during 1931 and 1932 in Illinois, and especially from numerous experiments at Urbana, were somewhat higher than usual, indicating that the conditions in general were ab- normally favorable rather than unfavorable. This may have modified the experimental results to some extent. For the growing of Golden Bantam, however, the excessive heat was distinctly unfavorable, and the yields were low. Furthermore, Stewart's disease was exceptionally severe. The authors believe, therefore, that the experimental data per- taining to Golden Bantam in this study are not so reliable as the data pertaining to Country Gentleman and Narrow Grain Evergreen. MATURITY STUDIES BASED ON EXAMINATION OF INDIVIDUAL EARS In order that any effects of soil variation on the development of ears might be minimized, data for the individual-ear studies were taken from plots of the mixed-population series distributed systematically in the field. As the silks emerged on these plots the ears were tagged and dated ; but only those ears tagged on the day the entire field reached mid-silking, as determined by the 75 percent theoretical stage of emergence, were used. Thus the stage of development of ears used in the individual-plant studies was directly comparable to the stage of development of ears in the mixed population. The only exception to this procedure occurred in 1931 in Narrow Grain Evergreen Series I, 516 BULLETIN No. 432 [April, (j.) 3anivH3droi 1937} YIELD COMPLEX OF SWEET CORN 517 D Q !1 [u O o 00 518 BULLETIN No. 432 [April, 1937} YIELD COMPLEX OF SWEET CORN 519 t* O o o y. a Q t/3 p x 520 BULLETIN No. 432 [April, in which the shoots were tagged two days before the field reached the 75-percent-emergent stage. Every time a picking was made from any of the mixed-population series, tagged samples were also harvested from the plots mentioned above. During 1931 these samples included 10 ears from each plot, and in 1932, 15 ears. Only the total weights of the samples were taken and recorded. These varied considerably, of course, but not to the extent of concealing the trends shown by the successive harvest dates. Relations Among Ear Components Changes in Weight of Ear Parts During Maturation. Changes in weight and moisture content which occurred in the component parts during the maturation period of the ear are indicated in Tables 1, 2, and 3. These changes in Country Gentleman Series IV and Narrow Grain Evergreen Series IV are plotted in Figs. 3 and 4. The curves were fitted by free-hand methods, but wherever the means gave evidence of a linear arrangement, the regressions were calculated according to Ezekiel's 8 * method of computation of the formula Y = a + bx. The data for all ten series of the three varieties were thus com- puted, but as the trends were quite similar, the charts of only the two series are presented here. Mean weights of the unhusked ears showed a slight tendency to increase up to a certain point during the maturation period, and then to decrease. This tendency seemed to be more pronounced in Country Gentleman (Fig. 3 A) than in Narrow Grain Evergreen (Fig. 4A), but the experiment with the latter variety may not have been carried far enough to exhibit the full course of weight changes. Mean weights of husked ears likewise increased as the period of maturity advanced, but whereas the weights of unhusked ears reached a peak and then de- creased during this period, weights of husked ears had no uniformly definite tendency to decrease. Mean weights of kernels increased steadily ; and cobs and husks declined fairly steadily. When changes in weight of dry matter only are considered, it is apparent that the dry matter in unhusked ears, husked ears, and kernels increased at a nearly uniform rate during the maturation period (Figs. 3B and 4B). This similarity of increase among the three items was clearly due to the increase in the dry matter of the kernels, since the dry weight of neither cobs nor husks increased appreciably. Moisture Content of Ear Parts During Maturation. Percentages of moisture in the three ear components are shown in Tables 1, 2, and 3 and in Figs. 3C and 4C. 1937] YIELD COMPLEX OF SWEET CORN 521 ....... ! cs r CM co co m*o -co co d *o oo oo * co t^<- co co r^ co co co (- -co t- r> to t>- o co to co r* r oo 5 to i-* cc co CM co o oo *o CM cc otoo-oooo to ~- o *o en to to -* o r- 10 CM os to *-* o> i^- CM os t- r* co r- t*. r- t- to o 5 i>. t>- r-> to co co to to co to r* h* r co co to to cc -r ro o -f r- to o eo CM - 10 1*- o* I s - ^< oo *c ^ ~*^coco as c^ co 10 <*< to CO CM t^* t^ kC CM OO CM *O O (- t ^- ** CO -^ CO OO CO '' O *O >O CM CO CO N CM C5 CJ -H COCOCOCMCM^ CM CO CM ^ CM CO * o peoa>CMpcM CM CM CM t* c c*4 oo oo oo ^ Okp^*^*OO^H co as o r^. i o tMt^ *CM cot^-eor^cocoW CO CM CM CM N CM CM CO CM C^ CM CM CO CO CM -CO CO CO CM CM CM CM CS ^r^rioeM-^tocc ^-oeop^to tO^< -10 -r en -^ i - r-- -^ o O CM t* OO CO CO O> *-" C C O *O O -" OO -OS CM O >O CO I-* -^ CO co ^< ^" *c 5 o co ec co ^" <* c -^ eo * ^* eo co ^ ^ c 5 eo t^- ^ eo eo us co <* eo o w o os ostoh-b* -< oo ^< o co os ^ [--i-i i- r^> ^ a> CM CM IQ co 'O o i eo oo oo -*t- -t- r- CO I- t^ t- t^ 00 CO CO CO t- t- 00 KtOt^OO t^- tO O 1^- 1^- 1>- 1^. *- -^ 1-- O M" OO -V O O O CO f ^T 00 ^ -^ Q O CO <-i O O OO 1^- Oi a> oo c t- to o - CM CM o os ** oo CM oo co t^. o> to 25 35 S3 r ~ *c5 c-j OP - co M- o CM r CM 10 CM ^- to cc o o> oi r- <- t-^ ai o o . c s Cl CO*CCT>I^O CO I ^. O OO C?i ^- QO OO CM ^-< CO -^ CO ^T CO ^^ - 1 CM CM CM CM CM CMCMCMCMCMCM CMCMCMCM CiCMCMCMCMCMCM - oi c-< CM o oo o co os I s - M- - M< co to co co c oo *o o *o 5 J i 522 BULLETIN No. 432 [April, | 9 -^o& ooo>~oa. ^.o . i S3 .... ^^3. S.OO^^ g.8^8^^; 1 i i wo-r.axN^-0 - O O r co^oo^o^ao 5Ot- a>oo<0 lOlflOOOXOO 00 t^ IN 10 to TO 0> i '5 a 3 c^.esio -oooo- | 1 Si 1 SSSSSSSE: 8g5!gS5Sg SSSSSS 755SSSg 1 . ooor-^.<eo o.^. ooo>^^v i n 888K88SS SK8S88K8 S5SSSS2 8S8R88S 8 a ^.c.o5oo o.--,o.ooo, o^o>g > cot^oxocom ooo us o> co o> * t^ CM ^f CM OS O W5 -* CM O> O CO ^ r>- CO TT CM 00 <- ^T O O> OS t* O OS -^ -^ -i i t i 1 k s COiOaOCOkOr^OOCO CDOOOOSCOCMOOaO CMt^-COCMCOtO OSCOkOOCiOCOOS o i 1 I 1 J i i 1 S?SSS2S;S SSS55SSS SSSSS2 SStSSSKgg a : 1 _ 1 j i L 3r COO ^ O '^ CM O CO OO O >O CO CO O ^^ t*- 1C CC ' C5 Tf Ci ^H w CO eo ^H OS t* 1 4 1 [ i I s -r. 1 4. e o a ies o> oo oo o> t. oo eoooo o> a> o t. ^- oo - OS ^^ CO OO O CM ^> COOO O CM CM CM CM CO eO CO CO ^^CMCMCMCMCMCO CM CM CM CM R CO -^ CM CM CM CM CM CO represents means 1 sUUiHiiiUMiMIHHUlUiiHH ..^.. . N . i-d***' * . . Note Seriea Moldy. 1937] YIELD COMPLEX OF SWEET CORN 523 I H O i E W , 1 1 _ 1 O5 O5 OO O3 CO O5 ^f O> OS r^t^t^r^t^ oooooooo . I 1 J 6 CODOOO-O- 05 00 t^t^-t>-cDr n COCO I O O9-H-H U9IOU9-W CO-WM ^ 1? J c -^ r i i - ~ > ~- ~ -: f 1 H 1 *< * 00 CM C31 O OO OS J ^ -^t ^f. IQ kc co ua IQ CD uj>ooioion oceivooec COCOOQO g CC CO O f OO ** OS W CO 'c II 3 re '- -i- ^- -r i-i-:'. MCOMCNCM OCNtNO) I l JS 0)00!MCO IQtMOCO " 1 1 M o t e* oa oo r oj CD CMtNtCO * 2 loiooiof o irar- E o* w co co *OO OOOSC-J ) of 15 ear RtJ c^ Qrj i 2S?5^S SSSSS I J 5 5 1 s : :::: :::: i J 4 8 i i i i !|J ! i i ^.COOOON^. ^gg S i Q "2? 5? 'E-* "E^ to CO X, 524 BULLETIN No. 432 [April, 1937} YIELD COMPLEX OF SWEET CORN 525 526 BULLETIN No. 432 [April, In 1931 moisture-percentage data were taken only for the kernels. The method used in 1931 consisted of drawing quadruplicate 50-gram random samples from the kernels cut from each 10-ear sample, and of drying these at 85 C. in aerated thermostatically controlled electric ovens for a period of 72 hours, at which time they had reached con- stant weight. In 1932 the same procedure was followed in drying the kernels, except that a drying temperature of 80 C. instead of 85 C. was used. In addition, all the cobs from each day's picking were dried at about 65 C. for 48 hours, and stored for several months, after which aliquot samples were taken and dried at 100 C. for 72 hours, at which time they reached constant weight. Moisture data were also taken in 1932 for husks, but because of inadequate preliminary drying some of the husk samples molded. The dry weights of the husks were determined by weighing after an initial drying period of 65 C. for 48 hours, followed by a second drying period of 72 hours at 100 C. Moisture content of husks, of cobs, and of kernels declined as maturity advanced (Tables 1, 2, 3 and Figs. 3C and 4C). This trend in the moisture content of the kernels confirms the findings of Appleman, 1 * and Culpepper and Magoon. 6 ' 14 * By measuring the mois- ture content of sweet-corn kernels at successive intervals beginning five days after the appearance of silks, Culpepper and Magoon 6 * found that in general the moisture content declined as maturity advanced. The rate of decline varied with the variety but correlated closely with field observations of moisture content in relation to maturity and with the consistency of corn canned at different stages of maturity. Moisture content of the kernels is, therefore, a good measure of maturity and, as stated before, has been utilized as a measure of maturity in the experiments under discussion in this paper. The percentage of moisture in the husks remained consistently higher than in either cobs or kernels, but the rate of decline was not materially different. Relative Ratios of Ear Parts During Maturation. The relative ratios of the ear components computed as percentages are given in Tables 1, 2, and 3, and the linear regressions in Figs. 3D and 4D. The relationships were similar in all three varieties, showing that the percentage of husked ears and of cut kernels (calculated on a basis of weights either of unhusked or of husked ears as indicated) increased as maturity advanced. Moisture, as shown above, consist- ently declined. The percentage of kernel weights increased at a more rapid rate than did the percentage of husked-ear weights (Figs. 3D and 4D) because percentages of cob weights declined instead of 1937] YIELD COMPLEX OF SWEET CORN 527 remaining constant. Both the fresh green weights and the calculated dry weights of husked ears and kernels increased and decreased at practically the same rates (Figs. 3 A and 3B and Figs. 4A and 4B). By plotting similar percentages obtained by Culpepper and Magoon, 6 * similar conclusions as to relative weights of ear components may be reached. The foregoing results show clearly the importance of maturity in relation to yields. Other things being equal, even a slight delay in harvesting sweet corn at any time during the period of edibility in- creases the yield of cut kernels and materially reduces the percentage of cob weights. It is obvious, therefore, that harvesting should be delayed as long as possible. Quality, of course, according to Culpepper and Magoon 6 * and others, decreases as maturity advances. The rela- tionship existing between delayed harvesting, field yields, moisture content, and increased toughness of the pericarp will be discussed later (pages 577 to 581). Altho the sweet-corn ears on which data presented in this section are based were of identical maturity as determined by the simultaneous emergence of the silks, such ears were not actually in the same state of physiological maturity. The variation was much less, however, than is common at canning factories, where the ears vary all the way from the early milk to the dented stage. According to these experiments, relatively immature ears reduce the yield of cut kernels, whereas older ears increase it. Therefore, such practices as soil fertilization, strain selection, use of first-generation hybrids, Hoffman's 11 * system of grading seed, or other practices which tend to reduce the range of maturity, will increase the quantity of cut corn obtained from given samples. Causes for Variations in Percentages of Net Yield. Altho, as shown by Culpepper and Magoon, 6 * sweet-corn varieties may, because of inherent genetic characters, vary widely in percentages of net yield, the trends displayed in these experiments by the several yield com- ponents remained virtually identical in all three varieties used (Tables 1, 2, and 3). Other factors which might cause variations in percent- ages of net yield are climate, seasonal conditions, and fertilizers. That climate, as modified by geographical location, exerts no consistent influ- ence on the relations between the ear parts has been demonstrated by Straughn. 20 * Time of planting, likewise, as shown by Magoon and Culpepper 14 * and as confirmed by the present experiments, seems to have no influence on the relative values of the ear components when the ears are . harvested at the same stage of maturity as determined 528 BULLETIN No. 432 [April, by moisture content. But fertilizers, particularly nitrogen salts, ac- cording to considerable evidence obtained by the authors of this bul- letin, may modify to a material extent the relationship existing between the various ear components within a variety without necessarily chang- ing the relationship between strains or varieties. Relation Between Maturity and Toughness of Pericarp Quality in green sweet corn is to a large extent dependent upon tenderness of the pericarp; and in canned corn tenderness is undoubt- edly the leading criterion of grade. Increasing toughness is very closely related to advancing maturity, according to Culpepper and Magoon, 6 * Haddad, 10 * and others, and cannot be ignored in any adequate interpretation of yields in sweet corn. Penetration Tests on Pericarp. For practical purposes toughness in sweet corn is best measured by the pressure required to puncture the pericarp with a needle of standard size. The force required was, in these experiments, measured in grams and is called "penetration value." Toughness is, of course, in direct relation to the penetration value. An instrument for measuring toughness is described by Culpepper and Magoon, 6 * but as it is subject to all the variations inherent in coil- spring balances, a special device was constructed for these experiments. In brief, the instrument devised consisted of a compressed-air chamber having at one end an elastic rubber diaphragm in contact with a plunger holding the needle. The pressure of the compressed air upon the plunger was registered on a manometer. The instrument was cali- brated with a laboratory balance, twenty readings being taken at each of a number of uniformly distributed points which, when plotted graphically, gave a straight line. From the equation for this straight line the manometer readings were converted into corresponding values in grams for the pressure required to penetrate the pericarp with a needle measuring .04 inch, or virtually 1 mm. (.0394 inch) in diameter. Thirty penetration readings, taken spirally from tip to butt on each ear of the 10-ear samples used in the individual-ear studies in 1931, (Table 4), showed considerable variation in the penetration values of the kernels on individual ears. Furthermore the mean penetration values of individual ears frequently varied widely within a 10-ear sample. More nearly uniform penetration values within each 10-ear sample might have been expected, as the ears were of the same physiological age measured by emergence of silks; that is, the silks of each ear emerged on the same day. The lack of uniformity in mean penetra- 79J7] YIELD COMPLEX OF SWEET CORN 529 SS-, H 2 2 < O 2 til O H 2 O < D M 1 2 O I J4 2 g si O O 'f t^ O * 00 - *~* |-s c oj 0)J2 6 O 1 *8 >O<^ -H-H-H-H-H+I-H-H Narrow I 1 -*-*oaoo (NfS M * VO O OOO fStSNtSfO'^'O'^ V c 08 3^. Ifl 't' 00 ** " <^ O> s: ES O si o >^ e V 6 t -I VO OV)OC t^ r^. t^ t^ NO in >o 5 & c o >O'OOOr- U Penetrati B? 32 "3 ~* (N (S H-H-H+I-H-H-H _ ' ; _ r - C ir t^ n m oc i~ (N (N fS (N (N n 'O 1 o u 5 00 fS ^ 00 O oQ^ >o -* r^ 00 w co : c^O ^O^ 00^0*00 t^-cj'5' ^ o o eo < >C^GO OO -r OO t^ OO r~ CO O C> r~ ~1 /; ^ u? OIMOO 0000*0 050505 oo e^ o> sor~oo 0000 - c^ - CO wcocoe 1000 - . CO ' I C -OM CT . -wic^ ^- *o c5 ^ c*ao (Ococ^eo oo co ^r i c O 5 *^ -H CO ^ ^ C*4 C4 -OO^H 01 co oa os o oooic^ -icoeoaa^' c*a co ^t^ oooaoso*oo oeoooc tO'-'oo^fr-o* iracseo cocooi^'-" oc^oc coeoaa^'f*-ao otoiooso^t^ ^-i CM ^ ^ 01 1-1 OJ ^ ^ ^ O* CO ^ r* t^ o* oo t^. co ** o eo 01 *n 01 coc^ost^oor^-oj o* to as -^ CO^''^O'-"CQO co'-o'^'Oeoc 3 5 OO Oa ^^ OO OO CO CO O r- r^ 'OCM O5COO^^ -r t CO - - ~ T ^-^reo^-ooo oooosoo^c ^r co ^r o c r i ^ 71 : OSCOt>-O>COOO '-t*QCOOOCO r-^!O^5 OS OO C t^- Q ^^ " I s - CO 9 4^ a>oiooocoo oitoO'oa^' r-oi-^co o3r--o-S4:o o'Oor^ -3t>-*tttO'^r^H *-"OO "Oo3 o * as <- too^cooot^oo oot^-^'t--c 5 3 CO CO < 00 OOOOiO CM ^ O O O O O O O O> O OS O5 O 22SSSS1S 2? - ' 1937} YIELD COMPLEX OF SWEET CORN 549 coop 'cooo'io " ! !2 ^? oooo 5? c* o o o -a* u? *f* h- oo c< Aa?) eo oa w -* c*a oo oo w oo 'CJcoco- -i "'HIJ; lOdm r_- -f-tl~s> U9OIO ' -! ^ O) O) to ; ^'u}^'~ o o> o> r~ t 111 a ^ CO 00 OO !0 CO O -OOCNCO COCO -CO NtNcMioreco N 'ree>co ci 't> :oooooo oSSooo oooo oo oo co co go a; IO^- aoa>co -co o t^ -^ c ~* in re co i coco 9000 co r^o>-H r~ooo)ra >o o r~ Q o> o us o o o os o CN ~ co ^ ^> c QO o oo oo o o -r T-I *r ic o r^ '-o ccco cMt^coioc t>- re oo co oo Si ^ -^ao-^-^-cooo oo ^i CM ua 01 -^ r NtMrecocorere _ 1IS i=is t~^rO CO'-'O CM ssssiss o io o o o -^ o^r -i-* -, _ co ao &* i us Me^coeo^ -^ Q 00 M * *r O -H m 550 BULLETIN No. 432 [April, oo -r o o oo co It t~ CM ^- CM ^ t r r - -r - ~ ~ t-tOO ^1 CO ^ CO M5 CM '<--!-.- Ocor-oouscot^ co o> oo jo o> r~ o> CM cb coco-" t^co jr r^ co co ^i us co CO CM --rl~r-?1 CO O !*( CM CO CO CO*IQ -lOtO O CO * OO CO CO "^"S-Jco ">** '^3 "" 2 3 " M CM 2 "' t ^ o 5 CM r* r* o o r^ t^. H co to t* oo t co 10 ro -^ * c^ co O CM r~ 5 ^- eo r cooiOsocMt coooos co oo oo 10 o ^r t CM o oo CM as CO CO CO CO CO CM CM CN CO CM CO CO CM CM CM CO CO CO O5 CN CM CO CO CM CM CM'fCOCM COIOI CO>Ofl CO U5 CM CM - CM CMi-r-H CM -COIOCMO 1O CO GO t b- CM CO - I- '. CO CM" CM O OS IQ : SSS522 : CMOO-i OCO CM CM Oi CM CO ?-- CNU9CO COCOOS^r I - = r -JTl .-,- CM CM 00 :"2SSS COlO.- ) "~t t^ CO 55 pt --HUJCOCM os CM p ce *nmt~- COTTCM apop^r pco -cocMpa) W-X" -COCOCOCO CM 'iOCO -COCOIO OSt-M OOO> O t-OOcd 00 O US KS CN CM i-i^M^TtiO wr CNCOr~OOCOCM O> t^ O 00 CO O ^H -t- -ji os t^ os =o CM co * ao co co r CM os CM - oo as os o co r~ oo co CM MS Scocoai-" co OOCOQOOCM COCMCOCO oo us co oo r- oo CMTCCOCMOOIOO> lO CO *O O CM *^ W3 O "3 ft CO ^^ CO C CO CO *i O CO CO IO CO >O C CO CO *3 ^i O CO ? t ^^ CO H3 O>-^CO"5t OS CO US r OS CO 00 Q CM * CO 00 COOOOCM^C cS C4 CM CM CO CO CO S. CM S C CM CM OSCMCMCM S CM CM CM CM ~ CM CM CM C s^jss sasana ^3s YIELD COMPLEX OF SWEET CORN 551 ? Cd "3 1 Q - w eC < 1 e 9 mi iiinn OH ^ . i I 63 -^S tClo ^2 N 2c^c5e3 "^SSS^SSJ 1 0, OS r-- O CO ^< O4 oo c^ CO r^- CO co O (N OO *O CO CO O O ** SSSSSgS S?Sgg?5S 8833 SRSSSSS SSSSSSS II > cooot-. w-r- -eocooo ooo> - r. -< ootcoiNe4 8 1 | Q - ti 3 8 1 K PL, - - r '~ ~ " ~ -* Tl ~ i.^ ^- -^ --i.': o CO -* t^ -H (O C4 COO<0O4OOOO s 1 0, -H h -OSO* -^" -OOU9CO * .... s N -OO -COX90000 'O .... + "8 1 n CL, i i i * j t 15 ai ! i 88X5ZS3 = SSJSaa 83S8 2S8SSSS 22SS1S S S \ \ s I ijjjjjjjljjjjjjljjjjljjjjjjjljjjjjjj i 1931 huaked ii 1 1< ii ii i i* $ V 552 BULLETIN No. 432 [April, g^ SRSSSS :S * O go o O^i"r- ~. -M oc icowoooooo >-: c :~ 71 -i >~ IOWNOO--O M o co t^eoooooa> co r- eo CM e o -* r^ r^ r co r*. i> f~r~rr^ r t~ t~ r~ r>- eo co b- r^ t>- r- 1~ h t> J! 3 O 00 > TO 0> W O CO O CO f- !M U5 CM CM OO CO CM CM O p OO IO fcl - ^ E .... . --w-oo- -oco fi ^t | 8 ::::::: i i :SSS :3S : : :SSSS : : :SSSS g COOCOOOOIO ~~ *~. f T I T] I ~ i^ C !- -1 ~|-J~r-7l M -cs i::;;;; cooJt^op'^rV = o * t^ oo co o> >o >- opoococo S t^t^coScoco cococeso r^ ^ > co co co o h- r- r- 2 S co co Q eu ' "22 . C J (2 Q |0n ...-4< . - _ . ,-_-,-, _ r . :: .-O> ri :::' :: ' ' :;;' g VCOlOlOlOr- VU5C3SM OOOCO- -OOOOM ..O^>- . -4* 10 ^^ e*3 eo e*? t~ o co w e^ cc Oi o -m-*tnir -o w CJ CO CO CO CO CO CO 1O CD 1C CO 1C CO to CO * CO CO CO CO ".- -S -^ =SSS = = = ^ = = = = " ^ = j== = = = = = = = = = o 8 I JS & H IJ ^ WB + oect^.o wa^osccooco *ooo-^o c^eseo * M'^-'C^oooto r--i^-cot>- oo c? oo h ^ i^^i^-aonooo I o I .5 j| .2 5 .8 ? .8 S 1 .8 ? J- 03 03 05 05 OQ 1937] YIELD COMPLEX OF SWEET CORN 553 -^osoopo co co r- op p o o> t- ^- eo o -* r-* opeOio^ co o so -coco * CO CO CO O t * i- CO * OJ h t -jr.-.- T r? .- r-7 r-? * - -i " -t ^Oraoooe^ coooeO'-'^'Ocoe* oopoaooor co ooe>*o * ~ -i r i - - - SSo'Io-8 SSSSSSS3 SSSSS3 SS88S8SS SS888&11 ao eo i-go co ^ r CM o o coc?4or>oooic^eo co^aOoi oooic^eo co o -^ " co c >ooo -^-p -^ON^^fC^-H co-*rppeo oo co co T ^^ co r c>j ^r eo <- mc^T< ri r i TJ- OO Ci ro c^ , _ -j ri OO 00 M o 3COOS - lOcOcOOS c< ^ CN o oo ~ w b-op -coooopto pr- o>-^- ' 5C^c^^)^ u^ooo>' 'Oco ^^ eoiooso^c -0000000031 < ^r ^^ c*i o ?j os cc c^ o b c^ -* 3 o 0000 O OOOOOJO NWNNNCO ?3 ' 99 r 554 BULLETIN No. 432 [April, IJ 1* I- s fclw E I 1 IB w CL, Is OOC6C-1 - I - -^ -. ssss 8o*S5 _< ^ (OUJO -OOC-100CO o> -(Mooto (M IOCS to -OC 5 5 g 3 is 5 < r o . c -H !> CO O t^ SSSSSSSS SSSSSSSS M > u3COOOO>tt>5CD ISS8 :is3 joo c^oto; >OO OOOO ! Sooooooo SIN^iCO I-H to CO '^* ' COCO-HO -O I - r gOstOTf! CO -HOC OO TT IN tO tO t^ tO 00 O i " CO O IM t^ 1O CO "H N -^ oo oo -H to o >OSN^< to - co to en to os < > i iO-<^ O O O O ^H o O -H _ _ _ _ T 1 ?] 7 1 KwKt^aoB OO CO >O tO CO to * o t^ 5 CO Oi C CC OO to Tioo>-a> s: = ?? * o2js9oo -ssiatTSs ~* tOOOOC^^tOOOO tOOOOCS'W tOOOO -^ -> N PJ (M CO C-) IM OJ N CO -" O C< C"J (M CO CO WM (M CO 8 9 "S. II Is* I ^ i c< 08 'C< dS 03 03 1937} YIELD COMPLEX OF SWEET CORN 555 tf) CO CO CO f 00 - COOOCOOCOCM ff~O>O50O O M- CM CO CM CM O> CM CO CO E 1 5 ; | a coiooax**! oor~r~ooooo cooooooce t~ r- oo -r co t o> M- M> eo CMCM CM COCO COtMCM CM 5 * >o 10 co oo oo t~ r^ co t^ co 10 locooo^coco -f coiomoous -r at J co ** co eo I 1 1 -g > ! 1 1 gSSSSS?5 S&a;a9: ^^3? Sfl8SS899 SaKSSSSS 1 ! j t w . . . -0-5^. . . .^0,COM . . .M.O.M.CO _ c,^ . s I i i s i i I 1 a, g ~ ] ! c i i s 1 : : . saft - - 2- . -aft : : : : .a : : : : : -S i ^T i r OSOOMOO ^OM oscoo -c^bf* a - tu r faC | i I centa = = 1 eu ^-1 CM-H . Ol | : -> I 1 O3^^'a . . . .COC-M. . . . .cosc^ . .cocooco oooo^r e*s - h 2 - 5 '5 1 a eu a O ! s. E "a ex. . . - --r -.c -.0 as ro -^ co *n eo o o - - ^ eo ^ c co M o* o o>cCO AOQOCO (?-J OO Q eor- .i-. M ?i^i ~ eoN^t . 1 ^ t>8 ; > ~ CU ~ ^OTI aou7 ^it-'-ro^c-ioooo oo -^> -^ o c-i ac -^5 t- -^ o t- -^ - F - 7 i r-r r -- r. 1 3 i 9 1 O *5 so S o ^- ^H ^r ^r * 10 o ^r c5 eOkO^to^<<-i ^t- * :c 3 co o us v co * so S r*. to co -^ 1 t 5 M 1 ii b 1 5 Q ti i 2SSg??Sg 22SSSS8S SSS^S 2-SgSSSSS 2288S8g ) V, where pickings of A o hard to be cut. i i i < Date of harv< LnnnninNMniinniinnnninnnn If a 4 *! 1 -il *! l-'i I *! 1 03 og eg BQ co * - 556 BULLETIN No. 432 [April, 1 T IM t-om -t~oooo OON 'MOsoO"- - ->-OO-" ooosec 5 ooecaoiooo r i a -H \ j QJA. a iiiiis B 1 ; * 1 1 "8 rf <> ^- o> t to oo o> -"OCooM o o> w r- oo o r~ oo n o r- r^ oo -r o * r * ccr-ootoojoo) c i>ra-rfi'WC-> --o -rmooooo r- o n --c r~ 3-, -r d d ( ^ E d 3iM oo ee> O3 o eo " ic oo oo i/d oo o oa to -^coco<-i * r~ oo oo h- * oo to oo o oocc o ; 1 d c | ok 69 d B 1 e ^ i a. f' CO 1C t^ O kO TT OO OO kO C O> ^H < N to ud O tO OS lO OO tf3 C to OO M O N O> -r r 1 .- r. '- I - S'S' ' i H H i^ ec o o < to -^ ^^ d oo oo to to o os ~- ^ o* r* c^ -^ oo oo -^ c^ oa oo t^> >o 01 oo * ^^ oo oo to 113 CO CO CO O rr r^ ?a rr - S 8- - V z _= > *J = B 1 Z s h 4 :- r r-. tr: T O CO *- tO CO OO O CO OO-^Ot^OlO CO t^ OO -4* OO O tQ O OO CO IO OO O CO Ud CO h 1 J E a .2 J *; 5 5 ^ -COCOO ! ^1< OJt.-iCO - ~IO . . . . i 8 H: . Z V E Q & *9>a -o> --.--. . - f . . trt^'' -CM -io .... -, y _ r -_ . . . r^aOT . . . . ol O 2 o g 1 64 eu f5 co"5NO 'io ^-co .... - - - . -i ecco ^r . . . . il -5 i 1 >F MATURITY ON KELAT (I I'J ^ s 1 1 os CN ^ co o es * co co oo o CM -*r co oo o ^-rc*cr^^; eooo oe^ M*COOO o co oo o tN (N (N CO " N os ^r * *o os 00 itAooioco o> o> ^- o <* r- S O> O OS 00 eO i- re c ~- O ~- OO f2 K S to O * - co - 3OOOO >c^r* o g *c CT> *o 01 I 4Q Ud ' 5 *O U7 S "^ 1C * O OO O IN to '-O CS to to . to woo oo oo 03ioSS -OQtOCOc cor ooo *o t oo co oo -i os os r o i C^ -f -tD *- OO OO OS OS ift *- OOCOC4^<^*t >^eo r-cooooc4c^ MOCO^OO^< > 0000 >o 9O OO ; C C ^. iii; O C5 O O ^ t SCJ-^fOC^iOC** ^"Ot^iftCOt 1 CO^-O>COC^OcOI CDO>^^^^*OC^O o^^^^c^toc^i cr*o> cr*"r-i oo ao -*f d r- oo ~ -* -^ O O * ec o - c c: -M o -r >o - T co o*ao>oocor- cs to oo CD co ^ *- O j - -r i - TJ -r o T O cOOl^cOt^tA -C4 O TJ ro M 1 O os *C CO CO CO N CO M ^- tO 00 O to U5 O CO t>- OS h- tO O CO CO CO * CO O CO OO r^co*oo>5oo CO^QCOOOK? o> eoic O III coeococo i-i-ic " C 8 a 1 ' s II gs * - > ill III III 558 BULLETIN No. 432 [April, B o U i as o O 2 en E O o (j rt 03 W > (d Ovt '500 "JO S u bi OO^UT* <*5 *(-. 11 Q cu bi fti +*** -000.0 t \r> r*> *r <*> >e *f&r*n ooooo oooo i 1 Qcu bi cu 'f m -fO *'f 00 -OOO (N ^^ r^ ^M ^- (N es & OOOO OOOO irs per ac s 3 l*5 00 O^'O'f t^ >O>OO> lOOOOf r~ooioi f> CMMPO<^f*> Ult^'OO "% u Pv4v H 23 fe-c M S Ja t w ?soo**O r r- t w OMOOifl O<5 Tons of ur QCU- bi cu Ot^^O I^U1^<*5 *t rj *t TJ> o 'Om'Ooo ooooo oooo Q cu u cu u-, u~, ~i r X --, i/it-OOO WOfO ooooo oooo 1 o o> o ^ f"l t~- o* ** ** OOOOOfO <~- fi u (4 t w !> -f -0*00 P cu bi h *TjTt fS St O>OO<^00 ^fS l^O fS'O INrOOui ^mmfoo OvoO'l'tN s of husk i (St^(S 0<^00 OOf-J 10 O>O si w (MOO (N "5 | 1 Q cu bi cu . _ -r _ -i -i ~4 OlOtNfOO 'OOOTfO> OO tNUTO CS1OTJIO> tststss Tftsr^to Q cu u cu 10 Tfro r- >oioo>a 0000 10 (N(N(Sf^^ HOPOTtHO ooooo oooo *1 21 OO>OOOI^ lOO't IOCNf5(NO Ov^OOlO Oi^iOfOOv OvtS IO I ^fJ^ilO^ (2-g 3 O>O O>OO< c^ ~-"*~* S 5 n9 aj.S ^1 ^ *^ r*5 1/^ r^ N f^s 10 r* HCStS ; : ; ; S! : : : : :S : : : : ^ o oo o ts f oo o fs ^ _ _ _ _^ _ ~, -., ~j S3 5 2^ fe< fc< co to 1937} YIELD COMPLEX OF SWEET CORN 559 o U w w I a X 3 CQ H a o z u s Q a. u c CQ a v in u 2 5*8 si s IE s; w' <^t>-.tO(N W W*iflO^ ^'i^'OiO 0" i 0. - 1! S|.s OO'i'tMio >OOOit^ 8 M "^"i^.w<5 ooo^m r " PI fl* O^'J-tSO O>Oo>O a J3 U OOm^fxO M NO -OOO - "OO | > w -10 . .O 2 ss i/im -f^N '-fs O OO* TfO^^ *5 f*5 00 O O IO ^ O IO MfN(N(N(M CStNCSCS | v r^ (S l/>\O (X S K S o 8 t^OON fS .1 t w OO <*) W H -MN 2-JS QSu ^tW5 .(S\O O -IOU1 | SiM fooo -t^Os O -r^io E-^s 1ST W ifllO^'lO'C W500>OH II 2 8 fik w Ot^ -^" O -"HO c- o- x x -f -r*r.t-x st " p cu "Si c f^t^tst^fs r w O*O o Q Cu OvOXN <3 -fO h <0 o 1 foioo a-iCoofN o^t^oo o^^oo *rTO5 5^T)-J^ 2 u ~ *r. -s. -s. S_"-.X\r. s h H C CH si i w .KSO'OO (NO -t- C c ^c vot^omo ov>o^io Q cu . oo '^'i' * 3 ss - -r ir s t^OOOO f*5^"l/)lO>O u^O*O*O n " O 2 W O *-l"^l W>^I"*W < V IX M^ u M r*s JS-^ it: o -H^.IOf-t^ >O. (S 8J si * "^ O\ W5iI^ w>lfl^ -NCNNtS(S fOf*5 fO^^Of*! ! S! -^>or ior (NStS(N NCStSN 5 'o V 15 Q -V (\j ' ' ' ' S? : : : : :S : : : : of liarvest -^ M . . . . ^ : : : : : 2 : : : : "< O 00 O < ^f XOtN'T -"< fc,^-(NtJM 8? 8^ t< fe" 1 (0 to Q * 00 O M * OOOCMTf >-<"-< ^,-rstN "> Sf *> Sf v 3 B P l r 560 BULLETIN No. 432 [April, to the fact that harvest began at different stages of maturity in each series and also to the influence of changes in the weather. The increases in the number of unhusked ears are easily explained, for the reason is obvious from Tables 8, 9, and 10. As the harvest season advanced, an increasingly larger percentage of the total number of ears which had been classed as immature or as culls (and thus in- cluding lower ears) in the earlier pickings had now become sufficiently 90 80 tso 30 20 10 PERCENT PRIME HUSKED EARS PERCENT DENTED HUSKED EARS PERCENT KERNELS FROM PRIME HUSKED EARS PERCENT KERNELS FROM DENTED HUSKED EARS PERCENT KERNELS, PRIME PLUS DENTED HUSKED EARS PERCENT MOISTURE. KERNELS FROM PRIME PLUS DENTED | EARS, WET BASIS COUNTRY GENTLEMAN I 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 DAYS AFTER 75 7, SILKING FIG. 10. RELATION BETWEEN MATURITY AND PERCENTAGES OF PRINCIPAL YIELD COMPONENTS: COUNTRY GENTLEMAN SWEET CORN, SERIES I, 1931 mature to be classed as usable unhusked ears. The results listed in these three tables do not, however, explain the significant decreases in number of sorted unhusked ears encountered during the later pickings in some series (Tables 11, 13, and 15). Were such decreases accompanied in turn by corresponding increases in the number of culls, the explanation would be self-evident. Since no attempt was made to count the number of unhusked culls it may only be inferred that the number could not have increased without also increasing the weight, which was recorded. Weight of unhusked culls did not increase. It is therefore probable that the number did not increase. The present data are inadequate for an explanation of the decline. 1937] YIELD COMPLEX OF SWEET CORN 561 There was apparently no direct relationship between moisture con- tent and the period when the number of sorted unhusked ears reached the maximum (Table 17). The mean moisture contents of Country Gentleman, Narrow Grain Evergreen, and Golden Bantam were, re- spectively, 65.7, 68.4, and 66.4 percent, but the variation within each variety was very large indeed. 70 550 0.40 10 I BASIS: SORTED I UNHUSKED EARS _ PERCENT PRIME HUSKED EARS* PERCENT DENTED HUSKED EARS PERCENT KERNELS FROM PRIME HUSKED EARS PERCENT KERNELS FROM DENTED HUSKED EARS PERCENT KERNELS, PRIME PLUS DENTED HUSKED EARS) PERCENT MOISTURE, KERNELS FROM PRIME PLUS DENTED EARS PERCENT MOISTURE. KERNELS FROM DENTED EARS MOISTURE PERCENTAGES.^ WET BASIS) COUNTRY GENTLEMAN IZ 20 22 24 26 28 DAYS AFTER 75% SILKING FIG. 11. RELATION BETWEEN MATURITY AND PERCENTAGES OF PRINCIPAL YIELD COMPONENTS: COUNTRY GENTLEMAN SWEET CORN, SERIES IV, 1932 Weight of Sorted Unhusked Ears per Acre The acre weights of sorted unhusked ears from Country Gentle- man (Table 11) and Narrow Grain Evergreen (Table 13) increased during the earlier stages of maturity, but the yields from Golden Bantam (Table 15) were contradictory, owing no doubt to the disturb- ing influence of Stewart's disease. In general, early increases in yield were followed later in many of the series by a tendency to decrease, but such decreases were significant only in isolated instances. The maximum weight of sorted unhusked ears occurred with mean moisture contents of 66.6, 68.3, and 66.4 percent in Country Gentle- man, Narrow Grain Evergreen, and Golden Bantam respectively (Table 17). These means are in fairly close agreement, and within 562 BULLETIN No. 432 M I si cu o Z H c/5 H * J2 u o I s n I OS Id - S 2 1 B ^- * o - * "O :S O -t~ t^. -o B B %e 5)d'*jd PB S M adod ^oi.e *o 3 ^r^tNO B . . .,, . . . ., .^ .- . 1 S O *O O -O >C O GO 00 OO O O CO "* ^ ^* l/> >O * C C 00 00 00 C tf - (*5 \O rs ^O (N O CM O (N >O **5 O fN O r* oo oo oo oo f*5 to oo O (SO 00 00 V) T C ^ C 00 ^ t- ^ * ^ . ^ .^ ^* O . OS -^ V c 10 -*O *O -O -O '^f (s ^H O CN CS S 00 O COiO^OO O O^ O-^ ^iO^ :S o :g :S .*. (S W 00") ntleman S ^ ^N , i ^* ^ ^* 00 **5 (NOO (SO 'C 00 O H I* 'c 3 CM O (N O f*5 to O O O O r- (SO o (SO(NO o o U c~ " oO oo 00 r r (so Tt , . . >> . n ."9 .'P . K >, a ^ a 1 i TI jlfjlllfl*l ?f^ mil 4 a.g'o.5'5 <._'S.S'5.S'o 'S.-S S._'o la s -M a a |a s co ON a. M Maximum percent Unhusked basis Moisture per Husked basis Moisture per 1937] YIELD COMPLEX OF SWEET CORN 563 3 C . . .(N - . -N (N 1 e 3 S -Jg _ ... -op -r- . -O -O o> & i B t> (N -t*- & ** O */5 O O O /5 g B Id s !!!!!! *!-*!!! *l*Ti O fe _ -. . ^i O 1 * OO'O 00 (N (N 1 00^*5 (NOO'*) -(N C C* fN 'I^(N (Nr^O^O^* a 3)0 'o<0* '" (5i^ S l> O ^O ^** O 00 O .sOQt>Of<. r^oo^fN^QO^ 1 x "f "t *O 'N X M O t B 3 ^^ O^O iJ)^^- >O>O -O>O O O >O i j cs>O O *O (N-O OO fSO CS -O (NOCNOrsio i SK :S3K :R SS S :8* SKSSRS i 'f :::::::: ::::: :::::: ' '"2 oi ' S"? 2 ' it practically equ iriety these were aximum, but the le grain not cons on unhusked-ear . the maximum bi Golden Bantam v; ighted total. L necessarily the n am style cut; who :entage calculated fc?*S "S ^^7 5 5"~1 *T* "T* ^^ i5 .**** '- ~ "* " ^ !* <5 Illlvl 564 BULLETIN No. 432 [April, varieties (Table 17) the -variation in moisture content was not nearly so great as the variation in number of sorted unhusked ears. The point where the weight of unhusked ears attained the maximum was appar- ently subject to environmental influences, however, and could not be considered very closely associated with a definite moisture content of kernels. Also, altho the period of maximum weight of unhusked ears corresponded fairly well with the period of maximum numbers of unhusked ears, these periods did not agree so well with the period 60 30 10 I I BASIS: SORTED UNHUSKED EARS PERCENT PRIME HUSKED EARS PERCENT DENTED HUSKED EARS PERCENT KERNELS FROM PRIME HUSKED EARS PERCENT KERNELS FROM DENTED HUSKED EARS PERCENT KERNELS, PRIME PLUS DENTED HUSKED EARS) PERCENT MOISTURE, KERNELS FROM PRIME PLUS DENTED EARS, WET BASIS NARROW GRAIN EVERGREEN I 19 20 21 22 23 24 25 26 27 28 29 30 31 DAYS AFTER 75% SILKING 32 33 34 35 36 37 FIG. 12. RELATION BETWEEN MATURITY AND PERCENTAGES OF PRINCIPAL YIELD COMPONENTS: NARROW GRAIN EVERGREEN SWEET CORN, SERIES I, 1931 of maximum yields of prime husked and prime plus dented husked ears (Table 17). This lack of agreement is important because it shows that yields (on the basis either of weight or of numbers) of unhusked ears may be unreliable as a measure of husked yields. Weights of sorted unhusked ears declined in the later stages of maturity (Tables 11, 13, and 15). Such declines must have been due to an actual decrease in weight arising from moisture losses, a de- crease which was undoubtedly greater than any possible increase ac- companying the deposition of food materials, which are laid down in the kernels alone. The calculated mean daily losses of moisture from cobs alone were very small (Tables 12, 14, and 16), but the mean daily 1937] YIELD COMPLEX OF SWEET CORN 565 losses of moisture from the kernels alone amounted to .96, 1.01, and 1.15 percent for Narrow Grain Evergreen, Country Gentleman, and Golden Bantam respectively. Furthermore the decrease in yield began at approximately the time of the appearance of dented ears in Country Gentleman and Narrow Grain Evergreen (Table 17), comparisons being made by means of the respective moisture contents. Dented ears lose moisture very rapidly (Tables 12, 14, and 16). PERCENT PRIME HUSKED EARS 1 D V PERCENT DENTED HUSKED PERCENT KERNELS FROM />/ PERCENT KERNELS FROM Dt PERCENT KERNELS. PRIME P >- PERCENT A """""* ^ ARS t/ME HUSKED EARS \ NTED HUSKED EARS MS DENTED HUSKED EARS ] IOISTURE, KERNELS FROM PR WET BASIS ^ A PERCENT MOISTURE, KE BASIS: SORTED UNHUSKEO EARS IMC PLUS DEN VNELS FROM D r D EARS. ENTED EARS, ^ WET BASIS T / , ^ ^ \ / > \ ^ _s* "^^ ' ^ , ' ^ s. -, m -* z^ /- NARROW GRAIN EVE .RGREEN ' or -* 90 70 ?.50 20 22 24 DAYS AFTER 75% SILKING 26 FIG. 13. RELATION BETWEEN MATURITY AND PERCENTAGES OF PRINCIPAL YIELD COMPONENTS: NARROW GRAIN EVERGREEN SWEET CORN, SERIES IV, 1932 Number of Prime Husked Ears per Acre In Country Gentleman and Narrow Grain Evergreen varieties, the number of prime husked ears per acre increased, usually, in successive pickings until either just before or just after the appearance of dented ears (Tables 11 and 13). The decrease in number was thereafter very rapid and was accompanied by a correspondingly rapid increase in number of dented ears. The mean moisture content of the kernels at the period when the number of prime husked ears reached the maximum was 70.6 and 69.4 percent respectively for Country Gentle- man and Narrow Grain Evergreen (Table 17). This was a remarkably close agreement. 566 BULLETIN No. 432 [April, According to data presented in Table 17, three yield components in Country Gentleman and Narrow Grain Evergreen number, weight, and percentage of prime husked ears reached maxima simultaneously. There is thus a very close relation among the yield components of prime husked ears, as contrasted with the much looser relationship among yield components of unhusked ears. This is a matter of con- siderable importance, for it means that when corn is sold on the basis 80 60 040 20 O PERCENT HUSKED CANNING EARS, BASIS SORTED UNHUSKED EARS A PERCENT KERNELS CUT OFF HUSKED CANNING EARS, BASIS SORTED UNHUSKED EARS O PERCENT MOISTURE ON THE WET BASIS - KERNELS CUT FROM CANNING EARS 19 23 DAYS AFTER 75% SILKING 25 FIG. 14. RELATION BETWEEN MATURITY AND PERCENTAGES OF PRINCIPAL YIELD COMPONENTS: GOLDEN BANTAM SWEET CORN, SERIES I, 1932 of husked ears, the grower, by harvesting at the proper time, obtains the maximum yield of prime-quality ears and the canner receives at the same time the highest possible percentage of prime husked ears in each ton of unhusked corn. The total number of husked ears (dented plus prime) increased for a time after dented ears appeared in each series, except in Series V of Narrow Grain Evergreen, where the total number reached the maxi- mum just before the appearance of dented ears (Tables 11 and 13). In all series except Series I of Country Gentleman, Series II of Narrow- Grain Evergreen, and Series I and II of Golden Bantam, in all of which the maximum total number of husked ears occurred at the last picking date, the total numbers of husked ears in the later pickings showed declines, some of which were statistically significant. As the date of maximum total number of husked ears coincided (except in 1937] YIELD COMPLEX OF SWEET CORN 567 Series V of Country Gentleman, Series IV of Narrow Grain Ever- green, and Series I of Golden Bantam) with the date of the maximum number of sorted unhusked ears, it is apparent that the decrease in the total number of husked ears was the direct result of the decrease in the number of sorted unhusked ears, for which no satisfactory explana- tion is available. Weight of Prime Husked Ears per Acre Acre-yields of prime husked ears from Country Gentleman and Narrow Grain Evergreen varieties, measured by weight, increased consistently (Tables 11 and 13) as maturity advanced to the maximum, at which time, according to data in Table 17, the kernel moisture content ranged from 68.4 to 72.0 percent in Country Gentleman and from 66.4 to 71.8 percent in Narrow Grain Evergreen, the respective means being 70.6 and 69.4 percent. As the desideratum in harvesting sweet corn for canning purposes is prime husked ears, it follows that the proper time to harvest these two varieties in order to obtain maxi- mum yields of best-quality ears per acre occurs when the moisture in the kernels reaches a mean of 70.6 percent in Country Gentleman and 69.4 percent in Narrow Grain Evergreen. This 70-percent-moisture stage virtually coincided with the first appearance of dented ears, at which time the percentages of moisture in the kernels were 69.6 per- cent in Country Gentleman and 70.1 percent in Narrow Grain Ever- green (Table 17). Thus, ordinarily, the proper time to harvest sweet corn for canning cream style is just when dented ears first appear. Sometimes when sweet corn is packed whole-grain style, a definite effort is made to harvest a little earlier than for cream-style, in order to secure a more tender pericarp. While this practice is apparently justified by the results of the penetration tests (pages 528 to 532), it entails a material reduction in yield which must be balanced against the gain in quality. There is some reason for believing that earlier harvesting in order to obtain better quality may be more justifiable with Country Gentleman than with Narrow Grain Evergreen. Ac- cording to the data from the individual-ear experiments (Tables 1 and 2) and the data given in Tables 12 and 14, the loss of moisture in Country Gentleman is more rapid than in Narrow Grain Evergreen and correspondingly the rate of maturity is more rapid in the former than in the latter, a difference that agrees with the experience of Illinois canners. Coupling this difference in the loss of moisture with the fact that Narrow Grain Evergreen is not only consistently tougher than Country 568 BULLETIN No. 432 [April, Gentleman, but also has a slower rate of increase in toughness of peri- carp (Figs. 5 and 6), there exists a definite basis for the premature harvesting of Country Gentleman when packing whole-grain style. By harvesting Country Gentleman only a day or two earlier than the date of maximum maturity a substantial gain in tenderness is obtained with only a moderate loss in yield. On the other hand the toughness of pericarp in Narrow Grain Evergreen increases with maturity at a relatively slow rate and premature harvesting therefore entails a con- siderable loss of yield with only a slight compensating gain in tender- ness. In other words, the yield losses due to premature harvesting in the two varieties are about the same; it is gain in tenderness which differs markedly when they are prematurely harvested. Percentage of Prime Husked Ears The percentage of prime husked ears increased in relation to ma- turity in the same manner as did acre-yields, the maxima being reached (Table 17) at the same time in each series. This concurrence is highly significant from the economic point of view alone, because it exposes one of the fallacies involved in calculating yields and in purchasing sweet corn on the basis of unhusked ears. When sweet corn is purchased on the unhusked basis, growers who realize that sweet corn gains weight rapidly during the maturation period often delay harvesting as long as possible. From the standpoint of weight alone, delay may be profitable (pages 561 to 565) ; but when both weight and quality are considered, the delay may often be detrimental. That the purchase of sweet corn on the basis of husked ears, with dockage for a more than nominal ratio of dented ears, is a fair method of purchase is evident from the results of these experiments, in which the period of maximum yield per acre and of maximum ratio of prime husked ears coincided exactly (Table 17). Total Weight of Prime Plus Dented Husked Ears Altho the weight of prime husked ears per acre declined very rapidly after reaching the maximum point, the total weight of prime plus dented husked ears continued to increase for some time longer (Tables 11 and 13). This continued increase was in keeping with the tendency for yields of sorted unhusked ears to increase beyond the time when yields of prime husked ears reached the maximum. Total yields of husked ears, however, in all series except Country Gentle- man II, continued to increase for some time after the maximum yield of unhusked ears was reached (Table 17). The mean weighted moisture contents of the kernels at the time of 1937} YIELD COMPLEX OF SWEET CORN 569 maximum yields of sorted unhusked ears were 66.6 percent in Country Gentleman, 68.3 percent in Narrow Grain Evergreen, and 66.4 percent in Golden Bantam (Table 17) ; and at the time of maximum yield of prime plus dented ears (total sorted husked ears of Golden Bantam) the respective moisture contents were reduced to 64.4, 63.0, and 65.2 percent. The correspondence between date of maximum yield and moisture content does not, however, seem to be fixed definitely. There was considerable variation in the moisture contents among the several series, the range for Country Gentleman being 57.9 to 67.8 percent, for Narrow Grain Evergreen, 60.6 to 65.4 percent, and from 60.5 to 70.0 percent in Golden Bantam. Percentage of Prime Plus Dented Husked Ears The percentages of prime plus dented husked ears computed on the basis of the sorted unhusked ears tended to reach the maximum at an even earlier date than did the yield on the basis of weight (Table 17). In Country Gentleman this point apparently lay either at the extreme limit of the picking period covered, or at some point beyond ; but in Narrow Grain Evergreen Series I, II, and III it lay within the period covered. This period of maximum percentage is of no particular importance, because sweet corn is inedible at that stage. It does show, however, that the stages of maturation succeed each other closely, and that each one is significant in a different way. Weight of Kernels Cut From Prime Ears From the standpoint of the canning industry the yield of cut corn from prime ears is more important than any other yield component, because the number of cases of canned corn per acre depends directly upon it. Both in Country Gentleman and Narrow Grain Evergreen the peak yield of prime ears and of cut corn from prime ears very nearly coincided (Table 17). According to these data, the harvest period that will yield the maximum edible product in either of these varieties is when the yield of prime husked ears reaches the peak i.e., when dented ears begin to appear in nominal quantities. Since Golden Bantam prime ears cannot be distinguished visually from those past prime there is no way of determining when the maximum production of prime ears is reached other than by moisture tests. Percentage of Kernels Cut From Prime Ears The ratio of cut corn from prime ears to sorted unhusked ears, expressed as percentages (Tables 12, 14, and 16), increased signifi- cantly as maturity advanced and reached the maximum ( Table 17) at 570 BULLETIN No. 432 [April, the same time as the acre-yield. The relationship between this and other yield components is shown in Figs. 10, 11, 12, and 13 and is discussed on pages 560 to 565. The maximum ratios of cut kernels for the four series of Country Gentleman that were cut whole-grain style averaged 35.8 percent (Table 12, Series I to IV), and the same was true for Narrow Grain Evergreen (Table 14). Means such as these support the conclusion that sweet corn should be harvested at the time when moisture content reaches 70 percent or, by inspection, when dented ears begin to appear. The percentages of cut kernels from prime ears were also calculated on the basis of weights of prime husked ears equaling 100 percent. The maximum ratios (Table 17) were reached at a much later stage in virtually every series, the moisture averaging 64.5 percent in Country Gentleman, 64.4 percent in Narrow Grain Evergreen, and 67.2 percent in Golden Bantam. The maximum was attained at a much lower mois- ture content by this method of computation because the weights of the cobs followed a trend different from that followed by the weights of the prime ears. The trend followed by weights of cobs may be deter- mined from Tables 11, 13, and 15 by making suitable calculations. 8 Weight of Kernels Cut From Prime Plus Dented Husked Ears The period of maximum yields of usable ears came at practically the same time as the period of maximum yields of kernels cut from usable ears (Table 17). The moisture content of the kernels from prime ears varied at this time from 63 to 64 percent, and the weighted total percentage of moisture averaged 64.2 percent in Country Gentle- man, 62.8 in Narrow Grain Evergreen, and 65.2 percent in Golden Bantam. Percentage of Kernels Cut From Prime Plus Dented Husked Ears The maximum percentage of kernels cut from prime plus dented ears, calculated on the basis of the weight of the unhusked corn equal- ling 100 percent, was obtained at approximately the same weighted total-moisture content as the maximum yield namely, 62.7 percent in Country Gentleman and 62.2 percent in Narrow Grain Evergreen (Table 17). In Golden Bantam, however, the moisture at this period was 67.2 percent, which was slightly higher than at the period of maximum yield, when it was 65.2 percent (Table 17). 'Subtracting the weights of the prime kernels from the weights of the prime husked ears gives the weights of the cobs. 1937] YIELD COMPLEX OF SWEET CORN 571 Time of Appearance of Dented Ears The first appearance of dented ears in Country Gentleman and Narrow Grain Evergreen occurred when the moisture content averaged 69.6 and 70.1 percent respectively (Table 17). This average moisture content was practically the same as that (70 percent) found when the maximum yield of prime husked ears was obtained (page 567), and consequently the first appearance of dented ears may be accepted as an accurate indication of the appropriate time to harvest sweet corn of the types used in these experiments. As a matter of general practice, however, sweet corn is usually harvested for canning at the time when dented ears are becoming more noticeable. Packers find that a fancy cream-style grade may be canned when, by weight, dented ears do not exceed 5 percent of the unhusked ears. The weighted mean moisture content of the kernels at this period was 68.6 percent in Country Gentleman and 67.4 in Narrow Grain Evergreen (Table 17). The moisture of the kernels from dented ears was considerably lower, averaging respectively 65.0 and 65.2 per- cent. The actual mean percentage of dented ears was 3.9 and 4.2 per- cent, respectively (Table 17). Yields of Culls, Green Ears, and Husks Unhusked culls in several of the series (Tables 11, 13, and 15) showed slight increases in weight for a short time during the matura- tion period, but in all cases the weights of culls definitely declined during the later stages of maturity. This decline was due to the fact that many of the unhusked culls were actually immature ears, and in the latter part of the maturation period many of these ears had matured and were included with the usable ears. The weights of husked culls also declined in the later stages of maturity (Tables 11, 13, and 15), probably, as stated above, because many of the so-called culls were actually undeveloped ears which should be classed as "green ears." The distinction is difficult to make and, furthermore, unimportant. The percentage of husked culls de- clined in the same manner as the weights (Tables 12, 14, and 16). Weights and percentages of husked green ears declined rapidly as the maturation period advanced. In accordance with expectation, green ears did not appear at all after the first few pickings (Tables 11, 13, 15, and 12, 14, 16). The husk ratios (Tables 12, 14, and 16) declined significantly as maturity advanced. This decline in percentage of husks (by weight) was, however, only a matter of mathematical function, as, with one 572 BULLETIN No. 432 [April, exception, there was no significant decline in mean weights of husks from either upper or lower ears (Tables 8, 9, and 10). That the husks of the individual ears actually lose some weight as they advance in maturity is shown in Figs. 3A and 4A, but the loss is small and apparently due to the loss of moisture (Figs. 3C and 4C). The real reason for the decrease in the husk ratios (Tables 12, 14, and 16) was the large increase in weights of kernels as maturity advanced, which in turn produced increases in weights of unhusked ears and caused percentages of husk weights to be proportionately smaller. Altho ear weights declined later on, this decrease was not sufficient to offset the decrease in the husk ratios. COMPARISON OF TWO STYLES OF CUT AND THE EFFICIENCIES OF CUTTERS In all the experiments the corn was cut whole-grain style, except certain pickings in Series V of both Country Gentleman and Narrow Grain Evergreen, in which it was cut cream style as soon as it had reached the proper stage of maturity for that type of pack. This change in style of cutting when maturity advances too far for whole-grain packing is the usual commercial practice, for altho ears may have be- come too tough for the whole-grain style of pack, they may still be at the right stage of maturity to pack a high-grade cream-style product. Except in Series V, therefore, the entire work on cut corn in these experiments might be open to criticism because the whole-grain style of cut was used thruout. Consequently comparisons of the two types of cut, using several models of cutters, were made. In this experiment small fields of each variety Country Gentleman, Narrow Grain Evergreen, and Golden Bantam were harvested at the prime canning stage, the moisture content of the kernels averaging approximately 70 percent and only a very few dented ears appearing. The ears were husked by machine and washed, sorted, and trimmed, according to the usual canning-factory procedure, in the canning laboratory of the Illinois Agricultural Experiment Station. The trimmed corn was then weighed out in samples of 20 pounds each and each sample was cut separately. The results (Table 18) are the means of numerous replications. The data on cuttings by the Burpee hand cutter were included in Table 18 because this cutter was used in 1931. However, before the data were entered in the various tables, the 1931 results were converted into values corresponding to those obtained from the Peerless whole- 1937} YIELD COMPLEX OF SWEET CORN 573 TABLE 18. CONVERSION TABLE: MEAN PERCENTAGES OF CORN CUT IN DIFFERENT STYLES BY DIFFERENT CUTTERS, AND INDEX NUMBERS ENABLING YIELD FROM ONE STYLE OF CUT AND CUTTER TO BE CONVERTED TO YIELDS FROM OTHER STYLES OF CUT AND CUTTERS Mean Type of cutter Style of cut percentage of cut corn (liusked Dev. P.E. Index No. Number of replications of test ears = 100) Narrow Grain Evergreen Burpee hand 65.75 10 92 2 131 75 21 Peerless Whole grain 58.62 .35 23 1 117 45 31 Peerless combination .... Sprague No. 5 Cream Cream 54.27 .26 49.90 .14 14.8 108.75 100 00 21 32 Country Gentleman Burpee hand 6* 66 32 41 6 no so 24 Peerless 57 00 + 27 26 2 115 86 24 Peerless combination ... Sprague No. 5 Cream 49.30 4- .14 48 78 16 2.4 101.07 100 00 24 24 Golden Bantam Peerless 56 00 32 4 104 09 11 Sprague No. 5 . 53 80 45 100 00 11 grain cutter. The results obtained during the two years are thus directly comparable. The probable errors of all the tests reported in Table 18 were very small, indicating that when once any of the cutters is adjusted it continues to cut uniformly. In presenting the results, the Sprague No. 5 cream-style cutter was used as the standard with which to compare the style of cut and efficiency of each of the other cutters. The Sprague No. 5 seems to have been less efficient than the other cutters, including the combination type, and the differences are significant in all but one instance. On the basis of these tests a ton of prime canning ears yields the most cut corn when cut whole-grain style. By converting the data for kernels in Series V (Tables 12 to 14) from cream-style values to whole-grain values thru the use of the factors in Table 18, the results of Series IV and V become similar, as Series IV is a replicate of Series V in all respects except style of cut used. Accordingly the authors conclude that the data on cut corn in this paper may, without altering the conclusions, be converted at will from values of one style of cut to values of another by using the index numbers in Table 18. 574 BULLETIN No. 432 [April, MOISTURE CONTENT OF THE COBS The decline in the moisture content of the cobs during advancing maturity paralleled rather closely the decline in the moisture content of the cut corn but remained at a somewhat higher level (Tables 12, 14, and 16). Since the moisture content of the cobs declined progressively, it can apparently be used as a measure of maturity. By methods used in these experiments, however, moisture content was more laboriously measured in cobs than in kernels. Furthermore moisture content of cobs, as a measure of maturity, has the disadvantage of being only an indirect indication of the maturity of the kernels. SILK EMERGENCE AS A MEASURE OF MATURATION Ability to estimate the maturity of an entire field of sweet corn by means of small but representative samples would be distinctly valu- able to grower and canner. The experiments which have been dis- cussed here suggest such a possibility on the basis of (1) the similar- ity between the respective moisture contents of single tagged ears whose date of silk emergence is known and samples taken from a mass popu- lation ; and (2) the validity of predicting maturity of a large population on the basis of silk counts from small samples. The results graphed in Figs. 5 and 6 show that the percentage of moisture was an accurate index of maturity, and that individual ear samples of known age as determined by silk emergence were almost identical with aliquot samples taken from a large population whose age in turn was known only by means of silk counts taken on a few small plots scattered at intervals thruout the field and then only on the basis of 75 percent of the theoretical stand. The evidence in Figs. 5 and 6 taken in 1931 is supported by the 1932 results. From a com- parison of the moisture contents of individual ears in Tables 1, 2, and 3 with the equivalent weighted means of the mass populations in Tables 12, 14, and 16 it is obvious that the two if plotted would show very similar trends. It would also be found that the plotted courses of the weights and percentages of cut corn and sorted husked ears, as well as of the weights of unhusked ears in relation to maturity, would resemble each other very closely. These relationships are based on certain theories about silk emer- gence. Therefore, before the relationships can be accepted as evidence it will be necessary to prove that the theories about silk emergence are correct. 1937] YIELD COMPLEX OF SWEET CORN 575 Silk emergence has been shown by Culpepper and Magoon 8 ' 14 * to be a fairly reliable basis for predicting the maturity of sweet corn. The same investigators have shown that the peak of the silking period (which they call the "mid-silking period") is an even better point from which to predict maturity, and that the length of the silking period corresponds rather closely to the length of the period of edibility. But with these facts known it is still essential to know how to utilize this knowledge as a practical basis for predicting maturity. For example, what is the mid-silking period and how can it be determined? Is the mid-silking period the peak point in the production of all the silks, including silks of suckers, or is it simply the peak of silk emer- gence on those shoots most likely to produce an ear? Obviously it is quite impossible to make any such separation because suckers and main stalks are likely to be almost identical in appearance. What is needed is a method for determining the mid-silking point easily and expeditiously. As stated above, maturity both of the individual ears (Tables 1, 2 and 3) and of the mass populations (Tables 11, 13 and 15) was measured from the time when silk emergence reached 75 percent of the theoretical stand. This was assumed to be the "mid-silking period." It is important to show that the two points were actually equivalent. The 75-percent point has worked well in experiments with sweet corn at the Illinois Station during the past ten years. The location of the two points in the present experiments may be made from the com- plete silking records appearing under the column headed "Total number unhusked ears" in Tables 8, 9, and 10, which is equivalent to the silks actually counted. The calculated results appear in Table 19. The actual median number of silks is shown in Table 19, but the selection of the date when the median number appeared was somewhat arbitrary owing to the fact that counts were made only at 24-hour intervals. However, August 10 and July 26 were selected as the dates on which 50 percent (the median) of the total number of silks had emerged. The actual number of silks appearing the day before and the day after are also shown. Thus in Country Gentleman the total silk emergence on August 9 was 271, August 10, 382, and August 11, 444. Clearly August 10, the date on which silk emergence totaled 382, was nearest the true median of 357. Similarly in Narrow Grain Evergreen the total emergence on August 10 was 386, which was closer to the median of 434 than was 498, the number of silks appearing August 11. In Golden Bantam the median total silk emergence was 250 and on July 26, 225 silks had actually emerged. Table 19 shows further that the number of silks corresponding to 576 BULLETIN No. 432 [April, 75 percent of the theoretical stand was 351 silks in Country Gentleman and Narrow Grain Evergreen and 324 in Golden Bantam. Since 382 and 386 silks were actually counted on August 10 in the Country Gentleman and Narrow Grain varieties respectively, this date would be the estimated median point, since both 382 and 386 were very close to the theoretical 75-percent point, 351. With respect to Golden Bantam, July 27 would be selected instead of July 26, because the total count on the 27th reached 330, which was very close to the theoretical 75-percent point, 324. Thus in Golden Bantam there appeared a dis- crepancy of one day which, however, was not serious. The question may be raised why 75 percent of the theoretical per- fect stand was selected as being equivalent to the median point. In the first place, allowance must be made for the fact that the actual stands are much less than the theoretical, altho this difference is count- erbalanced to some extent by the possible influence which missing plants and hills would have on the number of shoots produced by the remaining plants. In the second place, allowance was made for only one shoot per stalk, whereas according to the data in Tables 8, 9, and 10, 32.0, 33.4, and 29.8 percent, respectively, of the total silks of Country Gentleman, Narrow Grain Evergreen, and Golden Bantam emerged from lower shoots. The relationship between the 75-percent stage and mid-silking, when the foregoing conditions are taken into consideration, are as follows: If the theoretical stand is 2 plants per hill, the total theoretical stand for 500 hills would be 1,000 plants; and at one shoot per stalk the total theoretical number of shoots would be 1,000. Consequently the 75-per- cent point of silk emergence would be reached when 750 silks appear. But about one-third of these, or 250 silks, would be silks on lower shoots, according to the percentages given in the preceding paragraph, and consequently only about 500 the median point of the 1,000 plants would be silks on upper shoots. Therefore, 75-percent silk emergence of the theoretical stand equals the theoretical median silking of the upper ears, and since lower ears show only a slight lag it is usually also the mid-silking point of these. If the actual stand does not deviate too much from the theoretical, the method is satisfactory, as shown in Table 19. The data presented in Table 19 show that determining the theor- etical median silking point is for all practical purposes just as accurate as keeping a tagged record of all silks as they emerge. Since, in the mass-population experiments reported in Tables 11 to 16 (pages 548 to 559) this method proved to be accurate when applied to a few plots 1937] YIELD COMPLEX OF SWEET CORN 577 TABLE 19. ACTUAL MEDIAN POINT OF SILK EMERGENCE AND MEDIAN POINT DETERMINED BY CALCULATION Country Gentleman Narrow Grain Evergreen Golden Bantam Actual median date* Aug. 10 Aug. 10 July 26 Silks emerged day before 271 267 118 Silks emerged on date selected 382 386 225 Silks emerged day after 444 498 330 Date of silk emergence of 75 percent of theoretical stand Aug. 10 Aug. 10 July 27 50 percent of total actual silk emergence 357 434 250 75 percent of theoretical stand 1 ' 351 351 324 Fifty percent of actual number of silks, according to Tables 8, 9. and 10. b The theoretical stand of Country Gentleman and Narrow Grain Evergreen was 468 plants, and of Golden Bantam 432 plants. scattered at random thru the field, it is obviously practical to use similar small samples in order to determine the maturity of any large population. GENERAL DISCUSSION AND ECONOMIC ASPECTS OF FINDINGS IN PART I It is important for both grower and canner to understand as far as possible the trends of the complex forces which are at work during the maturation period of sweet corn. Such an understanding is just as important to the grower primarily interested in raising as much corn per acre as he can, as it is to the canner who must always keep in mind the quality of the corn. In the preceding text the relation of the principal yield components to maturity has been discussed in detail. The purpose in this section is to show how to apply the information gained. It is necessary to bear in mind, however, that these investigations deal only with time of maturity and very little attention has been paid to the relations between the yield components themselves. Such a study of yield components has been made and is reported in Part II of this bulletin. There are three methods of measuring maturity in green sweet corn. The first of these is the method used by Culpepper and Magoon, 6 * who showed that the hull or pericarp becomes increasingly tougher with age. The second is to determine the moisture content of the kernels, a method shown by the present studies to be related closely to toughness of hull. A third method is the alcohol insoluble-residue test recently proposed by Jenkins. 13 * Whatever the respective merits of these tests, the fact remains that they are of little value to the canner or grower because they are slow 578 BULLETIN No. 432 [April, and cumbersome. The test for toughness may be an exception, but no corn canner in Illinois uses it consistently. The thumb-nail test dis- cussed by Appleman 1 * is used to a considerable extent, but as individual ears vary widely, it is difficult to estimate the maturity of a whole field of corn by testing a few ears with the thumb nail. A better and more accurate test, and moreover one which supple- ments the thumb-nail test very nicely, is to determine the ratio of dented ears in a sample of corn. This test can be made in a few minutes, and that it is accurate is shown by the present investigations. Several of the most experienced canners in Illinois agree with this conclusion. They have found that fancy Country Gentleman corn cannot be packed from lots which contain, by weight, appreciably more than 5 percent dented ears. Ten percent seems to be the maximum. Some canners regard the appearance of more than a few dented ears as being detrimental to quality. The same conclusions have been reached with respect to Narrow Grain Evergreen, altho it should be understood that a strictly fancy grade of this variety cannot be packed at any stage of maturity because, as shown in the text, it has a much tougher hull than Country Gentleman at all times during the canning period. The rapid and accurate determination of maturity is absolutely essential for the purchase of sweet corn on a grade basis. To the grower the fact that yields increase as maturity advances is superficially of more direct importance than quality. The grower gen- erally believes that the longer he allows his crop to remain in the field the greater will be the total weight of ears harvested. The investiga- tions reported here deal with this point in considerable detail. It is true that the total weights of both unhusked and husked ears and of kernels increase in relation to maturity, but not indefinitely, as many growers suppose. The sequence in the maturity of the several yield components must be held firmly in mind in order to evaluate the subsequent discussion. The order in which a field of sweet corn matures (Table 17), as meas- ured by the moisture content at the time when the yield of each com- ponent reaches the maximum, is, according to these investigations, as follows: 1. Number, weight, and percentage of prime husked ears, which is followed very closely by 2. Weight and percentage of prime cut kernels, which coincides with the first appearance of dented ears. 3. Number and weight of sorted unhusked ears. This sequence holds true for Country Gentleman and Narrow 1937] YIELD COMPLEX OF SWEET CORN 579 Grain Evergreen, but in Golden Bantam, using Series II as the basis, the weight and percentage of cut kernels comes first and the weights of the sorted unhusked and sorted husked ears come together and are second. The fact that the sorted unhusked ears reach their maximum weights later than the other components is a point of major importance. It is not only the source of most of the disagreements between canners and contract growers, but also forms the chief argument against the practice of writing contracts for the buying of sweet corn on the basis of unhusked ears. On the basis of these experiments there was only one series out of ten (Country Gentleman Series II) where the weights of the unhusked ears showed a significant gain after the prime husked ears had reached their maximum weights. On the other hand, in nine out of the ten series (excluding Bantam) the maximum weights of sorted unhusked ears came after the maximum weights of prime husked ears, but the increases were not significant. In other words, altho statistically there is no particular gain for the grower to be had by delaying harvest, yet the drift is very strong in that direction. In nine out of the ten series that tendency was shown ; and in addition it is supported by popular belief. Since prime husked ears reach maturity first in the sequence, it is obvious that if a canner wants a good quality of raw product he must contract on the basis of husked ears or cut kernels, either of these bases permitting maturity sequences to be taken into consideration. There will then be no temptation for the grower to delay harvest, especially if the dockage becomes increasingly heavier as the percent- age of overmature (dented) ears increases. Some such basis for buy- ing is almost a necessity if tender whole-grain style of corn is to be packed. The experiments show clearly that if tenderness is desired, harvest must be early, because it is out of the question to try to separate ears from the sorting belt into "whole-grain" maturity and "cream-style" maturity. A given lot must be either one or the other; it cannot be both unless the canner uses a strain w r hich matures very unevenly. It is true, of course, that early harvesting pays much better with a rapidly maturing variety than with one which matures slowly. Thus in Country Gentleman maturity is rapid and the increase in toughness occurs at a uniform rate. Early harvesting sacrifices yield but gives a large increase in tenderness. In the slower maturing Narrow Grain Evergreen toughness increases at a variable rate and earlier harvest does not appreciably increase tenderness. Probably this is the reason 580 BULLETIN No. 432 [April, why rapidly maturing varieties are usually the ones which are canned whole-grain style. Any proposal to contract sweet corn on the basis of husked-ear or cut-kernel weights is not particularly welcomed either by the grower or by the canner. The grower is apt to be suspicious of the method and the canner objects to it because it involves sampling, which is expen- sive. However, sampling for this purpose is not any more expensive than the sampling necessary in buying cannery tomatoes on official grades. In addition, if the canner wishes to buy sweet corn on official grades, sampling is a necessary part of the procedure. It is, however, rather difficult to buy sweet corn on a graded basis ; and that this difficulty is recognized is indicated by the fact that the official grades are still experimental. Buying according to grades is done either on the net (husked) basis, or on the basis of cut kernels. In either case a sample must be taken, usually 50 pounds of unhusked corn from each load. In grading the corn the canner must first know what constitutes a normal sample. A normal sample differs with each variety ; but when it is known for a given variety, prices can be established on the basis of (1) maturity as determined by the presence of immature and dented ears, and (2) percentage of husked prime ears or of cut corn. The question is, what percentage of prime ears or of cut corn should a normal sample of each particular variety contain? It is essen- tial to establish this point, because the normal price must be based on the normal sample, and any dockages or premiums must be calculated accordingly. Some canners in Illinois who are buying on the husked- ear basis have established, by means of averages from a large number of samples, the fact that Narrow Grain Evergreen and Country Gen- tleman will husk out normally 60 percent prime ears by weight. Bases calculated from the experiments reported here by averaging the means of the maximum percentages of prime husked ears are 62.6 percent for Country Gentleman (Table 12), 64.6 percent for Narrow Grain Ever- green (Table 14) and 49.1 percent for Golden Bantam (Table 16). Thus it is evident that the Illinois base of 60 percent is practically correct. Too high a base works a gross injustice on the grower, whose corn is then expected to meet an unreasonably high standard. Conversely a base which is too low discriminates against the buyer. When, for example corn is contracted at $12 a ton unhusked, at a base of 60 per- cent, the canner expects to get 1,200 pounds (60 percent of 2,000 pounds) of prime husked ears for $12. The grower who brings in corn testing 55 percent prime husked ears receives payment according to the 1937] YIELD COMPLEX OF SWEET CORN 581 proportion 60:55 = 12: X, or $11 a ton. If the base were 50 percent he would receive $13.20 a ton (50:55 = 12: X), and if the base were 70 percent he would receive $9.43 a ton (70:55 =- 12: X). The whole matter is complicated by dockages for overmature (dented) ears, poorly filled ears, and immature ears. As mentioned previously, some canners believe that lots containing more than 5 percent of dented ears should be rejected entirely, but others think a sliding scale of values is better. In the latter event they would pay the base price for prime ears out of lots containing less than 5 percent of dented ears and assess increasingly high dockages against this base price for increasingly high percentages of dented ears. It is obvious, of course, that such a system would not work with varieties such as Golden Bantam, which does not have a characteristic denting. PART II: CORRELATION ANALYSIS OF THE YIELD COMPLEX IN SWEET CORN The investigations reported in Part I were limited to the changes which occur in yield components of sweet corn with respect to the advancing maturity of the crop. Considerable evidence of the existence of a close relationship between the yield components themselves was found, but no attempt was made to measure this relationship definitely, owing to the rapid changes which characterize the ripening period. In the experimental work reported in Part II the effect of advancing maturity (time) was eliminated by harvesting all the plots in the sev- eral series at identical stages of maturity. By means of this procedure the relations between yield components could be measured in terms of correlation and regression coefficients. EXPERIMENTAL METHODS Two varieties of sweet corn, Country Gentleman and Narrow Grain Evergreen, were used in these experiments. The strains used were the best of those selected by ear-row methods at the Illinois Station and were identical with the strains used commercially by canners in Illinois. The experimental data were secured from six series of experiments entirely distinct from those reported in Part I. Four of the six series were planted with Country Gentleman and the remaining two with Narrow Grain Evergreen. Nos. I, II, and V were grown in 1931, and Nos. Ill, IV, and VI in 1932. 582 BULLETIN No. 432 [April, Series I consisted of 268 plots of 10 hills each, planted 38 by 38 inches. These plots were planted at uniform intervals over a large field and served primarily as check plots in a yield trial of numerous first-generation sweet-corn hybrids of the same variety. Series II consisted of 70 plots of 44 hills each, planted 42 by 42 inches and located at uniform intervals in two 10-acre fields where rotation- fertilizer studies of sweet corn were being made. The 70 plots were unfertilized checks, half of which were grown the first year and half the second year after red clover. The plots were actually the two center rows of larger plots consisting of 8-by-22 net hills, each without border. Thus there was absolutely no border effect in Series II. Series III consisted of 68 plots, each 44 hills in size, forming part of the same rotation-fertilizer experiment as Series II. The only difference between Series II and III was that 33 of the 68 plots had moved along to another field in order to follow the rotation. The remaining 35 plots were identical with Series II but were planted a year later. Series IV consisted of 283 plots, each 10 hills in size, planted 38 by 38 inches. These plots were planted as checks at uniform intervals in a trial of one variety of first-generation hybrids. Series I and IV differed only as to the year when they were planted. Series V consisted of 110 plots, identical with Series I in all respects except variety. Series VI consisted of 91 plots, identical with Series V except that it was planted a year later. All the series were harvested at the prime canning stage, which coincides with the nominal appearance of dented ears. Thus, so far as maturity was concerned, the entire experiment was harvested at the same stage. In order to eliminate errors in snapping, all the shoots showing silks were harvested and sorted carefully into two classes unhusked culls and sorted unhusked ears. The latter class was considered to be equivalent to sweet corn as usually delivered to canning plants in Illi- nois. The other sorting methods were the same as those used in the experiments reported in Part I, page 513. All the corn was cut whole-grain style. It was not washed, nor was it trimmed to remove defective spots before cutting. Two old-style Peerless huskers were used to husk the ears and were set in such a manner that the shanks were cut off as close to the butts as possible without cutting off any portion of the butt itself. 1937] YIELD COMPLEX OF SWEET CORN 583 STATISTICAL METHODS Formulas and sources for the particular methods used in the cal- culations involved in this study were as follows: The total correlations were calculated according to the method described by Hottes and Huelsen. 12 * Significance of the correlation coefficients was determined by the probable errors according to the formula .6745 (1 - r 2 ) P.E. = V n The significance of the differences between the correlation coefficients was calculated according to Fisher's 9 * method (p. 182, ex. 32) ; and the weighted means of the correlation coefficients were determined by means of Fisher's example (p. 183, ex. 33). The regression coefficients and the regression lines were most conveniently calculated according to the method described by Rietz. 18 * The standard errors of the regression coefficients were calculated according to the formula S.E. = ay \/ n given by Burgess. 4 * The standard errors of the weighted mean re- gression coefficients were calculated according to the same formula as used for a weighted mean of means in Part I, page 514, standard errors being substituted for the probable errors in the formula. INTERPRETING THE CORRELATION COEFFICIENTS The coefficients presented in Tables 20 and 21 measure the degree of association between certain of the more important yield components. As each of these studies is confined to a highly selected variety, the correlations are probably physiological rather than genetic. Since, in addition, the observations are confined to the attributes of yield, the question may be raised as to how far the mathematical analysis may be carried and still avoid a nonsense interpretation from the biological standpoint. After giving the matter careful consideration the authors believe that partial correlations may not properly be used, for the simple reason that from the biological viewpoint it is rather doubtful whether the influence of one or more of the components can be stabil- ized or eliminated. For instance, in determining the first-order partial correlation between the weight of prime husked ears and the weight 584 BULLETIN No. 432 [April, of prime cut corn, it is doubtful whether the influence of cobs can be stabilized, because cut corn and cobs are simply parts of the prime ear, and it is hardly reasonable to suppose that the physiological influence of a fundamental part of the whole can be held constant in order to measure the association of another part with the whole. The authors believe that similar criticisms may be leveled against the use of multiple correlations, and they have therefore used only the simple or total correlations. EXPERIMENTAL RESULTS The correlations for the varieties Country Gentleman and Narrow Grain Evergreen, grown in 1931 and 1932, are presented in Tables 20 and 21. The coefficients are divided into five groups into which they seem to fall both naturally and logically, bearing in mind that the yield of sweet corn (ears) as it comes from the field may be broken down into the following components: 1. Unhusked culls 2. Sorted unhusked ears a. Husks b. Husked culls (including immature ears) c. Prime husked ears a a. Cut kernels bb. Cobs Yields of ears may be expressed according to number and weight. The weight per ear is the ratio between the weight per acre and the number of ears. Correlation Coefficients of Yield Components The correlation coefficients of the components ordinarily used in measuring the yield of sweet corn sorted unhusked ears, prime husked ears, and prime cut kernels were both high and significant (Section A of Tables 20 and 21, and Figs. 15 to 20). A great deal of variation, however, occurred between series and between the annual weighted means. It was assumed, in accordance with Fisher 9 * (p. 183) that when the difference between the values of r expressed as z was greater than twice the standard error of z, this difference was significant. The correlations between weight of prime husked ears and weight of prime cut kernels showed no significant differences when Series III and IV of the Country Gentleman variety (Table 20) were compared, nor did the weighted means show a significant difference. The Narrow Grain Ever- green variety (Table 21), also failed to show a significant difference. Altho many of the series, as well as the annual means, differed significantly (Tables 20 and 21), the range of variation in the coeffi- 1937] YIELD COMPLEX OF SWEET CORN 585 cients was not sufficient to obscure the fact that a very close relation- ship existed. The extreme range in all the correlations between weight of sorted unlnisked ears and weight of prime husked ears was .82 to .99, and all of the differences between series and between yearly means were significant. Therefore the yield of sorted unhusked ears should not be used for the purpose of predicting the yield of prime husked ears, because significant variations arise from differences in external conditions such as soil and weather. The correlations between weight of sorted unhusked ears and weight of prime cut kernels varied in a similar manner. Thus the weight of sorted unhusked ears, altho closely associated with the yield of cut kernels failed to be an accurate criterion of the yield. These considerations are of vital interest in interpreting the yields of sweet corn. Plainly the yields of unhusked ears cannot safely be used in determining the yield of prime husked ears or of kernels cut therefrom. The correlation surfaces and the regressions of several typical series from Section A of Tables 20 and 21 are shown in Figs. 15 to 18. The correlations between the weight of prime husked ears and weight of prime cut kernels differed significantly only when Series I and II were compared (Table 20), but the extreme range in both varieties was only from .96 to .98. Altho a difference between .98 and .96 is significant, that difference may arise to a large extent from the necessity of dropping decimals in making the calculations. The close degree of association is also shown in Figs. 19 and 20. The weight of prime husked ears is therefore a much better method of measuring the weight of cut kernels than is the weight of sorted unhusked ears. This conclusion is apparently in accord with the prac- tical experience of canners, who in purchasing corn find the husked- ear basis to be a good index of the yield of kernels. That this practice is open to certain objections will be shown later. The correlations between weight of sorted unhusked ears and per- centage of prime husked ears (Tables 20 and 21) varied widely in Country Gentleman but much less in Narrow Grain Evergreen. The annual mean coefficients (Table 20) did not differ significantly, and the weighted means of all series were very nearly alike for the two varieties. There was a slight tendency, subject to considerable varia- tion, for the percentage of prime husked ears to increase as the yield per acre of sorted unhusked ears increased. Apparently this slight degree of association disposes of the erroneous idea, which has long been current, that larger ratios of usable husked ears accompany increasing yields. 586 BULLETIN No. 432 [April, 00 CO >o-roi- 1000 cc >o i- r~ to oo oor^io o> ^ II W Mil; ,. ms s =sisis in si X et 1 |H ^ O U 3 sssi si sssssso Iss si u i 1 J>co N ^H OS n - t^- t^- CO OO I"- OO Tl > oooo oo ooooooo ooo oo ft. O 8 - S Fl l -H-H-H-H -H-H -H-H-H-H+I-H-H -H-H-H -H-H c^ooooo co-rf r^-cococooococo t^ooeo t-* ___ OO .5 1 "S U 03 | i;;;; ;4 ffi.'^jjt. Ill \\ !*: ^ ^ 1 e " 2 < CQ O Q W Note The errors of th 1937] YIELD COMPLEX OF SWEET CORN 587 1 e oo = ui*4>3*>~&e s*"" = I I 5COM5O- MOO N f- OOO ' f r i r i1 ii :::::!: ::! i! 1 : .8. : e ^ : ^ : ^ : : : : : J ;jj S^; fe^s^h^hii : : : | 1111 11 " J ' 1 6 J3 ^ N_ ^J3^J3^J3J3 "9T?T? ^^l III a! s :."5 3 X ^BPE: fcee ee'S'B'e^-S : e : 3 . jlll I12 u 3 W 588 BULLETIN No. 432 [April, 2 H q U o 6 as as H c U 03 S K S Si! 1 3^ t^- co eo r- SCM o o o os JSP co 00 co o co co r*. ' r r O CO O CO Z >~ if5 tn CO OS CM CO CO OO CO CO O CO CO CO * CO CM CO " ^ ^ S * '^ ^*"i^ ^^ pppp o o ppppppp pop H -H-H -H-H -H-H -H-H+I-H-H-H-H -H-H-H -H-H OS OO O* O> OS OS CO CM CO CO O O t^- *-" CM CM OS OS ' f f CM CM kA O CM ^H 1A CO CO IA CO CO CO CO CO kA d O pppp pp ppppppp pop op -H-H-H-H +I-H -H-H-H -H-H -H-H -H-H-H -H-H CMOOO500 OCM CO n JO f CO h- 00 0)NO> CgOO f i r II i : : * ^.J* 5 - ^_ _S-BJ9 i" * * 33 'P 1*1 < 111 1 13 -o -3 TJ -o 08 S2 H H I H v B a-a-a? ~r-a"a 3-?^^ fcfe^7"= HIS ell liJ^-S-^jl 3 Efg4 3 C C t- 8333 z?: YIELD COMPLEX OF SWEET CORN POUNDS SORTED UNMUSKED EARS 589 X 6.1- 7.1- 8.1- 9.t- 10.1-11.1- 12.1- 13.1-14.1- b.1- (6.1- 17.1- 18.1- 19.1- 201- 21.1- 22.1- 23.1- 24.1- 7 8 9 10 11 12 13 14 IS 16 17 18 Iff 2O 21 22 23 24 25 COUNTRY GENTLEMAN TS 8 23 29 31 35 36 24 23 21 IB 12 13 TOTALS OF ARRAYS FIG. 15. CORRELATION SURFACE: TOTALS OF ARRAYS, AND REGRESSION OF WEIGHT OF SORTED UNHUSKED EARS ON WEIGHT OF PRIME HUSKED EARS: COUNTRY GENTLEMAN SWEET CORN, SERIES IV The relationships existing with respect to number of ears are in- cluded (Section B, Tables 20, 21) because, for market growers who sell on the basis of count, the number of ears is more important than the weight. The correlations between number of sorted unhuskcd ears and num- ber of prime husked ears varied in a barely significant manner in Series I and II of Country Gentleman (Section B, Table 20), but the annual means did not differ significantly. In Narrow Grain Evergreen, how- 590 BULLETIN No. 432 POUNDS SORTED UNHUSKED EARS [April, X 9.1- 10.1-11.1- 2.1-13.1-14.1-15.1- ie.l-17.1- 18.1- 19.1- 20.1- 2 1J- 22> 23.1- 241- 251- 36.1 10 It 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 NARROW GRAIN EVERGREEN TD. TOTALS OF ARRAYS FIG. 16. CORRELATION SURFACE: TOTALS OF ARRAYS, AND REGRESSION OF WEIGHT OF SORTED UNHUSKED EARS ON WEIGHT OF PRIME HUSKED EARS: NARROW GRAIN EVERGREEN SWEET CORN, SERIES VI ever, there was a significant difference between the annual means (Section B, Table 21). All the coefficients in this group are high, and show that there was a fairly close relationship between number of sorted unhusked ears and prime husked ears. These correlations were subject to considerable variation because of external conditions which 79J7] YIELD COMPLEX OF SWEET CORN POUNDS SORTED UNHUSKED EARS 591 X 6.1- 7.1- 8.1- 9.1- 10.1- 11.1- 12.1-13.1-14.1-15.1-16.1-171- 18.1- 19.1- 201-21.1- 221- 231- 241- 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 COUNTRY GENTLEMAN 23 29 31 35 36 24 23 21 18 12 13 TOTALS OF ARRAYS FIG. 17. CORRELATION SURFACE: TOTALS OF ARRAYS, AND REGRESSION OF WEIGHT OF SORTED UNHUSKED EARS ON WEIGHT OF KERNELS CUT FROM PRIME EARS: COUNTRY GENTLEMAN SWEET CORN, SERIES IV were in turn the cause of fluctuations in the number of husked culls. Husked culls may be defined as sorted unhusked ears which upon husk- ing revealed their lack of sufficient development to be included with prime ears. All of the correlations between weights and numbers of sorted unhusked ears were high and differed significantly between the Country Gentleman series (Section B, Table 20), but annual means were prac- tically the same. In the Narrow Grain Evergreen variety (Section B, Table 21) there was no significant difference. It is apparent, therefore, that increased weight is to a large extent a function of increased number. Mean weights per ear which are important because of the general belief that the ear becomes heavier as the yield increases are shown in Section C of Tables 20 and 21. The correlations between weight of sorted unhusked ears and mean weight per sorted unhusked ear give a basis for this belief. The fluctuations in these coefficients were slight in both varieties and were not significant, leading to the conclusion that 592 BULLETIN No. 432 POUNDS SORTED UNMUSKED CARS [April, X 9.1- 1O.I-M.1 12.1- 13.1- 1 4.t- 15.1-16.1- 17.1- 16.1-191- 2O.1- 21.1- 22.1- 231-24.1- 25.1- 26.1- 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 NARROW GRAIN EVERGREEN "ZI TOTALS OF ARRAYS FIG. 18. CORRELATION SURFACE: TOTALS OF ARRAYS, AND REGRESSION OF WEIGHT OF SORTED UNHUSKED EARS ON WEIGHT OF KERNELS CUT FROM PRIME EARS: NARROW GRAIN EVER- GREEN SWEET CORN, SERIES VI the weight per sorted unhusked ear tends to increase with the yield in the Country Gentleman variety (Table 20). In the Narrow Grain Evergreen variety (Table 21) this tendency was very slight, leading to the belief that in some varieties there may be the possibility of securing high yields without an accompanying response in the weight per ear. In the Country Gentleman variety (Table 20) the correlations be- tween weight of sorted unhusked ears and mean weight per prime husked ear fluctuated widely, but the same correlations for the Narrow Grain Evergreen variety (Table 21) varied only slightly. Apparently the tendency for weight per husked ear to be associated with yield of sorted unhusked ears is subject to external conditions in Country Gen- tleman but possibly is not so associated in Narrow Grain Evergreen. In either case the mean degree of association is slight. The mean weight per sorted unhusked ear was associated with weight of prime cut kernels much more closely in Country Gentleman (Table 20) than in Narrow Grain Evergreen (Table 21), but there were no significant variations in either variety. This means that in 1937] YIELD COMPLEX OF SWEET CORN 593 w - 22 OS O CUU u. "3 P P w > < a * s si C^ K 5 O 594 BULLETIN No. 432 [April, H i I c/i 25 en < Is o 1937] YIELD COMPLEX OF SWEET CORN 595 Country Gentleman, where the average correlation was .554, a tendency existed for the yield of cut kernels to increase in relation to weight per unhusked ear. In Narrow Grain Evergreen, where the correlations averaged .359, this tendency was very slight. The mean weight per prime husked ear was similarly associated with yield of cut kernels, altho subject to significant variation in Country Gentleman (Table 20), where the mean correlation was .527 as compared with .392 for Narrow Grain Evergreen. The correlations between number of ears per acre and mean weight per sorted unhusked car fluctuated very widely in the Country Gentle- man variety (Table 20). This is also true for the two variables number of prime husked cars per acre and mean weight per prime husked ear. A low degree of association here indicates that weight per ear tends to fluctuate independently of the number produced. Zero or negative correlations might be expected. The correlations between mean ear weights of sorted unhusked and of prime husked cars were surprisingly low. These coefficients were also peculiar in that in the Country Gentleman variety (Table 20) the differ- ences between series were slight but the annual differences were very large. In Narrow Grain Evergreen also (Table 21) the annual differ- ence was large. Apparently in some years there may be a greater degree of association between the mean weights per ear than in others, but in a given year this association does not fluctuate very greatly. Data on husked culls are given in Section D of Tables 20 and 21. The correlations between weight of sorted unhusked ears and weight of husked culls were highly variable, but in nearly every instance they approached zero. There seems to be no tendency, therefore, for an increase in the yield of unhusked ears to be accompanied by an in- crease in the yield of husked culls. Results in Series III (Table 20) were different from all the other series with respect to husked culls a peculiarity which cannot be accounted for from the data at hand. The correlations between weight of prime husked cars and weight of husked culls varied materially, but with the exception of Series III (Table 20) they were negative. Thus in Narrow Grain Evergreen (Table 21) there was a slight tendency for the yield of culls to decline as prime ears increased. A similar tendency was apparent in all series of Country Gentleman except Series III. Prime cut kernels and husked culls were also slightly correlated with the same exceptions as noted in the previous paragraph. The correlations between weights of sorted unhusked ears and weights of cobs fluctuated significantly, but were uniformly high 5% BULLETIN No. 432 [April, (Section E of Tables 20 and 21 ). The correlations between the weights of prime husked ears and cobs varied comparatively little in Country Gentleman but significantly in Narrow Grain Evergreen. Weight of prime husked ears and weight of cobs were much more closely related than unhusked ears and cobs. Thus the weights of the cobs closely paralleled the weights of the ears. This means that cut kernels must give practically the same ratios, which is shown to be the case in Tables 20 and 21. Regression Coefficients of Yield Components The regression coefficients of the more important yield components in Table 22 are included for the purpose of extending the explanation of the yield complex in sweet corn. At the same time the regression coefficients show clearly the changed responses of sweet corn to vari- ations in seasonal and environmental conditions. For instance, accord- ing to Section A of Table 22, Series I responded quite differently than Series II to external conditions. Virtually all of the regression co- efficients show considerable variability, which indicates that their use for prediction purposes is quite limited. Two Interpretations of Regression Coefficients. The regression coefficients may be interpreted in two distinct ways. One of these is the conventional method whereby it is assumed that for every unit increase in the independent variable the dependent variable increases to the extent of the coefficient. Except for the coefficients in Section C of Table 22, an interpretation may also be made on the percentage basis. Thus the regression coefficient of weight of prime cut kernels on weight of prime husked ears in Series I was .51, which can be interpreted to mean that 51 percent of the weight of the prime husked ears consisted of the weight of prime cut kernels and that the balance was the weight of cobs. The percentage method of interpreting regression coefficients brings out the complementary nature of many of the components. Carrying out the example stated in the foregoing paragraph, the com- plementary regressions from Series I, Table 22, were as follows: Weight of prime cut kernels on prime husked ears = .51 Weight of cobs on prime husked ears = .50 Calculated total regression =1.01 Here cobs are complementary to prime cut kernels, and the sum of the regressions should equal unity, which it practically does. Similar com- putations made for the other series show that the total varies around unity. 1937] YIELD COMPLEX OF SWEET CORN 597 888 -H-H-H 88888 88 -H-H-H-H-H -H-H SSS S 88883 oo -H-H-H -H -H-H-H-H-H -H-H S OO *O OS OO O ** S * ooSo SSS s 53 H-H-H -H -H-H-H-H-H -H-H t^oo o S" CO ""S ^ o3 s -H-H-H -H -H-H-H-H-H -H-H tor^c-* m u? to e* c* oco 0.55 S 8888S -H-H-H -H -H-H-H-H-H r - .--i r 0MU) 00 002 3=00 588 -H-H-H -H-H-H-H-H sss? 8.8 -H-H oS -H-H-H -H -H-H-H-H-H StO h* *T*OI9W9 co><5 o> C^>no626' = = -H-H S3S -H-H-H 2 1^.00 es- 58888 So -H-H-H-H-H -H-H ~ H 'i = 598 BULLETIN No. 432 [April, Another comparison (Series I, Table 22) may also be made, viz.: Weight of prime cut kernels on weight of sorted unhusked ears = .29 Weight of cobs on weight of sorted unhusked ears = .33 Calculated total regression = .62 Actual weight of prime husked ears on weight of sorted unhusked ears. . = .59 The total calculated regression should equal .59 instead of .62 as shown above. Some of the other series are more consistent. Use of Correlation and Regression Coefficients for Estimating Yields. Another point of importance shown by the data in Table 22 is the variability of the regression coefficients within each variety. These fluctuations were most likely due to changes in external con- ditions and they correspond to similar changes noted in Part I. There- fore the value of regression coefficients for the purposes of prediction or estimation even within a variety seems to be very limited. The mean coefficients are of some value provided a sufficient number of variates grown under a wide range of conditions is included. The results given in Table 22 also show rather clearly that regression coefficients are applicable only to the variety in question. However, the relationships shown in Tables 20, 21, and 22 are undoubtedly linear, as is indicated by the scatter diagrams in Figs. 15 to 20. A comparison of the regression coefficients (Table 22) with the correlation coefficients (Tables 20 and 21) shows that the latter may prove to be quite unreliable, especially for the purposes of prediction. Thus in Series V and VI (Table 21) the respective correlations for weight of prime husked ears and weight of prime cut kernels were .98 and .97, with a difference only 1.6 times the standard error. The re- spective regression coefficients (Table 22) are .48 .02 and .57 .02, the difference being .09 .028, which is significant. This would mean that in every 100 pounds of prime husked ears produced by Series VI there would be 9 2.8 pounds more cut corn than in Series V. Other comparisons might be made which would show even larger differences. This comparison emphasizes the fact that in results of this kind cor- relation coefficients alone, even tho approaching very closely to unity, should not be accepted without supporting evidence, especially for prediction purposes. The regression coefficients in Table 22 show that as the yield of unhusked ears increased, there was a uniform altho fluctuating tend- ency for the mean weight per ear to increase. The regression co- efficients of mean weight per sorted unhusked ear and of mean weight per prime husked ear on weight of kernels per acre confirm this con- clusion. The regression coefficients of mean weight per prime husked 1937] YIELD COMPLEX OF SWEET CORN 599 ear on mean weight per sorted unhusked ear were highly variable and most of them considerably below unity, indicating that much of the increase in weight per sorted unhusked ear is due to factors other than mean weight per prime husked ear. The husks being the only remain- ing component, it follows that except for the somewhat minor in- fluence of husked culls, a considerable part of the increase in mean weight per unhusked ear is due to husks. A comparison of the two varieties for which data are given in Table 22 shows that with respect to mean weight per ear the Narrow Grain Evergreen variety was considerably less influenced by total yields than was Country Gentleman. This might be interpreted to mean that in adverse seasons when yields are low the mean weight per ear would be less unfavorably affected in the Narrow Grain Ever- green variety and, conversely, in favorable seasons Country Gentle- man ears would increase in mean weight to a considerably greater degree. DISCUSSION OF CORRELATION ANALYSIS The analysis of sweet-corn yields by means of correlations em- phasizes the fact that while there is a close association between the more important yield components weights of sorted unhusked ears, prime husked ears, and prime cut kernels that association is subject to a great deal of variation because of external conditions. Thus it appears to be inadvisable to attempt to interpret the yield of one com- ponent in terms of another. The association between prime husked ears and prime cut kernels, however, is both very close and relatively stable, a fact which ap- parently indicates that the yield of prime husked ears is considerably the best measure for predicting the yield of cut kernels. It seems probable that in the long run this method for predicting the yield of cut kernels may be sufficiently reliable for commercial purposes. According to the regression coefficients given in Table 22, however, the percentages of cut kernels were rather variable from series to series and, therefore, the use of prime-husked-ear weights for predicting the yields of prime cut kernels may be inaccurate in individual cases. The use of the weights of sorted unhusked ears for predicting the yields of prime husked ears or of prime cut kernels is not recommended because of the very wide fluctuations of the regression coefficients. From the viewpoint of the canner these data show clearly that purchasing sweet corn on the unhusked-ear basis is definitely inferior to purchasing it on the husked-ear basis, the present practice of many 600 BULLETIN No. 432 [April, factories in the Middle West. The latter method is apparently much more accurate and seems to answer the purpose quite well, but if absolute accuracy is essential, as in experimental work, the cut kernels themselves should be weighed. The coefficients throw light on the general question of the effect of favorable growing conditions upon the components as reflected by increased yields. The low correlations between weight of sorted unhusked ears and percentage of prime husked ears obtained in these experiments show only a slight tendency for a better husking ratio to be associated with yield. Higher yields, however, were associated in general with fewer husked culls. At the same time higher yields were usually accompanied by heavier ears, especially in the Country Gentle- man variety, a consideration which is obviously of the greatest im- portance ; and increased weight per ear was associated with increased yield of cut kernels, a tendency which was far more pronounced in Country Gentleman than in Narrow Grain Evergreen. Mean weight per ear was variably associated with number of ears produced per acre, especially in the Country Gentleman variety, an association which indicates that under some conditions an increased yield expressed as number of ears may not be accompanied necessarily by an increased weight per ear. The tendency for mean gross weight per ear to be associated with mean net weight was lower than expected, and apparently was subject to wide variations from season to season. The yield of husked culls was inversely associated, usually with yields of prime husked ears and cut kernels. Some exceptions were noted, however. The association between unhusked-ear and husked- cull weights was usually very slight. The cobs showed a strong tendency to vary in the same direction as the sorted unhusked and prime husked ears. This indicates that in spite of fluctuations the components of sorted unhusked ears namely, prime ears, cut kernels, and cobs all tend to vary in the same direction, but this tendency is modified more or less by such factors as husked culls. The regression coefficients are unusually important in an analysis of this kind. They show in a measurable manner the fluctuations which are due to external conditions and which are not apparent from a comparison of the correlations. The regression coefficients also em- phasize the limitations of one component as a measure of any one or more of the others. In addition, the regression coefficients show that high correlations between yield components must be interpreted in a very restricted sense. 1937] YIELD COMPLEX OF SWEET CORN 601 SUMMARY AND GENERAL CONCLUSIONS PARTS I AND II The chemical changes which characterize the ripening process of sweet corn are quite well understood, but all that is known about the accompanying physical changes has heretofore been derived from studies of individual ears. The investigations which have been reported in this bulletin had for their principal object a study of the changes which take place in the yield complex of mass populations of maturing sweet corn. Three varieties Country Gentleman, Narrow Grain Evergreen, and Golden Bantam were used. These experiments were conducted in 1931 and 1932, during a period of excessively high mean temperatures and subnormal rainfall. Such conditions are interpreted as being generally favorable for sweet corn, except for Golden Bantam. Studies of Individual Ears During Maturation Moisture Content. Experiments with individual ears selected for simultaneous silk emergence confirmed previous observations on the loss of moisture from the kernels accompanying the ripening process. Husks and cobs lost moisture in a similar manner, but the percentage of moisture in the husks remained consistently higher than in the cobs. The moisture content of kernels, of husks, and of cobs exhibited linear relationships with respect to advancing maturity ; and consequently the moisture content of any or all of these components may be used for measuring maturity. Moisture content of cobs or husks would, of course, be but an indirect measure of maturity of the kernels. Toughness of Pericarp. Penetration tests with a specially designed penetrometer confirmed previous work in indicating that toughness of pericarp in sweet corn increases as maturity advances. Wide varia- tions in the mean penetration values of ears silking on the same day and picked at the same time indicated that silk emergence is not an entirely accurate index of physiological maturity. Penetration tests also brought out differences in toughness of pericarp among the different varieties. Narrow Grain Evergreen rated consistently higher in toughness than Country Gentleman at any comparable time after silking. This difference in toughness is interpreted to mean that Narrow Grain Evergreen is inferior in quality to Country Gentleman. As a measure of maturity, penetration tests were of doubtful value during the later stages when the kernels were flaccid and denting had 602 BULLETIN No. 432 [April, occurred, for dented kernels tended to give lower penetration values than normal kernels. This lowering of penetration values of dented kernels indicates that the turgidity of the endosperm modifies penetra- tion values very materially, thus leading to false interpretations of the toughness of the pericarp. Statistical Analysis of Chemical Analyses. A statistical analysis of a large number of published chemical analyses of maturing sweet corn showed that moisture content of the kernels is very closely related to maturity as measured by chemical composition. This finding, cor- roborated by the fact that moisture content was also closely related to maturity as determined by the period after silk emergence, and as related to toughness of the pericarp, led to the conclusion that moisture content of the kernels is an accurate means of measuring maturity. The statistical methods used in analyzing the chemical data were applicable for the estimation of total sugars and starch during the period of edibility of sweet corn even tho only a few chemical analyses were available. Estimation of the other constituents should be possible by the use of similar methods. Studies of Mass Populations Successive decreases in moisture content of prime ears sorted out of a mass population at each picking are due entirely to errors of selec- tion which arise from the inability to separate, on the basis of ap- pearance, ears which have just reached the prime canning stage and those that are about to dent. For this reason canners who wish to pack corn whole-grain style should snap the ears early in preference to attempting to sort out the less mature ears from corn snapped for cream-style canning. Unhusked Ears. The unhusked ears in a mass population in- creased both in number and in total weight as maturity advanced. During the later stages of maturity the number of ears per acre, with certain exceptions, tended to remain constant, but the weight tended to decline. The period when the weights of the unhusked ears attained their maximum was subject to considerable fluctuation as a result of environmental conditions and seemed not to be associated definitely with any particular moisture content. Decreases in weight during the later stages of maturity seemed to arise from losses of moisture alone, which were very rapid at that period. Prime Husked Ears. Prime husked ears in a mass population in- creased in number until either just before or just after the appearance of dented ears. There then followed a sharp decrease, accompanied 1937] YIELD COMPLEX OF SWEET CORN 603 by a rapid increase in the number of dented husked ears. The total number of sorted husked ears usually varied in the same manner as the number of sorted unhusked ears. Total weight of prime husked ears in the mass population also in- creased and reached maximum production at the period when dented ears first appeared. The moisture content at this time was 70 percent. The percentage of prime ears calculated on the unhusked-ear basis varied in the same manner as the acre-yields. As prime husked ears are the most valuable of the yield com- ponents, with the exception of cut kernels, the fact that acre-yields and percentages of prime husked ears coincided is highly significant, and is an irrefutable reason for canners to purchase sweet corn on the husked-ear instead of on the unhusked-ear basis. Sorted Husked Ears, Weights of sorted husked (prime plus dented) ears in the mass populations increased as maturity advanced, accompanied by rapid changes in the ratios of prime and dented ears. Peak production of sorted husked ears was reached after sorted un- husked ears attained their maximum. Climaxes in Maturity. The ripening period of sweet corn was characterized by a series of three distinct climaxes, which occurred in the following order: 1. Period of maximum yield and percentage of prime husked ears, moisture content 70 percent. 2. Period of maximum yield of sorted unhusked ears, highly vary- ing moisture content averaging from 66.4 to 68.3 percent. 3. Period of maximum yield of sorted husked ears, moisture con- tent ranging from 63.0 to 65.2 percent. A fourth climax in yield occurred at a still later date. This climax was the period when the percentage of husked (prime plus dented) ears reached the maximum. This point had no economic significance because the ears were for the most part inedible at that time. Kernels From Prime Ears. The yield and the percentage (un- husked ears = 100) of kernels cut from prime ears in a mass popula- tion are more important as yield components than prime husked ears, but as these three components vary so closely in relation to each other they may be considered collectively. For "cream style" maturity it is advisable to harvest the three varieties of sweet corn used in these experiments when the moisture content averages 70 percent. For whole-grain maturity the moisture content should exceed 70 percent. The percentage of kernels cut from prime ears (husked ears = 100) 604 BULLETIN No. 432 [April, reached the maximum at a much later period than the maximum weight or the maximum percentage of kernels computed on the un- husked basis. The average moisture content at this time ranged from 64.4 to 67.2 percent, indicating that the highest yield of cut kernels from a ton of prime husked ears is obtained only when the corn has passed the best (70 percent) edible stage. Kernels From Sorted Husked Ears. The acre-yield of kernels cut from sorted husked (prime plus dented) ears reached the maximum at the same time as the yield of the ears, the ears at that time showing 63 to 65 percent moisture content. This production peak was followed very closely by the production peak as measured by the ratio of cut kernels to weight of unhusked ears, which occurred when the ears had moisture contents of 62 to 67 percent. Time of Appearance of Dented Ears. The first appearance of dented ears in the Country Gentleman and Narrow Grain Evergreen varieties was noted when the moisture averaged 70 percent. This co- incided with the first climax. The appearance of denting in these varie- ties may be used, therefore, as a reliable index for determining the proper time to pick sweet corn for cream style canning. Other Components. There was a slight tendency for unhusked culls to increase in yield during the earlier stages of maturity, but the later stages were characterized by marked declines. Husked culls decreased both in yield and in percentage during the later stages of maturity, owing most likely to the fact that many of the culls were simply immature ears which if undisturbed, would have developed later into usable ears. Green or immature ears were not found after the first few pickings. The percentage of husks showed a marked decline as maturity advanced. Weights During Maturation Period. All the experiments concern- ing the maturing mass complex of sweet corn showed that the mean weights per ear tended to increase for a period, following which de- creases occurred. The reduction in the mean weights of the early maturing ears seemed to be due to the fact that the earliest silking ears were precocious and weighed less. The increases in weight per ear of the shoots silking later were due to the rapid development of the kernels. The decreases in weight per ear which occurred still later were due to a combination of loss of moisture and the inclusion in the usable class of larger percentages of late-silking ears which proved to be lighter in weight per ear. 1937] YIELD COMPLEX OF SWEET CORN 605 Silk Emergence and Maturity. Silks tended to emerge according to a skew curve, with the lower shoots lagging a day or two behind the upper. This lag tended to widen the range of maturity included in any picking of sweet corn and showed why plant breeders should select strains maturing only a single ear to the plant if this can be done without decreasing the total yield. For most practical purposes the mid-silking point may be estimated by counting silks until the total number reaches 75 percent of the theo- retical stand. In addition, such counts made on a few small random areas seemed in these experiments to provide an accurate estimate of the silk emergence in a whole field. Shoots silking earliest, except lower shoots, produced a higher per- centage of unhusked culls than shoots silking in the median periods. Late-silking shoots, both upper and lower, produced more culls than the earlier shoots. Upper shoots produced heavier ears and a lower percentage of culls than lower ears. When all the ears in a given planting were picked the same number of days after silk emergence, the moisture content, provided the weather conditions were reasonably uniform, was maintained at prac- tically a constant level. The maturity of individual ears in small random samples paralleled the maturity of the mass population. It seems permissible, therefore, to use small samples in order to estimate the maturity of a whole field. Correlation Analyses Correlation analysis of the yield complex of Country Gentleman and Narrow Grain Evergreen sweet corn showed that the principal yield components were closely associated with each other. Sorted unhusked ears, prime husked ears, and prime cut kernels were closely correlated. Prime husked ears and cut kernels, and also cobs, which are comple- mentary to kernels, showed the closest correlations. The percentage of prime husked ears had only a slight tendency to increase in relation to unhusked-ear weight. The number of sorted unhusked ears was closely correlated with number of prime husked ears and with weight of sorted unhusked ears. Jncreases in mean ear weights were not generally associated with higher acre-yields, altho there was a varying tendency in this direction, Country Gentleman showing a stronger tendency than Narrow Grain Evergreen. The closest relation between mean weight per ear and total weight per acre appeared in three pairs of correlated variables: (1) weight per acre and mean weight per sorted unhusked ear; 606 BULLETIN No. 432 [April, (2) weight per acre of prime cut kernels and mean weight per sorted unhusked ear; and (3) weight per acre of prime cut kernels and mean weight per prime husked ear. Mean weight per ear was only slightly correlated with number of ears per acre ; and mean weight per sorted unhusked ear was variably correlated with mean weight per prime husked ear. The average coefficients of the latter correlation were unexpectedly low. Increases in the yields both of ears and of cut kernels were usually accompanied by decreases in the weight of husked culls. The yields of prime husked ears and of the kernels cut from them cannot be accurately estimated from the yields of sorted unhusked ears. The yield of cut kernels can be estimated, however, from the yield of prime husked ears with sufficient accuracy for commercial, but not for experimental, purposes. The correlations were believed to be nongenetic, because they were characterized by wide fluctuations from series to series, from year to year, and from variety to variety. The correlation coefficients, even when consistent, failed to give as satisfactory a measure of the rela- tionship as the regression coefficients. 1937] YIELD COMPLEX OF SWEET CORN 607 LITERATURE CITED 1. APPLEMAN, C. O. Forecasting the date and duration of the best canning stage for sweet corn. Md. Agr. Exp. Sta. Bui. 254. 1923. 2. BORST, H. L., and THATCHER, L. E. Life history and composition of the soy- bean plant. Ohio Agr. Exp. Sta. Bui. 494. 1931. 3. BRUCE, DONALD, and REINEKE, L. H. Correlation alinement charts in forest research: A method of solving problems in curvilinear multiple correla- tion. U. S. Dept. Agr. Tech. Bui. 210. 1931. 4. BURGESS, R. W. Introduction to the mathematics of statistics. Boston. 1927. 5. BURNETT, L. C., and BAKKE, A. L. The effect of delayed harvest upon yield of grain. Iowa Agr. Exp. Sta. Res. Bui. 130. 1930. 6. CULPEPPER, C. W., and MAGOON, C. A. Studies upon the relative merits of sweet corn varieties for canning purposes and the relation of maturity of corn to the quality of the canned product. Jour. Agr. Res. 28, 403-443. 1924. 7. . A study of the factors determining quality in sweet corn. Jour. Agr. Res. 34, 413-433. 1927. 8. EZEKIEL, MORDECAI. Methods of correlation analysis. New York. 1930. 9. FISHER, R. A. Statistical methods for research workers. 4th ed. Edinburgh and London. 1932. 10. HADDAD, E. S. Morphological development of sweet corn pericarp in two inbred lines and their Fi hybrid. Purdue Agr. Exp. Sta. Bui. 347. 1931. 11. HOFFMAN, I. C. The relation of size of kernels in sweet corn to evenness of maturity. Jour. Agr. Res. 31, 1043-1053. 1925. 12. HOTTES, CHARLES F., and HUELSEN, WALTER A. The determination of quality in sweet corn by means of the optical measurement of leached materials. Jour. Agr. Res. 35, 147-166. 1927. 13. JENKINS, R. R. Alcohol-insoluble residue as an index of quality of sweet corn. Proc. Amer. Soc. Hort. Sci. 32, 587-592. 1934. 14. MAGOON, C. A., and CULPEPPER, C. W. The relation of seasonal factors to quality in sweet corn. Jour. Agr. Res. 33, 1043-1072. 1926. 15. - . Response of sweet corn to varying temperatures from the time of planting to canning maturity. U. S. Dept. Agr. Tech. Bui. 312. 1932. 16. MILLS, F. C. Statistical methods applied to economics and business. New York. 1924. 17. MYERS, M. I. Determining the date of silking in experiments with corn. Jour. Amer. Soc. Agron. 22, 280-283. 1930. 18. RIETZ, H. L. (edit.) Handbook of mathematical statistics. Boston. 1924. 19. SAYRE, C. B., WILLAMAN, J. J., and KERTESZ, Z. I. Factors affecting the quality of commercial canning peas. N. Y. (Geneva) Agr. Exp. Sta. Tech. Bui. 176. 1931. 20. STRAUGHN, M. N. Sweet corn investigations. Md. Agr. Exp. Sta. Bui. 120. 1907. 21. UHLAND, R. E. Time of harvesting soybeans in relation to soil improvement and protein content of the hay. Missouri Agr. Exp. Sta. Bui. 279. 1930. 22. WALLACE H. A., and SNEDECOR, G. W. Correlation and machine calculation. Iowa Agr. Col. Off. Pub. 30, No. 4, rev. 1931. 23. WOODWORTH, C. M. Genetics and breeding in the improvement of the soy- bean. 111. Agr. Exp. Sta. Bui. 384. 1932. 608 BULLETIN No. 432 LIST OF TABLES AND CHARTS TABLES PACK 1. Effect of maturity on component parts of individual ears, Country Gentleman 521 2. Same, Narrow Grain Evergreen 522 3. Same, Golden Bantam 523 4. Penetration values and percentages of moisture 529 5. Penetration values, normal and dented kernels, same ear 532 6. Relation between dry matter and total sugars, total polysaccharids, and sugar/starch ratios 537 7. Correlations between total solids and other constituents 539 8. Relation between date of silk emergence and physical composition of ear, Country Gentleman 540 9. Same, Narrow Grain Evergreen 542 10. Same, Golden Bantam 544 11-12. Effect of maturity on relations between yield components, Country Gentleman 548, 550 13-14. Same, Narrow Grain Evergreen 553, 555 15-16. Same, Golden Bantam 558, 559 17. Time when yields reached maximums, and moisture content of kernels. . 562 18. Conversion table for different styles of cut and cutters 573 19. Median points of silk emergence, actual and calculated 577 20. Correlations between yield components, Country Gentleman 586 21. Same, Narrow Grain Evergreen 588 22. Regression coefficients of yield components 597 FIGS. PACK 1-2. Temperature, rainfall, and sunshine at Urbana, Illinois, 1931 and 1932 516- 19 3. Relation of maturity to principal yield components, Country Gentleman.. 524 4. Same, Narrow Grain Evergreen 525 5. Relation of maturity to toughness of pericarp, Country Gentleman.... 530 6. Same, Narrow Grain Evergreen 531 7. Relation between total polysaccharids calculated as starch, and total solids 535 8. Relation between sugar/starch ratios and total solids 536 9. Relation between total sugars calculated as invert and total solids 537 10-11. Relation between maturity and percentages of yield components Country Gentleman, Series I and Series IV 560, 561 12-13. Same, Narrow Grain Evergreen, Series I and Series IV 564, 565 14. Same, Golden Bantam, Series 1 566 15. Correlation surface: sorted unhusked ears on weight of prime husked ears, Country Gentleman, Series IV 589 16. Same, Narrow Grain Evergreen, Series VI 590 17. Correlation surface: sorted unhusked ears on weight of kernels cut from prime ears, Country Gentleman, Series IV 591 18. Same, Narrow Grain Evergreen, Series VI 592 19. Correlation surface : prime husked ears on weight of kernels cut from prime ears, Country Gentleman, Series IV 593 20. Same, Narrow Grain Evergreen, Series VI 594 3050 4-3710791 AUTHOR INDEX 609 AUTHOR INDEX 1. ANDERSON, H. W. See KADOW 16 2. BADGER, C. J. See BAUER 4 3. BARTLETT, R. W., and CASKEY, W. F. Milk Transportation Problems in the St. Louis Milk- shed 421-172 4. BAUER, F. C., LANG, A. L., BADGER, C. J., MILLER, L. B., FARNHAM, C. H., and JOHNSON, P. E. Crop Yields From Illi- nois Soil Experiment Fields In- cluding the Crop Season of 1935 145-244 5. BIGGER, J. H. See DUNCAN 9, 10 6. BURLISON, W. L., See PIEPER 24 7. CASKEY, W. F. See BARTLETT 3 8. DAWSON, W. M. .See LINDSTROM 19 9. DUNCAN, G. H., HOLBERT, J. R., MUMM, W. J., BIGGER, J. H., and LANG, A. L. Illinois Corn Performance Tests: Results for 1935 281-340 10. DUNCAN, G. H., HOLBERT, J. R., MUMM, W. J., BIGGER, J. H., and LANG, A. L. Illinois Corn Performance Tests: Results for 1936 389-420 11. FARNHAM, C. H. See BAUER 4 12. HEDGES, T. R. See NORTON 23 13. HOLBERT, J. R. See DUNCAN 9, 10 14. HUELSEN, VV. A., and MICHAELS, W. H. The Yield Complex of Sweet Corn. I. Effect of Ad- vancing Maturity II. Rela- tions Between Yield Compo- nents 505-608 15. JOHNSON, P. E. See BAUER 4 16. KADOW, K. J., and ANDERSON, H. W. Further Studies on Zinc Sulfate in Peach Sprays, With Limited Tests in Apple Sprays 129-144 17. LANG, A. L. See BAUER 4, DUNCAN 9, 10 18. LINDSTROM, D. E. Forces Affect- ing Participation of Farm Peo- ple in Rural Organization: A Study Made in Four Town- ships in Illinois 77-128 19. LINDSTROM, D. E., and DAWSON, W. M. Selectivity of 4-H Club Work: An Analysis of Factory Influencing Membership. .245-280 20. MICHAELS, W. H. See HUELSEN 14 21. MILLER, L. B. See BAUER 4 22. MUMM, W. J. See DUNCAN 9, 10 23. NORTON, L. J., and HEDGES, T. R. Prices of Illinois Farm Products From 1931 to 1934. . 1-76 24. PIEPER, J. J., and BURLISON, W.L. Foreign Strainsof Alfalfa and Red Clover What Is Their Adaptability to Illinois?. . .473-504 25. Ross, R. C. Soybean Costs and Production Practices. . . .341-388 610 INDEX INDEX Aledo experiment field yields, long-time summaries of 201-202 Alfalfa hay, foreign and domestic strains of compared 483-487 Alfalfa seed, characteristics of foreign strains of in Illinois 480-481 performance of foreign strains of in Illinois 482, 483-487 practices to use during shortages of 495-496 regulation of importation of 479-480 trends in production, importation, and acreage of 476-478, 497-499 Antioch experiment field yields, long-time summaries of 203 Apple sprays, field tests with zinc sulfate in 136-138 Apples, price trends of 23, 24, 30, 36, 39, 69 Barley, price trends of 8, 30-31, 39, 69 production trends of 31-32, 48-50 Bloomington experiment field yields, long-time summaries of 204 Broomcorn, production trends of 48, 51 Butterfat, price trends of 16-17, 25, 28, 30, 35-36, 39, 69 Carlinville experiment field yields, long-time summaries of 205-206 Carthage experiment field yields, long-time summaries of 206-208 Cattle, beef and dairy, price trends of 12-14, 25, 27, 30, 39, 69, 72, 73 production trends of 34, 52-54 Chickens, price trends of 17-18, 25, 28, 30, 35, 39, 69 Clayton experiment field yields, long-time summaries of 208-209 Clover hay, red, foreign and domestic strains of compared 487^194 Clover seed, alsike, annual U. S. production and importation of 497, 502 Clover seed, red, characteristics of foreign strains of 481-482 performance of foreign strains of in Illinois 482-483, 487-494 practices to use during shortages of 495^196 price trends of 22, 30, 36, 39, 70 regulation of importation of 479480 trends in production, importation, and acreage of 476-478, 497, 500-501 Clover seed, sweet, annual U. S. production and importation of 497, 503 CLUB WORK, 4-H, AN ANALYSIS OF SELECTIVITY FACTORS INFLUENCING MEM- BERSHIP 245-278 See Contents 246 purpose and plan of study 248-251 Summary and conclusions 277-278 Community activities, individual participation in, factors associated with. . 109-111 survey of in four Illinois townships 99-108 Corn, price trends of 5-6, 24-26, 30, 31, 39, 70 production trends of 31-32, 48-50 response of to soil treatments compared with that of oats, of wheat, of hay. . 1 79-180 yields of on Illinois soil experiment fields, in 1934 154-155 in 1935 161-163 for four-year period 170-172 CORN PERFORMANCE TESTS, RESULTS IN ILLINOIS FOR 1935 281-340 See Contents 282 contributors to 339-340 hybrid tests, two-year summaries of 31 7-320 with second-generation seed 317, 321-322 location of fields and soil types 284-288 measure of entries 292-293 number and kind of entries. . . .283-284 INDEX 611 planting and arrangement of entries 290-292 results of grain tests 294, 296-309 of silage tests 314-316 of soil-adaptation tests 294-295, 310-313 seasonal conditions and insect problems 288-290 Summary 322 CORN PERFORMANCE TESTS, RESULTS IN ILLINOIS FOR 1936 389-420 See Contents and Index 390 hybrid tests, two-year summaries of 396, 415-416 location of fields 392 measure of entries 394-395 number and kind of entries 391 planting and arrangement of entries 394 results of grain tests 395, 397-414 of silage tests 396,417^118 of soil-adaptation tests 396-397, 419-420 seasonal conditions and insect problems 393 Cowpeas, price trends of 8-10, 30-31, 70 production trends of 48, 51 Crop production, changes in on Illinois farms 48-51 Crop residues, influence of in increasing crop yields and values 183-185 use of in soil treatment, see Index 242-243 CROP YIELDS FROM ILLINOIS SOIL EXPERIMENT FIELDS, INCLUDING THE CROP SEASON OF 1935 145-243 See Contents 146 analysis of for 1934 152, 154-160 for 1935 152, 161-167 for 1932-1935. 152-153, 170-180 economic analysis of 149 influence of various materials on 181-200 measure of by digestible nutrients 148-149 by money values 148 by yield index 149 Dixon experiment field yields, long-time summaries of 209-210 Eggs, price trends of 18-19, 25, 28, 30, 35, 39, 70 Elizabethtown experiment field yields, long-time summaries of 211 Enfield experiment field yields, long-time summaries of 211-212 Ewing experiment field yields, long-time summaries of 212-214 Exports of Illinois farm products, changes in 54-59 Farm families, mobility of in four Illinois townships studied 83-86 relation of mobility of to participation in community organization 110 survey of activities of 88-91 Farm products, trends in prices of from 1921 to 1934 4-28 Farm purchasing power, changes in from 1931 to 1934 44-48 Farms, size and value of in four Illinois townships 81, 82 Fertilizer and soil treatment materials, see Index 242-243 FOREIGN STRAINS OF ALFALFA AND RED CLOVER, THEIR ADAPTABILITY TO ILLINOIS 473-504 See Contents 474 Summary and Conclusions 494-495 Hartsburg experiment field yields, long-time summaries of 214-216 Harwood township, study of rural organization in 79-126 Hay, price trends of 20-22, 25, 28, 30, 36, 39, 70, 71 production trends of 48-49, 51 response of to soil treatment compared with that of corn, of oats, of wheat. . 179-180 yields of on Illinois soil experiment fields, in 1934 159-160 in 1935 168-1 70 for four-year period 177-178 612 INDEX Hogs, price trends of 10-11, 25, 27, 30, 34, 39, 71 production trends of 52-54 Horses, price trends of 15-16, 25, 27, 30, 39, 71 production trends of 34, 52, 53 Illini township, study of rural organization in 79-126 Joliet experiment field yields, long-time summaries of 217-218 Kewanee experiment field yields, long-time summaries of 219-221 Lambs, price trends of 14-15, 25, 27, 30, 34, 39, 71 Land tenure, relation of type of to participation in community activities. . . 109-110 survey of in four Illinois townships 82 Lard exports, trends in 56-57 Lead-arsenate - lime injury reduced by use of zinc sulfate in sprays 131-139 Leadership in rural community organizations, development of 117-120 qualities of 120-123 Lebanon experiment field yields, long-time summaries of 221-224 Legume programs for Illinois farmers 495-496, 504 Legumes, small-seeded, U. S. production and importation of 497 use of as crop residues in soil treatment, see Index 242-243 Lespedeza seed, annual U. S. production of 502 Limestone, influence of in increasing crop yields and values 185-186 use of in soil treatment, see Index 242 Livestock and livestock products, price trends of 10-20, 34-38 See also Cattle, Chickens, Eggs, Hogs, Horses, Lambs, Sheep, Wool Livestock production, changes in on Illinois farms 51-54 Manure, influence of in increasing crop yields and values 181-183 use of in soil treatment, see Index 242 Manure system compared with residues system on crop experiment fields 154-180, 190-194 Manures, green, see Crop residues McNabb experiment field yields, long-time summaries of 224 Milam township, study of rural organization in 79-126 Milk, price trends of 16, 30, 35, 39, 71 quality of related to program of delivery 464-466 MILK TRANSPORTATION IN THE ST. Louis MILKSHED, PROBLEMS OF WITH SUGGESTED SOLUTIONS 421-472 See Contents 422 adjustments in needed 468, 472 costs of to producers and distributors 441-447 ways to reduce 450460 hauling routes for, revision of needed 460-464 index to tables and figures .469-470 scope of information on 424-428 Summary 466-468 system of in use 428-441 Milk trucks, cost of operating in the Dayton milkshed 448-450 Mineral fertilizers for soil treatment, see Index 242 Minonk experiment field yields, long-time summaries of 225 Monetary policy, effect of on prices of farm products 60-63 Morrow plots, long-time summaries of crop yields on 239 Mt. Morris experiment field yields, long-time summaries of 226-227 Newton experiment field yields, long-time summaries of 227-228 Nitrogen, use of in soil treatment, see Index 242 Oats, price trends of 6, 25, 26, 39, 72 production trends of 31-32, 48-50 response of to soil treatment compared with that of corn, of wheat, of hay. . 179-180 INDEX 613 PACK yields of on Illinois soil experiment fields, in 1934 157 in 1935 166-167 for four-year period 1 75-1 76 Oblong experiment field yields, long-time summaries of 229-230 Oquawka experiment field yields, long-time summaries of 231 Peach sprays, field studies of zinc sulfate in 134-136 laboratory studies of zinc sulfate in 132-133 Philo township, study of rural orgaization in 79-126 Phosphate, influence of in increasing crop yields and values 186-187 Phosphorus, use of in soil treatment, see Index 242-243 Population, study of in four Illinois townships in relation to community organiza- tion 81,83-86 Pork exports, trends in 56-57 Potash, influence of in increasing crop yields and values 187-188 Potassium, use of in soil treatment, see Index 242 Potatoes, price trends of 22-24, 30, 36, 39, 72 PRICES OF ILLINOIS FARM PRODUCTS, FROM 1931 TO 1934 1-73 See Contents 2 comparison of for individual products for 1931-1933 and 1934 28-44 monthly index of 69-73 outlook for after 1934 63-65 reasons for general rise of in 1933-34 59-63 Summary 65-68 trends in from 1921 to 1934 4-28 Raleigh experiment field yields, long-time summaries of 232, 233 Residues system compared with manure system on crop experiment fields. . . . 154-180, 190-194 RURAL ORGANIZATION, A STUDY OF FORCES AFFECTING PARTICIPATION OF FARM PEOPLE IN 77-127 See Contents 78 implications of the study 126-127 Summary 124-126 Rye, price trends of 8, 9, 30, 31, 39, 72 production trends of 31-33, 48, 50 Schooling, survey of in four Illinois townships 86-88 relation of to participation in community organization 110-1 1 1 Sheep, price trends of 14, 30, 35, 39, 72 production trends of 53, 54 Soil conservation program, unadapted seed not accepted for 504 Soil productivity, wide range in levels of 181-182 effect of treatment on 195-197 Soil treatment, economic evaluations of systems of 188-195 relation of to crop quality 196-200 SOYBEAN COSTS AND PRODUCTION PRACTICES 341-388 See Contents 342 analysis of for 1934 in two Illinois counties 347-351 enterprise study of for 1928-1929 on central-Illinois farms 344-346, 352-354 sources of data used in study of 344-347 Summary 386-388 Soybean costs, before-harvest practices influencing 355-365 changes in during 1928-1934 351-352 harvest practices influencing 365-376 Soybeans, harvesting of for hay 377-379 income and profit from studied 376-377 place of on central-Illinois farms 379-386, 388 price trends of 10, 30, 31, 73 production trends of 31-34, 48-51 614 INDEX yields of on four Illinois soil experiment fields, in 1934 158 in 1935 . 167 Sparta experiment field yields, long-time summaries of 232, 234 Spray injury, prevention of by use of zinc sulfate 131-139 Spray schedule used in apple-orchard experiment 144 Spray schedules used in peach-orchard experiment 141-143 St. Louis milk transportation problems, suggested solutions for 422-468 SWEET CORN, YIELD COMPLEX OF 505-608 See Contents 507 correlation analysis of 581-600 effect of maturity on 511-581 list of tables and charts 608 literature cited 607 Summary and general conclusions 601-606 Tenancy, percentage of in four Illinois townships 81, 82 Toledo experiment field yields, long-time summaries of 234-237 Trade habits of farm households in four Illinois townships, analysis of 91-98 Transportation of milk, problems of in the St. Louis milkshed 422-468 Unionville experiment field yields, long-time summaries of 237-238 Urbana experiment field yields, long-time summaries of 239-240 Voluntary organizations, reasons for farm membership in studied 112-114 West Salem experiment field yields, long-time summaries of 241 Wheat, price trends of 6-7, 25, 26-27, 30, 31, 39, 73 production trends of 31-33, 48-50 response of to soil treatment compared with that of corn, of oats, of hay. . . 179-180 yields of on Illinois soil experiment fields, in 1934 156 in 1935 164-165 for four-year period 173-174 Wheat exports, trends in 57-59 Wool, price trends of 19-20, 30, 36, 39, 73 ZINC SULFATE, FURTHER STUDIES ON IN PEACH SPRAYS WITH LIMITED TESTS IN APPLE SPRAYS 129-144 See Contents 130 literature cited 140 spray injury reduced by use of 131-139 Summary and conclusions 138-140 422-4321935-37 ^0112019529269^