THESIS Date 0-7 639 May 17. 1985 IIUNNIWII"!Hillllllllllllllllm Will I 310616 4019 _‘ AA A‘— LIBRARY Michigan State University H f This is to certify that the thesis entitled COMBINING ABILITIES OF THREE QUALITY PARAMETERS IN FIVE SOFT WINTER WHEATS (TRITICUM AESTIVUM L. AESTIVUM) presented by Susan Gildehaus Aylward has been accepted towards fulfillment of the requirements for M.S. degree in Crop and Soil Sciences -‘QALZC %\ &‘\ Major professor Everett H . Everson MSU is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES \. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINE§ will be charged if 555} is returned after the date stamped below. _._._ c__________I- COMBINING ABIL PARAMETERS IN (TRITICUM A ITIES OF THREE QUALITY FIVE SOFT WINTER WHEATS ESTIVUM L. AESTIVUM) by Susan Gildehaus Aylward A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Crop and Soil Sciences 1985 Copyright by SUSAN GILDEHAUS AYLWARD 1985 ii ACKNOWLEDGEMENTS I would like to extend my sincere gratitude to all those who so graciously assisted in this project. They are: To my adviser and committee members, Everett Everson, Tom Isleib, and Barbara Sears. To the USDA soft wheat Quality Laboratory in Wooster, Ohio; especially Patrick Finney, Charles Gaines, and John Donelson. To Faz Haghiri, Bert BishOp, Ross Brazee, and Bob Fox of the Ohio State Agriculture and Research Development Center in Wooster, Ohio. To Steve St. Martin and Bruce Griffing of Ohio State University. To my helpful friend Wen for his valuable computer expertise. iii TABLE OF CONTENTS ACKNOWLEDGEMENTS . . . . . . . . . TABLE OF CONTENTS . . . . . . . . . . . . . . LIST OF TABLES . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . . . . Hardness Measurements PSI Test — Particle Size Index . . . . PSA Test - Particle Size Analysis . . ESI Test - Endosperm Separation Index Kernel Hardness Kernel Embryology and Tissue Development Starch Granule (Amyloplast) Development Protein Body Development . . . . . . . Starch/Protein Interface in the Mature Wheat Kernel . . . . . . . . . . . . Morphology of the Nature Wheat Kernel Physical Properties . . . . . . . . . Biochemical Basis of Hardness . . . . Genetics of Hardness . . . . . . . . . Factors Influencing Hardness . . . . . Diallel Mating Design . . . . . . . . . . Heterosis and Maternal Effects . . . . . 12 l6 18 20 22 22 30 36 37 4O 45 METHODS AND MATERIALS RESULTS AND DISCUSSION 0 F2 Generation Analysis of Variance F Particle Size F Particle Size F Particle Size Index F Endosperm Separation Index 2 CONCLUSIONS APPENDIX . . BIBLIOGRAPHY (PSA) (PSI) (PSA) Analysis . . Analysis . . (ESI) Analysis . . . Analysis Page 48 55 57 59 66 71 75 78 82 89 Table 1. Table 2. Table J. LIST OF TABLES Page Parental lines and 3 year averages (1979— 1981) for the endosperm trait Particle Size Index (PSI) . . . . . . . . . . . . . . 53 ANOVA including Combining Ability Analysis (34) for the 3 traits Particle Size Analysis (PSA), Particle Size Index (PSI), and Endosperm Separation Index (ESI) from a randomized complete block experiment in 2 locations in a 5 x 5 diallel cross of winter wheat for the F2 generation: Mean Squares . . . . . . . . . 58 Mean squares for general (goa), specific (sca), maternal, reciprocal combining abilities, and error for the F1 and F2 hybrids of the endosperm trait Particle Size (PSA), F2 hybrids of the endosperm trait Particle Size Index (PSI), and F2 hybrids of the bran trait Endosperm Separation Index (ESI), all involving a 5 parent diallel of winter wheat . . . . . 62 vi Table 4. Table 5. Table 6. Table 7. Page Estimates of general combining ability effects (gca) for the 3 characters Particle Size Analysis (PSA) in the F1 generation in 1 location and F generation 2 in 2 locations, Particle Size Index (PSI) in the F2 generation in 2 locations, and Endosperm Separation Index (ESI) in the F generation in 2 locations from all 2 possible crosses involving 5 winter wheat parents . . . . . . . . . . . . . . . . . . . 63 Estimates of Specific Combining Ability (sca) effects for the endosperm trait Particle Size Analysis (PSA) measured from Fl hybrids in one location and F2 hybrids over 2 locations; and for the endosperm trait Particle Size Index (PSI) measured from F2 hybrids in two locations from all possible crosses involving 5 winter wheat parents . . . . . . . . . . . . 64 Estimates of reciprocal combining ability effects for the trait Particle Size Analysis (PSA) measured from F1 hybrids, from all possible crosses involving 5 winter wheat parents where maternal and reciprocal degrees of freedom are pooled, df = 10, according to Griffing (33) . . . . . 65 Heterosis for the endosperm trait Particle Size Analysis (PSA) in the F2 generation when measured by the sum of the single cross hybrid (SC F2) — midparental differences (MPSC) for each parent . . . . . 7O vii Table 8. Table 9. Table Table Table Table 10. ll. 12. 13. Page Heterosis for the endosperm trait Particle Size Index (PSI) in the F2 generation when measured by the sum of the single cross hybrid (SC F2) — midparental differences (MPSC) for each parent . . . . . . . . . . . 74 Heterosis for the bran trait Endosperm Separation Index (ESI) in the F2 generation when measured by the sum of the single cross hybrid (SC F2) - midparental differences (MPSC) for each parent . . . . . . . . . . . . . . . . . 77 Parent and hybrid mean values where n = 3 for the endosperm trait Particle Size Analysis (PSA) measured in the F2 generation over 2 replications in 2 environments . . . . 82 Parent and hybrid mean values where n = 4 for the endosperm trait Particle Size Index (PSI) measured in the F2 generation over 2 replications in 2 environments . . . . . . ._83 —— Parent and hybrid mean values where n — 4 for the bran trait Endosperm Separation Index (ESI) measured in the F2 generation over 2 replications in 2 environments . . . . 84 Hybrid mean values where n — 4 for the endosperm trait Particle Size Analysis (PSA) measured in the F1 generation as a randomized complete block design over I replication . . . . . . . . . . . . . . . . 85 viii Table 14. Table 15. Table 16. Parental and hybrid mean values pooled over 2 replications in 2 environments, where n = 12, for the endosperm trait Particle Size Analysis (PSA) in the F2 generation . . . . . . . . Parent and hybrid mean values pooled over 2 replications in 2 environments, where n = 16, for the endosperm trait Particle Size Index (PSI) in the F2 generation . . Parent and hybrid mean values pooled over 2 replications in 2 environments, where n = 16, for the bran trait Endosperm Separation Index (ESI) in the F2 generation . . . . . . . . . . . . . . ix Page . 86 . 87 Figure Figure Figure Figure Figure Figure Figure Figure (A) LIST OF FIGURES Electron micrographs of fractured wheat kernels at magnifications used to study bran width . . . . . . . . . . . . . . . . Structure of a wheat ovule before fertilization . . . . . . . . . . . . . Changes in the shape of the developing wheat endosperm from anthesis to maturity . . . . . . . . . . . . . . . . . Mature grain showing protein content gradation from the subaleurone to the midendosperm region . . . . . . . . A— and B—type starch granules 36 days after anthesis . . . . . . . . . . . . . . Early stages of A—type starch granule formation five days after anthesis Starch and protein deposits 27 days after anthesis . . . . . . . . . . . . . . Differentiated aleurone cells and storage protein deposition 15 days after anthesis . . . . . . . . . . . . . . Page 11 14 14 15 15 17 l7 l9 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 9A. 9B. 10. ll. 12. 13. 14. 15. 16. 17. Development of rough endoplasmic reticulum and accumulation of storage protein in develOping wheat endosperm cells . . . . . . Mature grain showing membrane remnants, MR, and storage protein forming a matrix between starch granules . . . . . . . . . . Purified protein and starch preparations from a hard and a soft wheat . . . . . . . . Freeze—etch replicas of fractured endosperm tissue from hard and soft Wheat varieties O O O O O O O O O O O O O 0 Transverse section of a wheat grain . . . . Scanning electron micrographs of purified starch granules prepared under non-aqueous conditions from hard and soft wheat varieties . . . . . . . . . . . . . . . . . Scanning electron micrographs of hard and soft wheat kernels which have been fractured transversely . . . . . . . . . . Electron micrographs of the starch—protein interface . . . . . . . . . . . . . . . . . Graphic depiction of the Microtrac Particle Size Analyzer . . . . . . . . . . . . . . . Particle Size Analysis (PSA) general combining effects (gca) in the F1 vs. the F generation . . . . . . . . . . . . . 2 xi Page 21 21 26 27 28 29 29 35 54 69 ABSTRACT COMBINING ABILITIES OF THREE QUALITY PARAMETERS IN FIVE SOFT WINTER WHEATS (TRITICUM AESTIVUM L. AESTIVUM) BY Susan Gildehaus Aylward A ffifiwz parent diallel cross between diverse parents within the soft. wheat class yum; grown in 21 randomized complete block design over two locations in Michigan. Combining abilities of the traits Particle Size Index (PSI) , Particle Size Analysis (PSA) , and Endosperm Separation Index (ESI) were assessed on the F2 field data. The residual greenhouse generated Fl seed was also tested for PSA on the laser powered Microtrac Particle Size Analyzer. Results showed highly significant general and specific combining ability effects in approximately equal ratios for PSA and PSI in the F2 generation. The large Fl PSA maternal effect disappeared after one round of segregation. ESI showed highly significant general combining ability only, suggesting that it is a separate genetic trait. No F hybrid combination showed greater softness than 2 the parent cultivar Houser. Nicking for increased hardness was the main heterotic effect. A suggestion is made for PSA to replace PSI for kernel hardness determinations. INTRODUCTION Understanding the genetic consequences that result from breeding strategies such as inbreeding, crossbreeding and selection are the primary objectives of the plant breeder. Plant breeding programs ‘must accommodate both simple and polygenic inheritance of desired and undesired characters. Qualitative variation is characterized by discrete classes whereas quantitative variation forms a continuous array of metric values ranging from one extreme to the other. Nearly every function or organ including the most economically relevant in crop species shows inheritance of a quantitative nature. Even the expression of those traits strictly inherited qualitatively is in most cases modified by genes with small quantitative effects. Breeders need methods to identify good cross combinations from which to select superior genotypes even in such highly improved crops as wheat. For this purpose, the combining abilities of the parents of individual crosses are assessed with the eventual goal of channeling these better genetic combinations into the germplasm of a new economic variety. Soft wheat has been increasingly .used not for breadbaking, inn: for cookies, cakes, and crackers. This significant forty year trend has mandated the evaluation of breeding lines based on their performance in pastry products. It has also stimulated research into the chemistry of wheat and flour products. Soft wheat quality tests are primarily concerned with milling, baking, and hardness parameters. These parameters and total flour yield are essential factors in assessing soft wheat quality. At all stages of the breeding program, from early generations to advanced uniform: nursery trials, quality tests are performed. Early generation screening tests are performed on as little as twenty grams of wheat to determine protein content, hardness, and alkaline ‘water retention capacity. In later test generations where the wheat is abundant, cakes are baked in duplicate as the ultimate quality evaluation. In the soft white wheat breeding program at Michigan State University, breeding for quality has enjoyed a high priority. As a .result, some of 11MB industry's highest quality lines have been developed here. The Michigan State wheat breeding program maintains an ongoing liaison with the U.S.DHA. Soft Wheat Quality Laboratory (SWQL) at the Ohio State Agricultural Research and Development. Center (OARDC) in Wooster, Ohio. The SWQL functions as the necessary link between the breeder and the milling and baking industries. The purpose of this research project is to study time combining abilities of the quality parameter endosperm hardness, which the industry uses to grade wheat for industrial use. Three soft white wheats developed at Michigan State University, one soft white wheat developed at Cornell University, and one soft red wheat developed at Purdue University were used in this study with the ultimate breeding' goal. of determining 'which. contributed superior endosperm softness to its progeny and then incorporating superior gene combinations into time ongoing’ wheat hybridization system. LITERATURE REVIEW Hardness Measurements Wheat varieties differ ix) the vitreosity, hardness, and milling behavior of their mature air-dry endosperms. Biffin (10) used a subjective visual method of classification based upon the vitreousness of the kernel. The conventional wisdom was that the more vitreous, the harder time wheat. However, he found that vitreosity was not necessarily related to quality, since some soft wheats were translucent. Visual assessment, although not eliminated, has been repudiated since soft wheats of high protein content can appear at least somewhat vitreous. The reason for this vitreousness is the layer of protein rich endosperm cells just below the aleurone layer. Because of the relative lack of starch grains in that area, the protein layer appears uniform and thus vitreous. Soft wheats typically contain many starch grains in the outer endosperm cells, and these small grains scatter and absorb light which results in an opaque or floury looking grain. Since Biffin's time hardness tests have been devised that measure resistance by an individual grain to penetration by a stylus Cfifl and resistance to crushing (65). However, these methods are limited by interkernel variability and nonuniform endosperm (l3). Methods using bulk samples include those that measure resistance to grinding (56), time to grind a given sample (17, 57), and particle size resulting from grinding (81, 96, 97). Williams (93) and Simmonds (76) have both suggested using near infrared reflectance spectroscopy for hardness tests. O'Buchowski and Bushuk (68) compared several methods and found that some (ME the methods measured different properties of the grain. Those methods based on pearling the grain depended strongly (Ni bran properties while those based on mulling properties .measured tflua endospermi characteristics. They found that bran had a definite influence on the results of grain hardness evaluation. Particle Size Index (PSI) Test In the PSI test, a weighted sample of grain is ground and sifted under standard conditions and the weight of the sample passing through the sieve is measured. Soft wheats produce a large number of small particles with poor packing density and poor flow properties. This results in poorly sieving wheats. Hard wheats on the other hand tend to fracture into regularly shaped particles of larger size, w) I“ which pack well and exhibit good flow properties. This results in ease of sieving. Several grinders have been used such as the Hobart grinder, Labconco Heavy Duty mill, Quadrumat Jr., and the Brabender grinder. These grinders have varying types of grind and.sx> it has been difficult to standardize the PSI test. Researchers have been satisfied with the rank ordering of wheats, accepting the variability of the grinder as long as the order was similar. In a PSI comparison, Williams (94) established the degree to which individual laboratories were able to differentiate classes of wheat based on the PSI test. Three methods of grinding were used, the Labconco grinder, the Falling number KT—30, and Near Infrared Analysis (NIR). The Labconco grinder provided the least differentiation between samples and had the highest test coefficient of variation. NIR showed the highest differentiation among samples and the highest correlation to PSI values. Coefficients; of correlation among all cmdlaborators showed excellent agreement (r=.99 or better). the concluded that tflmz PSI test is 21 valid reproducible method of measuring hardness in wheat and of differentiating wheat types. He also pointed out that NIR is probably the most practical rapid test for wheat hardness since it has the advantage of providing protein and moisture data as well. Miller gt g1. (60) found that in a comparison between the Brabender grinding time method, PSI mett the high Yam stating apprecie paramete without found t (BFY) v first correla laborat increas Quadrum (Finney Quadrun PIOpert obtains flOur qualit that t PSI method and NIR, correlations between PSI and NIR were the highest. Yamazaki (101) summed up 40 years of PSI research by stating that the value of kernel hardness can best be appreciated when it is directly associated with a milling parameter rather than standing alone as a kernel property without any relevance to processing quality. It has been found that PSI correlates highly with break flour yield (BFY) which is the amount of flour liberated after the first set of break roles. Stenvert (81) found a correlation of r=.5 between BFY and PSI using ea Buhler laboratory mill. Yamazaki (101) found the r value increased to .85 using a Labconco grinder. With a Quadrumat Jr. mill the correlation was increased to r=.95 (Finney et al., unpublished). It. is concluded that the Quadrumat Jr. mill most closely approximates the milling properties relating to endosperm fracturing and the BFY obtained on an Allis Chalmers mill. Particle Size Analysis (PSA) Miller gt gt. (57), Yamazaki and Donelson (99), and Chaudary gt gt. (19) have reported on the relation between flour particle size and cake quality. They found that cake quality increased as flour mean particle size decreased and that this relationship was highly correlated. Methods used to measr Counter spectrosc Coulter distribul negative analysis negative PSI in B’ has cert over se‘ grade, Q An laser 11 Scattere in diame area of develope the sc distrib1 This is Placed proper A Singl. infer t microns to measure particle size distribution are by Coulter Counter analysis and by near infrared reflectance spectroscopy (NIR). Donelson and Yamazaki (23) used Coulter' Counter' procedures to analyze the particle size distributions of coarse soft wheat flours and found high negative correlation coefficients between Coulter Counter analysis and PSI. Using NIR, Williams (93) found a negative correlation of r=—.96 between particle size and PSI in Burr milled wheat. The results indicated that wheat has certain varietal fracture patterns that are consistent over several types of milling including Allis straight grade, Quadrumat, and Allis patent pin-milling. An alternate method of measuring particle size is with laser light scattering technology (91, 92). Light which is scattered by small particles ranging from 2 to 176 microns in diameter can be used to determine the cross sectional area of those particles. A measurement technique has been developed which determines the volume mean diameter from the scattered light, and variance of the particle distribution regardless of the nature of the distribution. This is accomplished with a unique filter, which when placed in the Fraunhofer diffraction plane, transmits the proper amount of light as a function of scattering angle. A single measurement of the scattered light flux is used to infer the particles' volumes with diameters from 2 to 176 microns. For an in depth discussion of elementary diffractj above mer Beft to a s separatit effect endosper flakes separati this seE aPproxim bran aft Wheats w microsco thicknes (See Fi< microns, Whether yield, milling‘ Flour Y: Jr- art particle diffraction theory and spatial filter theory, refer to the above mentioned articles (91, 92). Endosperm Separation Index (ESI) Test Before milling wheat, the grains are usually tempered to a standard moisture content which optimizes the separation of bran from endosperm. This procedure has the effect of making the bran tougher to grind and the endosperm softer. Hence the bran is primarily removed as flakes during the milling operation. The endosperm separation index (ESI) measures the ease or difficulty of this separation on an Allis mill. It is defined as the approximate quantity of endosperm remaining attached to the bran after break and reduction passes (94). Good milling wheats will have lower ESI values. Using scanning electron microscopy, Lineback gt gt. (51) showed that bran layer thickness differed significantly in some wheat cultivars (see Fig. 1). Mean bran thickness ranged from 50 to 76 microns. They suggested further research to establish whether correlations exist between bran thickness and flour yield, cleanness of endosperm separation and ease of milling. ESI is estimated fix: a microtest, the "Percent Flour Yield (PFY)", in which flours ground on a Quadrumat Jr. are sieved through a 54—mesh screen. The bran particles remaining on top of the 54—mesh screen are measured . purposes ( microtest E81 and I classed a two diffe and kerne 10 measured and then calculated as the PFY score. For our purposes of intuitive understanding, I shall refer to this microtest as the ESI test. There is no correlation between ESI and PSI indicating again that bran separation can be classed as a: separate milling parameter from PSI. Hence two different traits can be measured, ie. milling potential and kernel hardness. 0 Figure l ll Figure 1. Electron micrographs of fractured wheat kernels at magnifications used to study bran width. Lineback gt gt. (1978) (51). Kernel H; er; The separate male gam sac thrc fuses wj gamete j 2). The latter i Chromosc from th hybrids genetic Whether triploh from thi genes c eXPrBSS Ones (4 Protein 1977 (7 Af Synchro 12 Kernel Hardness Kernel Embryology and Tissue Development The wheat embryo and the endosperm arise from two separate fertilization events. Fertilization occurs as two male gametes from one pollen grain penetrate to the embryo sac through the micropylar opening, whereupon one gamete fuses with the haploid egg megagamete and the other male gamete fuses with the two haploid polar nuclei (see Fig. 2). The former develops into the diploid embryo and the latter into the triploid endosperm containing three sets of chromosomes, two derived from the maternal parent and one from the paternal parent. The endosperm of reciprocal hybrids between lines with differing alleles will differ in genetic constitution al al a2 and a2 a2 a1, depending upon whether the maternal parent possessed al or a2. Thus the triploid inheritance adds to the genetic complexity arising from the allohexaploid nature of wheat (31). Products from genes carried on the maternally derived chromosomes are expressed in twice the dosage as the paternally derived ones (44). This has been demonstrated in gliadin storage proteins by Mecham gt gt. 1978 (55) and Qualset and Wrigley 1977 (74). After the formation of the triple fusion nucleus, synchronous free nuclear division, cellularization, and maturatic begin dex 10-14 da approximi The majo granules soluble leaves a These c2 into the cheeks 1 small st when the Later t) whereup< The res Cells In then e Finally grains which 1 Practic acCumu] squeleze and th protei] 13 maturation occur (see Fig. 3). Endosperm storage products begin developing when the tissue is completely formed, ie. 10—14 days after fertilization; and they continue for approximately six weeks whereupon the grain is ripe (72). The major constituents of wheat endosperm cells are starch granules and storage protein. Starch is synthesized from soluble carbohydrates which originate in the stems and leaves and which are conducted to the base of the ovary. These carbohydrates then traverse the ovary and, diffuse into the endosperm (72). It is in the region of the two cheeks that starch grains are first observed. First the small starch bodies are seen near the endosperm cell nuclei when the endosperm has just filled the embryo sac cavity. Later they are seen at many cytoplasmic points of origin, whereupon they increase rapidly in number and size (72). The reserve. cells are filled in varying stages. Those cells most distal to the embryo are initially filled, and then eventually those cells adjacent to the embryo. Finally, the kernels become densely packed with starch grains of various sizes except for the aleurone layer, which remains devoid of starch grains (see Fig. 4). In practically all of the endosperm cells in which starch accumulates, the protoplasmic contents die and become squeezed into the spaces between the densely packed grains, and then the cytoplasmic remnants appear embedded in a protein matrix (see Fig. 5). Figure 2. Figure 3. Figure 2. Figure 3. Structure of wheat ovule before fertilization. Antipodal cells A. Nucellar epidermis NE. Nucellar tissue N. Central endosperm mother cell C. Polar nuclei P. Synergids 8. Egg cell E. (Reproduced with permission from Mares gt al. 1975). Simmonds gt al. 1981 (79). _ 1 . mm mum, @‘3 i‘m . . ’ . Changes in the shape of the developing' wheat endosperm from anthesis to maturity. p = pericarp. e = endosperm. m1 = meristematic layer. twc = thick—walled cells. a1 = aleurone. (Reproduced with permission from Fyers. 1970. 1974). Simmonds gt al. 1981 (79). Figure 4. Figure 5. Figure 4. Figure 5. 15 Mature grain has gradation in protein content from the subaleurone to the midendosperm region of the cell. Simmonds gt gt. 1981 (79). Thirty-six days after anthesis, A- and B- type starch granules completely fill the long prismatic endosperm cells; storage protein forms a matrix between them. Simmonds gt gt. 1981 '(79). Star Star storage 5 as two 1 develop A-type microns rarely e that amy days of number < starch granule the tot B-type 25-50% Fig. 7 B"type eDdOSpg their EXperi. 26). 16 Starch Granule (Amyloplast) Development Starch, the principal component of carbohydrate storage in the mature seed, occurs in the mature endosperm as two populations of granules differing in size. They develop within plastids bounded by double membranes. The A-type starch (primary granules) range from 20 to 45 microns in diameter and B-type starch (secondary granules) rarely exceed 10 microns (27, 28) (see Fig. 6). It appears that amyloplasts reproduce by division during the first six days of endosperm development (18, 41), after which the number of A-type amyloplasts remains constant (14). A—type starch granules contribute 3% of the total number of granules in the mature endosperm, but account for 50-75% of the total starch weight because of their size (27, 28). B—type granules account for 97% of the numbers and from 25-50% of the weight of starch in the mature grain (see Fig. 7). Briarty gt “gt. (15) estimated the numbers of B-type granules to be 900 per cell, or about 94 million per endosperm. Initially, B—type granules are spherical, but their final shape is determined by the dense packing they experience as the grain reaches physiological maturity (24, 26). They form by budding off of A—type amyloplasts. Figure 6. FiQUre 7 Figure 6. Figure 7. l7 iwgwwyvb «s at. «v f ‘ ‘n-gnn ( * r r5 , ‘ur '2‘, ,_ Five days after anthesis, intact nucellar epidermal cells (NE), large, open, and highly vacuolated endosperm cells, and early stages of A- type starch granule formation. Simmonds gt gt. 1981 (79). Twenty-seven days after anthesis, starch and protein deposits almost completely fill the cell volume. Small B— type starch granules are greatly in evidence. Simmonds gt gt. 1981 (79). Pro‘ The from 0.3 Their a mature compress bodies a matrix c which ti stated ' its pro- the co lipopro body be endOSpe amino 5 still t WOrld. 18 Protein Body Development The protein bodies in endosperm are spherical, ranging from 0.5 to 15 microns in diameter (42) (see Fig. 8). Their shape becomes extremely distorted however in the mature tissue, because of kernel desiccation and compression by starch granules. Many smaller protein bodies are fused during growth, so that the storage protein matrix of the mature wheat grain tends to be a continuum in which the starch grains are embedded (9). Adams gt gt. (2) stated that wheat differs from most other cereals in that its protein bodies cannot be recognized at maturity due to the compression and fusion that have occurred. The lipoprotein membranes originally surrounding the protein body become enmeshed into the general protein mass. Wheat endosperm has poor nutritional development in the essential amino acid lysine, but because of its abundance wheat is still the most important source of protein for much of the world. Figure 8. Figure 8. 19 ”004’ ‘ {‘1 _V .4 a2." Fifteen days after anthesis, aleurone cells A are clearly differentiated and storage protein deposition is well under way. Simmonds gt gt. 1981 (79). its The the prot the sta These c derived such as soluble (77) (SI and pr membrane the prc Barlow desicca dependi hardnes Compres the Ce remnant 9F), E physics It the ma that t biOCheI 20 Starch/Protein Interface in the Mature Wheat Kernel The molecular interface between starch granules and the protein matrix is complex. Carried on the surface of the starch granule are amylose and amylopectin chains. These chains are in molecular contact with components derived from the desiccated remnants of the plastid stroma such as glucose and its short polymeric chains, water soluble proteins, and remnants of the endoplasmic reticulum (77) (see Fig. 9C). This material is surrounded by lipid and protein components derived from the amyloplast membrane, and is in close contact with membrane remains of the protein bodies and the rough endoplasmic reticulum. Barlow .2:..§l~ (7) have found that upon dehydration and desiccation, an adhesive bond forms that varies in strength depending upon the inherent and genetically controlled hardness of the grain. Simmonds (76) found that the compressed matrix consists of remnants of protein bodies, the cell nucleus, endoplasmic reticulum, and cellulosic remnants of the cell walls at cell boundaries (see Fig. 9F). Both starch granules and protein bodies are separated physically from contact by lipoprotein membranes. It is not known how intact these membranes remain in the mature cell. It is at this starch/protein interface that the prevailing hypothesis of grain hardness offers biochemical explanations. 21 .")D “~' " u: $1 a r- 1 Figure 9. Development of rough endoplasmic reticulum and accumulation of storage protein in developing wheat endosperm cells from about eight days after anthesis to maturity. A) Ten days after anthesis. Extensive development of rough endoplasmic reticulum RER. Scale bar = ]. pm. B) Seventeen days after anthesis. Close association of developing' protein bodies PB with rough endoplasmic reticulum. Scale bar = 10 um. C) Twenty days after anthesis. Distended section of rough endoplasmic reticulum filled with protein. Scale bar = 1 pm. D) Seventeen days after anthesis. Numerous protein bodies containing osmiophilic inclusions I are present in the endosperm. Scale bar = 1 pm. E) Thirty—one days after anthesis. Trapped membrane remnants form osmiophilic zones where several protein bodies have fused. Scale bar = 1 pm. F) Forty days after anthesis. Mature grain shows membrane remnants MR and storage protein compressed and forming a matrix between starch granules S. CW endosperm cell wall. Scale bar = 1 pm. Simmonds gt gt. 1981 (79). The grain we contains 10% and conditic endosper amounts However, techniq1 Th1 to the . endOSpe endOSpe cell V0 for end 12% prc hardnes with a Kernel which 6 22 Morphology of the Mature Wheat Grain The starchy endosperm comprises roughly 80—86% of the grain weight (40). At physiological maturity, wheat grain contains about 40% water, a figure which drops to between 10% and 15% at harvest depending upon environmental conditions. Milling techniques yield between 70—80% endosperm and since these flours usually contain small amounts of bran, the theoretical purity is never achieved. However, with high quality grain and good milling techniques, it has been and continues to be improved. There is a gradation in protein content from the outer to the inner endosperm cells, which results in a nonuniform endosperm (25). In fact, approximately 25% of the endosperm protein is contained in the 11% outer endosperm cell volume of the subaleurone region (30). Typical values for endosperm composition are 70% starch, 13% moisture, and 12% protein by weight. Whole grain density increases with hardness as reported by Stenvert gt gt. (82). Physical Properties One objective of wheat breeding is to develOp kernels with appropriate hardness for the end use of the grain. Kernel hardness can best be characterized as a syndrome which encompasses a number of physical properties including flour re wheatmea also re separate sifting resistar tend to opposed flour. soft an there i residue wheat. during relativ process more we Convert reason hardneg indust] SUffic: fermen. f0rmat granul 23 flour release on a single grinding, specific volume of the wheatmeal and resistance to abrasion (66). Hardness is also reflected in how cleanly the grain constituents separate, fix: the resulting fragment sizes, and i1) their sifting behavior. Harder wheats are associated with higher resistance to abrasion, grinding, and crushing and they tend to have a granular meal with a sandy texture as opposed to soft wheats which yield a characteristic fine flour. Figures 10A and 108 show starch granules from a soft and hard. wheat respectively. It can be seen that there is more non starch residue in the hard wheat. This residue may contribute to the sandy texture of the hard wheat. Hard wheats incur more damaged starch granules during milling operations than do soft wheats. The relative proportion of damaged starch granules affects the processing behavior of the flour. Damaged granules absorb more water than intact granules and thus are more readily converted into sugars by beta-amylase. It is for this reason that the proportion of damaged starch and hence the hardness of the grain from which it was milled have industrial importance. Breadbakers seek a flour with sufficient starch grain damage to allow expansion during fermentation, but not enough damage to interfere with the formation of the continuous protein network around the granules (81). Soft wheat bakers seek less starch grain damage unfermex A occurs 1 wheats becomes those b content through content of the that t] from t} this d adherir commun: and po( bran 5 manipu T after increa Becaus matriX granu] rEmair 24 damage in their flours since their products are largely unfermented. A significant difference in fracturing properties occurs between hard and soft wheats (see Fig. 11). In hard wheats the boundary between cell wall and cell contents becomes the line of weakness resulting in fracture along those boundaries. This occurs because their endosperm cell contents are very hard. Soft wheats tend to fracture through the cells themselves because the endosperm cell contents are not as tightly held together. Another effect of the shearing forces during milling of a hard wheat is that the subaleurone endosperm tends to separate cleanly from the aleurone cells (see Fig. 12). In the soft wheats this does not occur and so starchy endosperm is left adhering to the bran. However, Andrews (personal communication) has observed superior milling soft wheats and poor milling hard wheats leading to the hypothesis that bran separation and kernel hardness can be genetically manipulated as different traits. The increase in damaged starch of hard wheats occurs after the initial splitting along the cell wall plane when increasing pressure causes fracturing across the cell. Because the starch granules are so firmly embedded in the matrix, the later cleavage results in fractured starch granules and storage protein (see Fig. 13). The cell wall remains attached to the released particles because cell walls ar occurrin little 2 that th material sieving soft whe break a sufferi through granule wheats and fr storage firmly 25 walls and contents form a coherent whole with breakage occurring along the weakest point. In soft wheats there is little adhesion between cell walls and cell contents, so that the walls tend to mill into separate sheets of material which can sometimes cause problems in subsequent sieving and dressing procedures. The starch granules of soft wheats are not as tightly held in the matrix, and they break away more easily under increasing pressure, thereby suffering less damage (see Fig. 14). The fracture occurs through the cell contents and around individual starch granules. Symes (83) found that during pin—milling, soft wheats fractured into discrete particles of storage protein and free starch granules, while hard wheats yielded a storage protein fraction which contained starch granules firmly attached to the protein matrix particles. Figure l FiQUre *‘ “b ’~*I a ‘9‘ to: 4.. ‘ "412,41“ -fl' 7’ Figure 10A. Purified protein and starch preparations from the soft wheat Arthur. Figure 10B. Purified protein and starch preparations from the hard wheat Eagle. Figure 1 Figure 11. 27 Freeze—etch replicas of fractured endosperm tissue. (a) Hard wheat (or. Gabo) showing extensive cross—fracturing. (b) Soft wheat (cr. Soft Falcon) showing exposed starch granules revealed by surface fracturing. (S = starch granules; P = protein; O—«-shows direction of shadowing.) Simmonds 1974 (78). Figure 1 28 (u‘ICJ l|'|(i In“ {KN _|,.~ run. “In!” . I'll! AHlHM .A-H' \ultru- UuNuH Wumh ~A\~t~ Figure 12. Transverse section of a wheat grain. MacRitchie (1980) (53). Figure 1 Figure 29 Figure 13. Scanning electron micrographs of purified starch granules prepared under non—aqueous conditions from soft (a) and hard (b) wheat varieties. Simmonds 1974 (78). Figure 14. Scanning electron micrographs of hard (a) and soft (b) wheat kernels which have been fractured transversely. Simmonds 1974 (78). Ma being differs have 1c the I chemice Barlow light 1 transm. micros staini granul rapid solubl gave Evider retict that diffe] their and t COnst; Samp1. c35085. 3O Biochemical Basis of Hardness Many researchers have attempted to explain hardness as being based upon biochemical and physicochemical differences within the kernel (8, 76, 77, 82, 96). They have looked at the interrelationship between the starch and the protein constituents both microscopically and chemicalLy in order to explain differences in hardness. Barlow (8) obtained information on this interface using light microscopy, soluble protein extraction and staining, transmission electron microscopy, scanning electron microsCOpy, freeze etching, and fluorescent antibody staining. He found that immediately surrounding the starch granules was a region of water soluble proteins capable of rapid swelling upon hydration (see Fig. 15). These water soluble proteins were associated with carbohydrates which gave rise to glucose upon hydrolysis. He also found evidence that amyloplast membrane, and endoplasmic reticulum residue exist around the starch granules, and that starch granules prepared from hard and soft wheats differed in the amount of protein material adhering to their surfaces. Furthermore, between the membrane remnants and the granules were spaces of varying width, relatively constant within a sample, but variable between different samples. Freeze fracturing revealed that the fracture crossed through the starch granule in proportions that roughly the PSI starch wheats the sax tests floresc surrour demonst than it A amount: by wei protei total equiva test ( fl. (7 found Carbof Small also COmpOI Consh Water 31 roughly paralleled the hardness of the wheat as measured by the PSI test (see Fig. 11). Micropenetrometer tests (8) on starch and matrix material purified from soft and hard wheats revealed that the hardness of each constituent was the same for both classes of" wheat. Immunofluorescence tests on microtomed endosperm tissue stained with florescein confirmed this zone of water soluble material surrounding the starch granule. Jones and Dimler (43) demonstrated a higher proportion of water soluble proteins in starch rich fractions obtained by air classification than in the fine particle protein rich fractions. A milled hard wheat starch granule carries large amounts of matrix resulting in protein levels of up to 8% by weight. Included are both matrix and water soluble proteins. Simmonds (78) found that extractions of the total quantity of water soluble material were roughly equivalent to the kernel hardness as measured by the PSI test (see page 5 for explanation of PSI test). Simmonds gt gt. (77) further examined the water soluble material. They found a carbohydrate to protein ratio of about 2:1. The carbohydrate ‘was composed. primarily! of glucose, although small quantities of xylose, arabinose, and. mannose were also present. Although they did not find any specific compounds at the starch protein interface that could be considered an adhesive, they found a larger amount of the water soluble material. It has been pointed out by Yamada and 01( very st Us hardnes from 5 protein the pr either differ hardne yielde wheats as a starch cleanl wheat. throng geneti 1 Starct of gr PrOtef COmp0} Storag endOS] appea 32 and Olden (97) that carbohydrates and glycoproteins act as very strong adhesives in biological systems. Using nearly isogenic lines differing in kernel hardness, Wrigley (96) extracted the water soluble proteins from starch granules that had been purified from storage protein by solvent flotation. Electrophoretic analysis of the protein matrix material failed to show differences in either the gliadin or glutenin groups, ruling out differences in the matrix composition as an explanation of hardness. However, starch granules from the hard wheats yielded two to three times more material than the soft wheats. He suggested that the water soluble material acts as a cementing substance between storage protein and starch. When adhesion is weak, starch is liberated more cleanly with less adhering protein as in the case of a hard wheat. It appeared that it was a varietal trait, and that through the amount and composition of this material the genetic control of grain hardness is expressed. The above evidence suggests that the adhesion between starch granules and the protein matrix is the determinant of grain hardness rather than the composition of the protein matrix or the intrinsic hardness of the separate components. It appears that the bonding between starch and storage protein may be stronger in hard wheat, so that the endosperm cell contents comprise a coherent whole. It appears to be a quantitative rather than qualitative differe materia (65) t sectior compone surroux 11 hypotht postul that i then v would entrap would protei Spaces latte: electx E Stare} Yet S Cemen. in he exPla Stare iSOla 33 difference, with the greater amount of water soluble material acting as a cementing force. According to Moss (65) the major gene discovered by Symes (85) (see next section) must control the production of one, or a group of components, in the zone of water soluble material surrounding the starch granules. In addition to the above starch protein adhesion hypothesis of grain hardness, Stenvert and Kingswood (82) postulated that the adhesion concept is unnecessary and that if the protein matrix as a whole is not continuous, then variation in the strength of the endosperm structure would result. They stated that starch granules physically entrapped by a continuous matrix as in the hard ‘wheats would result in a different separation of starch from protein than would a discontinuous structure with air spaces unfilled by matrix, as is found in soft wheats. The latter would release starch granules. They used scanning electron microscopy to support these suggestions. Both hypotheses suggest variations in adhesion between starch and the protein matrix as explanations for hardness, yet Simmonds and others (8, 78, 96) postulated a specific cementing substance which is present in greater quantities in hard grain, and Stenvert and Kingswood based their explanation on the degree of physical contact between starch and matrix. This "cement" has not as yet been isolated, and further work needs to be done in this area to establi and K: tentat: 34 establish conclusive proof for this hypothesis if Stenvert and Kingswood's hypothesis tentative. is to be considered only Figure Figure 15. 35 Electron micrographs of starch—protein interface: (a) general View showing membrane residue (MR) surrounding each starch granule (S); (b) higher magnification showing membrane residue in greater detail; and (c) micrograph illustrating network of presumed water-soluble material (WSM). Simmonds 1974 (78). ICE 1 hardne or m1 phenom 85) t convez This and 56 genes hardn relat subje concl betWe unabl SubSe trait reCox stri< Signi 36 Genetics of Hardness In 1908, Biffin stated that cultivars of wheat vary in hardness (10). Prior to 1935, workers postulated one, two or multiple gene differences in hardness (l). The phenomenon of grain hardness has been shown by Symes (84, 85) to be genetically controlled. By backcrossing he converted a hard cultivar into a soft one and vice versa. This effect appeared to be controlled by one major gene, and several modifying genes. He stated that grain hardness of a new wheat cultivar would be influenced by the hardness of the donor parent and by the degree to which modifying genes are carried over. Worzella (95) concluded that hardness was inherited. as a quantitative character' with relatively few genes involved. Millington and Remilton (62) and Nakagawa (66) used subjective, visual inspection and reached differing conclusions. Thompson and White (88) tried to distinguish between hard and soft wheats by using percent bran but were unable to reach any conclusions. These two criteria have subsequently been shown to be independent of the genetic trait hardness. In Syme's study (84) the exact parental type was recovered much less frequently than if the difference was strictly due to monofactorial inheritance. He found four significantly different levels in the soft class, indica which that i for tt hardne preser contrc chromc I Chine: posses involx chara< locate Zhiror betwer and t] self1 on t} chrom. envir. inter 37 indicating that at least two additional genes are present which influence grain hardness differences. He concluded that it is possible that the same major gene is responsible for the large effect in all cases, with different levels of hardness being the result of modifying genes. Konzak (47) presented other evidence suggesting that the genetic control of hardness is complex and involves more than one chromosome. Using substitution lines of the cultivar "Hope" in Chinese Spring, Law gt gt. (50) showed that chromosome 5D possessed a single gene Hun for grain hardness. Lines involving ditelocentric SD's failed to segregate for this characteristic, strongly suggesting that this gene is located on the short arm of chromosome 5D. Ternovkaya and Zhirov (87) discovered a positive functional relationship between the number of D chromosomes (from 7—14) in the seed and the density of the endosperm by studying the progeny of self—pollinated pentaploid wheat hybrids. They speculated on the possibility of creating kernels with controlled chromosome sets. Factors Influencing Hardness Baenziger‘ gt gt. (4) found highly significant environmental, genotype and genotype by environment interaction components of variance for hardness as measured by the 12 env enviro cultiv qualit suffic Parish temper They i atmosp wheat was hi hardne grain E kernel softer increa micro; Contex reSpox BeCau: endOs} Yamaz, Shouh 38 by the PSI test. They tested 22 soft wheat cultivars over 12 environments and found that the cultivar means from an environment were significantly correlated with the regional cultivar means. They concluded that for preliminary quality evaluations, data from one environment is sufficient for ranking cultivars with respect to PSI. Parish and Halse (70) have shown the influence of temperature and humidity during ripening on grain hardness. They found that hard wheat became harder in a more humid atmOSphere during the later stages of ripening, while all wheat became harder if the temperature during this period was higher. They found that even light rain affected grain hardness with especially dry sites and years producing grain lower in PSI than other sites. Several workers have found PSI to be affected by kernel moisture content (30, 35, 45, 63). In each case, softening of the grain was reported as moisture levels increased. Grosh and Milner (35) found by micropenetrometer tests that hardness decreased as moisture content increased. Orth (69) found that soft wheats responded more to moisture content than did hard wheats. Because moisture affects the PSI test, with higher endospernl moisture content resulting in a softer‘ wheat, Yamazaki gt; gt. (101) recommended that PSI measurements should be made at similar grain moisture. Alternatively, data 5 making H wheats for ti found granul hardne soft wheats correI Ofa incre urea ( 39 data should be adjusted to a uniform moisture basis before making measurements. Historically, it was generally accepted that hard wheats were higher in protein and that this was the reason for their hardness. However, researchers (89, 95, 101) found no correlation between protein content and granularity. Trupp (89) concluded that protein content and hardness have different genetic causes. Symes (84) found soft wheats of higher protein content than some hard wheats. Sampson (75) showed the absence of any significant correlation between hardness and protein content in lines of a soft by hard hybrid. Altman gt gt. (3) found that increasing grain protein percentage with foliar applied urea did not affect kernel hardness. Dialle P widest dialle crosse can b elemeI diag01 full where recip full the r inher Popul Parer quant the Prime Pare: Perfl abil inte rela 40 Diallel Mating Design Perhaps no other mating design has enjoyed the widespread use and controversy in plant breeding as has the diallel cross. A diallel is the set of all possible crosses between a given number, p, of parental lines. It can be represented by an r) x r) matrix, where the ijth element represents the hybrid between and the leading diagonal (ii) elements represent and selfed parents. A full diallel includes the parental lines, a set of crosses where for each ijth element i 03...“. MO “CGOHQAH H HM> w .%Hm>fluoommou .mHo>oHIa He. can mo. one Do moccoamacmflm owocom . «in... Hos.m Hos.m ram. gonna mm.m~ .mw.s «.Na.wH am Ammo mew aam mam as am am om> a .sam.ms «.Hm.ms os.ma om.m mm.mm o Hmoondflomm was was am as am wm> w ..mm.mHH «.ms.mma Na.em om.m os.aa a Hangman: wow wow mom as wee om> a .«Hm.sa .*H~.sv .amm.sam Hm.m .«om.som m mom «ma sea was was ass oa> a ..ma.sa .amv.ma ..mHH.m~m .ma.mm ..am.oam a «ow ma mommono mcofid Honor Hones ao>CA Ham .AHmmV Euommoocm uflmuu coup may wo woflnbxc N xmocH cofluwnmmom a one .AHmmv xmecH mmflm maofluwmm ufimmu Euwmmoocm on» mo mcfiunxn m Euodmopco ogu mo momma»; mm cam Hm may now bongo paw .moflufiafibm acacfinaoo m .Aammv wwwxfiocm oNflm oaoflaumm uflouu Hmooumfloou .Hmcumuofi rfimowv camflcmmm rAmomV Homecom Mom moundqm coo: .m wHQMB Table 63 Table 4. Estimates of general combining ability effects (gca) for the 3 characters Particle Size Analysis (PSA) in the F1 generation in 1 location and F2 generation in 2 locations, Particle Size Index (PSI) in the F2 generation in 2 locations, and Endosperm Separation Index (ESI) in the F2 generation in 2 locations from all possible crosses involving 5 winter wheat parents. PSI ESI F2 PSA Fl PSA Gl —2.98 0.83 3.15 0.53 c2 —o 68 o 39 1.19 —1.44 c3 1.59 —o.32 —2.17 1.42 G4 0.86 -0.67 —0.97 1.30 c5 1.20 —o.21 —1.2o -l.82 SE(gi-gj)+ .62 .29 .83 .53 LSD 1 54 74 2.08 2 60 +Standard error of the difference between two effects. Table 64 Table 5. Estimates of Specific Combining Ability (SCA) effects for the traits Particle Size Analysis (PSA) measured from F1 hybrids in one location and F2 hybrids over 2 locations; and for the trait Particle Size Index (PSI) measured from F2 hybrids in two locations from all possible crosses involving 5 winter wheat parents. Parents 2 3 4 5 1 Fl PSA -3.27 2.33 -0.25 1.17 F2 PSA -5.45 2.82 2.37 0.24 PSI 4.22 —l.90 -1.37 -0.97 2 Fl PSA -l.22 2.34 2.12 F2 PSA 1.84 1.72 1.86 PSI -2.42 -0.74 -1.08 3 Fl PSA 0.03 —l.l7 F2 PSA -3.34 -1.34 PSI 2.18 2.11 4 Fl PSA -2.l4 F2 PSA -0.78 PSI -0.08 tandard ErrorJr Fl PSA F2 PSA PSI SE(sij-sik) .75 1.18 0.87 SE(sij-skl) .53 .83 0.62 LSD 2.60 2.08 1.54 +Standard error of the difference between two effects. Tab] 65 Table 6. Estimates of reciprocal combining ability effects for the endosperm trait Particle Size (PSA) measured from F1 hybrids, from all possible crosses involving 5 winter wheat parents where maternal and reciprocal degrees of freedom are pooled, df = 10, according to Griffing (33). Parents 1 2 3 4 5 Mean 1 7.98 5.82 -O.37 1.24 3.67 2 -7.98 -1.11 -2.14 -O.45 -2.70 3 -5.82 1.11 1.04 0.76 -0.73 4 0.37 2.14 -l.04 -l.73 -0.07 5 -1.24 -0.45 -0.76 1.73 —0.18 Mean -3.67 2.70 0.73 0.07 0.18 0 mate Sinc inhe cont cont gen< eff and Fig bet ger 66 F2 Particle Size Analysis (PSA) Table 3 shows that in the second generation the maternal and reciprocal effects have all but disappeared. Since the large Fl maternal effect vanishes after one round of random segregation, this effectively rules out maternal inheritance for this endosperm trait. Hence the genotypic contribution of the endosperm trait particle size is controlled predominantly by the triploid endosperm tissue genotype. Selection for PSA would be correspondingly more effective in this generation than in the F1 since more gca and sca are apparent after the first round of segregation. Figure 17 shows visually that there is no relationship between the gca's of the two generations. In the F2 generation, the gca variance is again almost equal to the sca variance (see Table 3), with the sca having slightly more effect. This suggests that the character is controlled by additive and non—additive gene effects in roughly equal proportion, with non—additive gene action having slightly more effect. In the F2 generation, Table 4 reveals that the largest PSA gca effects were exhibited at 3.15 by Arthur, which had also shown the largest maternal and reciprocal effects in the F1 (see Table 6). Along with Arthur, Tecumseh at 1.19 was the only other parent exhibiting positive gca as seen in Table 4. The fact that parents exhibit negative gca sugg alle Hi 67 suggests that they had a higher proportion of negative alleles for this character. Houser had the highest negative gca at -2.17 followed by Augusta at -1.20. Table 5 reveals that the crosses Houser/Arthur at 2.82 and Hillsdale/Arthur at 2.37 exhibited the greatest amount of sca. An intermediate amount was shown by Houser/Tecumseh at 1.84, Hillsdale/Tecumseh at 1.72, and Augusta/Tecumseh at 1.86. The reciprocal effects were not significant for this trait and therefore will not be discussed further. When mean performance was measured against the midparent for an estimate of heterosis as in Table 7, the combinations using Arthur, Tecumseh, and Houser had the largest positive change, with Arthur predominating. All five cultivars had relatively large effects with Hillsdale exhibiting the least amount of change. Positive heterosis however, a term typically associated with yield components in wheat, is not in fact reflected in these results. Since a higher value reflects a harder wheat as in this case for PSA, what is reflected is actually negative heterosis. In this case the additive and non-additive effects are combining to nick for greater hardness, not softness. The softest wheat according to inean parental performance is Houser at a PSA of 56.7 which shows a positive change of 4.1 in hybrid combination. The next softest wheat was the cross Hillsdale/Houser with a mean PSA value of 58.9. Hillsdale show the least amount of negative heterosis with TGCI 68 a change of 2.53, but even in spite of Hillsdale's high relative negative heterotic value, Houser remains the softest wheat. This is due to the hybrids' greater tendency toward hardness as measured by the amount of heterosis over the midparent. Moreover, according to gca effects, Houser would be a good Choice due to the absence of additive genes for hardness. Negative sca effects would be desirable for soft wheats also as in the case for Tecumseh/Arthur at -5.45 and Hillsdale/Houser at -3.34. Figu 69 4 F1 PSA GCA -1F I _3 J 1 l F2 PSA GCA Figure 17. F1 vs. F2 Particle Size (PSA) general combining ability effects (gca). Table Pare 1 SC 70 Table 7. Heterosis for the endosperm trait Particle Size Analysis (PSA) in the F 2 generation when measured by the sum of the single cross hybrid (SC F2) — midparental differences (MPSC) for each parent. Parents 1 2 3 4 5 2(SC FZ—MPSC) 1 SC F2 63.2 64.3 69.2 69.9 67.5 +5.65 MPSC 63.0 60.0 63.3 62.0 2 SC F2 62 7 66 2 67.3 67 2 +4.38 MPSC 59.7 63.0 61 8 3 SC F2 56 7 58.9 60 6 +4.10 MPSC 60.0 58 8 4 SC F2 63.3 62.4 +2.53 MPSC 62.1 5 SC F2 60.8 +3.25 I’x} effec mater sign: gene Tabl exhi Tecr 71 F9 Particle Size Index (PSI) Analysis The F2 PSI showed highly significant gca and sca effects in a gca:sca ratio of 1:1 and showed nonsignificant maternal and reciprocal effects. This means that there are significant sources of both additive and non—additive genetic variability for the trait in equal importance. Table 4 reveals that Augusta at 1.20 and Houser at 1.59 exhibited the greatest gca effects and Arthur at -2.98 and Tecumseh at —0.68 showed the least. Tecumseh/Arthur at 4.22 exhibited the greatest sca effects, with Hillsdale/ Houser at 2.18 and Augusta/Houser at 2.11 intermediate between the former combination and the remaining crosses as seen in Table 5. The mean performance measured against the midparent values for this trait also showed negative heterosis, as did PSA (see Table 8). In this case, a negative value actually reflects negative heterosis, so that the least negative result reflects the maximum tendency toward softness. Again, Arthur at —3.29 showed the greatest tendency toward lower PSI values and hence harder wheats. Tecumseh at —l.98, Houser at —2.02, and Augusta at ~l.72 were intermediate with Hillsdale at -0.70 showing the least amount of negative heterosis. The softest performing wheats were the parent cultivar Houser at 40.95 with the highest gca value of 1.59 followed by the crosses Hillsdale/Houser at 40.25 with an sca of 2.18 and ‘ Augi the Tec1 as he 72 Augusta/Houser at 40.51 with an sca of 2.11. The low value of Arthur decreases the softness of the hybrids in spite of the fact that the highest sca occurred in the cross Tecumseh/Arthur at 4.22. Augusta was the fourth softest wheat at 39.33. Characters normally dealing with the fitness of the plant are those most likely to exhibit heterotic improvement, such as yield and yield components. It might be expected that a harder endosperm would be more likely to contribute to the overall evolutionary fitness of the plant in terms of greater hardiness under adverse environmental conditions and resistance to pest damage and other tissue invasions that would render soft endosperm less fit. In fact the tendency toward harder endosperm as a result of hybridization has been established. Perhaps selection for increasingly soft wheats over years of breeding for cake and cookie quality has resulted in the development of lines with reduced combining abilities for additive and non— additive gene combinations. In other words, breeding for high PSI wheats has meant selecting lines with poor genetic nicking for the character endosperm hardness. If the cement hypothesis is correct, then perhaps the amount of this hypothetical substance has in fact been selected against since the time that breeding for soft wheat quality has been emphasized. Consistent with diallel theory, hybrid combination between diverse parents resulted in hybr incr CIOS by 9 tool Hous comk 73 hybrid vigor for hardness. Perhaps it represents an increase in the amount of this alleged cement in those crosses exhibiting the highest sca effects. Efforts should be made to analyze the material according to Simmonds (77), by which a thorough chemical analysis of the material which took several years to develop would be made. Again, according to the mean performance, the cultivar Houser at 40.95 remains the softest wheat with all hybrid combinations resulting in harder wheats. Table Parent 1 SC I MPSC 2 SC I MPSC 3 SC MPSC MPSC 5 SC 74 Table 8. Heterosis for the endosperm trait Particle Size Index (P81) in the F2 generation when measured by the sum of the single cross hybrid (SC F2) - midparental differences (MPSC) for each parent. Parents 1 2 3 4 5 2(SC FZ—MPSC) 1 SC F2 35.18 36.16 32.32 32.12 32.85 -3.29 MPSC 35.72 38.08 35.58 37.25 2 SC F2 36.26 34.10 35.05 35.04 —l.98 MPSC 38.60 36.12 37.79 3 SC F2 40.95 40.25 40.51 —2.02 MPSC 38.47 40.14 35.99 37.59 —0.70 37.66 39.33 -l.72 F Endc Si was exe divers: accord; parent among signif and r variar endosg additi relatj an i1 Posit at 0. extre Effec in Ta Chara Three at 0 Arthi effe 75 F2 Endosperm Separation Index (ESI) Analysis Since it was decided to run ESI after the experiment was executed, the parents were not selected based on their diversity as were those of the other two traits. However, according to the Anova in Table 2, the differences between parents was highly significant. Table 3 shows that the among crosses combining ability analysis revealed significant gca effects, but nonsignificant sca, maternal, and reciprocal effects. Gca accounted for 78% of the variance. This suggests that bran separation from endosperm is largely controlled by additive and additive by additive epistatic variability. Hence, finding the relative gca's of the lines would be useful in breeding for an increase or decrease in this character. The only positive gca values as seen in Table 4 are those for Arthur at 0.83 and Tecumseh at 0.39. Because parents were not chosen to represent the extremes of expression for this character, the heterotic effect based upon hybrid mean performance was small as seen in Table 9. However, positive heterosis is seen for this character in that higher values in fact reflect higher ESI. Three cultivars exhibit this positive effect, with Houser at 0.17 exceeding the others. This was followed by both Arthur and Augusta at 0.10 which exhibited less positive effects. Arthur also showed the highest parent mean perforI and th higher perfor also 0 develc upon : its ne effect 8C8. e: by u: Highs addii used due as . Pare exhi gca 76 performance at 76.58 when compared with the other parents and the hybrids. Obviously, breeders wishing to develop higher ESI varieties would select Arthur based on its mean performance, on its relatively high heterotic effect, and also on its highest gca value of 0.83. Breeders wishing to develop low ESI varieties would select Hillsdale, based upon its relatively low mean performance value of 74.05, its negative heterotic value of -0.08, and its negative gca effect of —O.67. Since this trait did not show significant sca effects, the individual crosses will not be considered. The F5 PSA correlated significantly with PSI. However, the association between ESI and both PSA and PSI was not significant. This reinforces the previous observations that bran separation is a genetically separate character from endosperm hardness. Its different genetic effects also reinforces this hypothesis. The results suggest that this trait could be improved by using varieties such as Arthur that are high in gca. Higher ESI lines could be developed on the basis of additive genetic variability, where high ESI cultivars are used iji a recurrent selection crossing system. However, due to the absence of non~additive components of variation as evideficed by non-significant sca, heterosis between parents of diverse ancestry would not be expected to exhibit superior combinations other than those based upon gca alone. Table ‘. Parent 1 SC F MPSC 2 SC 1 MPSC 3 SC MPSC 4 SC MPSC 77 Table 9. Heterosis for the bran trait Endosperm Separation Index (ESI) in the F2 generation when measured by the sum of the single cross hybrid (SC Fr) - midparental differences (MPSC) for each parent. Parents 1 2 3 4 5 2(SC Fz—MPSC) 1 SC F2 76.58 75.94 75.69 75.68 75.75 +.10 MPSC 76.47 75.33 75.31 75.53 2 SC F2 76.36 75.31 74.93 75.57 -.14 MPSC 75.22 75.20 75.42 3 SC F2 74.08 73.95 74.63 +.17 MPSC 74.06 74.28 4 SC F2 74.05 73.96 -.08 MPSC 74.27 5 SC F 74.49 +.10 (I) CONCLUSIONS Particle Size Analysis (PSA) as measured by NIR or the Microtrac Particle Size Analyzer is a precise determination of wheat endosperm particle size showing advantages over Particle Size Index (PSI) measurements. PSA exhibits less environmental interaction, shows a greater range of gca, and only requires small sample amounts. These findings support the adoption of PSA for kernel hardness determinations. The three characters Particle Size Analysis (PSA), Particle Size Index (PSI), and Endospernl Separation Index (ESI) exhibit considerable genetic diversity as revealed by the Anova. Particle Size Analysis (PSA) and Particle Size Index (PSI) show combining ability gca:sca ratios of 1:1.2 and 1:1 respectively, indicating that both additive and non—additive genetic variability are important to the inheritance of these traits in roughly equal proportion. 78 4. 79 A triploid genetic model shows no advantages over a diploid model for combining ability analysis in the triploid endosperm trait, hardness. The F1 maternal effect for Particle Size Analysis (PSA) dominates other combining ability effects, but this disappears after one round of segregation. Arthur/Tecumseh and Augusta/Hillsdale are the choice hybrids to generate in a hybrid wheat program based upon their superior softness in hybrid combination. Yield testing should be considered next for these hybrids to determine their usefulness in a hybrid wheat program. The F1 Particle Size Analysis (PSA) maternal and reciprocal effects disappear in the F2 generation, indicating that this trait is controlled by the triploid endosperm tissue genotype. Selection for lines in which the genes are to be fixed should begin in the F2 generation. However, subsequent inbreeding of lines would fix only the additive and epistatic gene interactions which behave additively. Hence, the gca of a line would be more important than the sca in generations after the F 1. 80 Because of the heterotic tendency to nick for increased hardness, no hybrid combinations were softer than the parent cultivar, Houser, for the trait Particle Size Analysis (PSA). Hillsdale/Houser was second in softness in this generation. Parents with low gca and hybrid populations exhibiting low sca or low average heterosis should be selected when developing wheats for softer PSA. Particle Size Index (PSI) also exhibited the tendency toward harder wheats when hybridized. The softest wheat again was the parent cultivar, Houser, with the crosses Augusta/Houser and Hillsdale/Houser following closely. Parents with high gca and hybrid populations exhibiting high sca or high average heterosis should be selected when developing wheats for softer PSI. It is possible that endosperm hardness is a fitness character, subject to hybrid vigor. Endosperm softness has resulted from years of selecting out those gene combinations that nick for greater hardness because the emphasis was upon increasing cake and cookie quality. That is, it is possible that higher quality was developed at the expense of evolutionarily fit endosperm. 10. 10. 11. 81 Further chemical analyses should be made to verify the cement hypothesis of endosperm hardness. It is indicated that Endosperm Separation Index (ESI) is controlled by additive and additive epistatic genetic variability and hence fixation of these effects could be expected to be fixed into pure lines. Lack of correlation with both PSA and PSI in addition to different patterns of genetic variability as evidenced by the absence of non—additivity point to the separateness of this trait as a genetic character. Therefore, one could increase or decrease ESI while keeping PSI or PSA constant. For an increase in ESI, Arthur would be a good choice based upon its relatively high heterotic effect, and high gca value. Hillsdale would be used to decrease ESI, based upon its relatively low heterotic effect and its low gca value. Finally, finding the gca's of lines would be useful when trying to increase or decrease this character. In a hybrid wheat program, a high x high ESI combination or low' x low combination should. be selected due to the additive gene effects and lack of sca effects. APPENDIX Table 2 Male Female M0201 M0280 M0284 M029 M030 82 Table 10. Parent and hybrid mean values where n = 3 for the endosperm trait Particle Size Analysis (PSA) measured in the F2 generation over 2 replications in 2 environments. 925518 env/rep M0201 M0280 M0284 M0295 M0300 I M0201 1/1 64.47 65.81 70.06 71.35 65.71 1/2 58.61 59.09 65.84 70.49 63.81 2/1 67.88 64.65 68.52 69.88 72.88 2/2 61.73 61.65 68.62 72.46 65.94 M0280 1/1 63.39 62.52 64.38 64.33 66.01 1/2 59.46 59.70 68.45 70.22 67.83 2/1 71.97 64.30 70.91 63.87 68.41 2/2 68.03 64.47 70.41 72.68 68.19 M0284 1/1 67.10 61.85 56.44 60.40 61.36 1/2. 66.58 64.54 55.12 59.38 59.84 2/1 74.93 63.65 57.57 55.58 59.82 2/2 71.58 65.58 57.49 61.33 64.30 M0295 1/1 67.68 64.63 56.23 61.89 62.25 1/2 66.95 68.55 62.83 61.97 66.38 2/1 68.79 67.30 55.42 62.64 58.98 2/2 71.63 66.81 59.76 66.80 61.41 M0300 1/1 66.36 66.87 60.04 61.33 60.89 1/2 66.98 58.89 58.72 64.30 60.53 2/1 72.27 70.07 60.62 60.48 62.19 2/2 66.46 71.40 60.34 64.04 59.80 Table Male Femal M0201 M028! M028 M025 M03 83 Table 11. Parent and hybrid mean values where n = 4 for the endosperm trait Particle Size Index (PSI) measured in the F generation over 2 2 replications in 2 environments. Male Female env/rep M0201 M0280 M0284 M0295 M0300 M0201 l/l 34.30 36.70 31.26 31.15 31.11 1/2 37.24 34.79 33.36 33.16 32.67 2/1 34.58 38.05 33.14 32.68 30.26 2/2 34.63 36.76 33.12 30.48 32.45 M0280 l/l 37.23 36.68 33.97 34.11 34.86 1/2 35.59 35.39 34.10 32.98 34.05 2/1 36.44 36.84 33.76 34.81 38.85 2/2 33.79 36.14 32.76 34.57 32.45 M0284 l/l 33.47 34.77 40.06 39.03 39.36 1/2 33.37 34.05 37.48 36.97 42.54 2/1 29.47 35.81 41.86 40.28 42.94 2/2 31.41 33.64 44.41 40.26 42.48 M0295 1/1 32.20 36.11 39.69 36.94 36.54 1/2 31.65 34.49 37.39 34.56 36.33 2/1 34.50 36.47 47.87 36.35 40.12 2/2 31.06 36.92 40.55 36.14 37.82 M0300 l/l 34.06 34.61 38.83 36.16 36.81 1/2 32.87 36.23 37.72 34.09 36.41 2/1 35.22 35.91 40.41 41.39 40.83 2/2 34.23 33.55 39.85 38.35 43.27 84 Table 12. Parent and hybrid mean values where n = 4 for the bran trait Endosperm Separation Index (ESI) measured in the F2 generation over 2 replications in 2 environments. Male FEEETE env/rep M0201 M0280 M0284 M0295 M0300 M0201 1/1 77.39 75.30 76.39 76.02 77.84 1/2 77.58 77.94 77.13 76.44 77.77 2/1 74.53 74.87 75.63 74.69 74.48 2/2 76.83 74.91 74.98 73.91 74.46 M0280 1/1 77.56 77.46 77.30 77.02 77.53 1/2 77.90 77.75 76.50 76.36 77.44 2/1 75.18 75.33 75.42 75.32 73.33 2/2 73.88 74.90 73.53 73.51 73.86 M0284 1/1 76.83 76.70 76.58 76.49 77.27 1/2 76.71 76.87 76.75 76.33 75.89 2/1 74.72 72.86 71.16 72.25 72.54 2/2 73.14 73.37 71.86 71.98 72.83 M0295 1/1 77.76 76.21 75.55 75.43 75.30 1/2 77.13 76.24 75.34 76.00 74.82 2/1 75.37 72.50 72.24 72.06 71.86 2/2 74.18 72.36 71.46 72.74 73.33 M0300 1/1 76.91 77.23 76.68 76.64 75.44 1/2 77.06 77.47 77.43 76.41 77.20 2/1 74.02 74.22 73.38 72.30 74.29 2/2 73.55 73.51 71.05 71.11 71.04 85 Table 13. Hybrid mean values where n = 4 for the endosperm trait Particle Size Analysis (PSA) measured in the F1 generation as a randomized complete block design over 1 replication. Male FEEETE M0201 M0280 M0284 M0295 M0300 TYi. M0201 77.79 84.10 75.22 75.11 312.22 M0280 61.83 71.65 74.05 73.29 280.82 M0284 72.46 73.86 77.79 73.18 279.29 M0295 75.95 78.33 75.71 69.61 299.60 M0300 72.63 72.40 71.67 73.07 289.77 TY.j 282.87 302.38 303.13 300.13 291.19 1479.70 Yi.+Y.j 595.09 583.20 600.42 599.73 580.96 86 Table 14. Parental and hybrid mean values pooled over 2 replications in 2 environments, where n = 12, for the endosperm trait Particle Size Analysis (PSA) in the F2 generation. Male . FEEETE M0201 M0280 M0284 M0295 M0300 TYl. M0201 63.17 62.79 68.26 71.04 67.08 269.17 M0280 65.71 62.74 68.53 67.77 67.61 269.62 M0284 70.04 63.90 56.65 59.17 61.33 254.44 M0295 68.76 66.82 58.56 63.32 62.25 256.39 M0300 68.01 66.80 59.93 62.53 60.84 257.27 TY.j 272.52 260.31 255.28 260.51 258.27 1306.89 Yi.+Y.j 541.69 529.93 509.72 516.90 515.54 87 Table 15. Parent and hybrid mean values pooled over 2 replications in 2 environments, where n = 16, for the endosperm trait Particle Size Index (PSI) in the F2 generation. M—a—le— M0201 M0280 M0284 M0295 M0300 TY' Female ‘ 1' M0201 35.18 36.57 32.72 31.89 31.62 132.80 M0280 35.76 36.26 33.64 34.11 35.02 138.53 M0284 31.93 34.56 40.95 39.13 41.83 147.45 M0295 32.35 35.99 41.37 35.99 37.70 147.41 M0300 34.09 35.07 39.20 37.49 39.33 145.85 TY.j 134.13 142.19 146.93 142.62 146.17 712.04 Yi.+Y.j 266.93 280.72 294.38 290.03 292.02 88 Table 16. Parent and hybrid mean values pooled over 2 replications in 2 environments, where n = 16, for the bran trait Endosperm Separation Index (ESI) in the F2 generation. & 110201 M0280 M0284 M0295 M0300 TYi. Female M0201 76.58 75.75 76.03 75.26 76.13 303.17 M0280 76.13 76.36 75.68 75.55 75.54 302.90 M0284 75.35 74.95 74.08 74.26 74.63 299.19 M0295 76.11 74.32 73.64 74.05 73.82 297.89 M0300 73.38 75.60 74.63 74.11 74.49 299.72 TY.j 302.97 300.62 299.98 299.18 300.12 1502.87 Yi.+Y.i 604.14 603.52 599.17 597.07 599.84 BIBLIOGRAPHY 10. BIBLIOGRAPHY Aamodt,(L S. and J.EL Torrie. 1935. Studies on the inheritance of and the relation between kernel texture and protein content in several spring wheat crosses. Can..L Res. Sect.(L 13:202. Adams,(L A” Norveillie,L.,and N.Liefenberg. 1976. Biochemical properties and ultrastructure of protein bodies isolated from selected cereals. Cereal Chem. 53:1-12. Altman, D.DL, McCuiston, W.I“ and w.EL Kronstad. 1983. Grain protein percentage, kernel hardness, and grain yield of winter wheat in foliar applied urea. Agron. J. 75:87-91. Baenziger, P.S.,Clements, R.I“, McIntosh, M.S” Yamazaki, w. T. Starling, 12 J., Sammons, D. J., and J. W. Johnson. 1984. Effect of cultivar, environment, and their interaction and stability analyses on milling and baking quality of soft red winter wheat. Crop Sci. 25:5—8. Baily, T. B” Qualset, C.CL, and D.E3 Cox. 1980. Predicting heterosis in wheat. Crop Sci. 20:339- 342. Baker, R. J. 1978. Issues in diallel analysis. Crop Sci. 18:533—536. Barlow, K.IL, Buttrose, M.Su Simmonds, D.IL, and M. Vesk. 1973a. The nature of the starch—interface in wheat endosperm. Cereal Chem. 50:443—454. Barlow, K. K., Buttrose, M.S., Simmonds, D. H., and V. Vesk. 1973b. The nature of the starch—protein interface in wheat endosperm. Cereal Chem. 50:443—454. Bechtel, D. Bu Gaines, R.Iu, and Y.P0meranz. 1980. Changes in protein bodies of maturing wheat. (Abstr.) Cereal Foods World 25:516—517. Biffin, R. H. 1908. On the inheritance of strength in wheat. J. Agric. Sci. 3:86. 89 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 90 Bingham, J. 1966. Paternal effect on grain size in wheat. Nature 209:940-941. Bitzer, M. J.and S.EL Pu. 1972. Heterosis and combining ability in sourthern soft red winter wheats. Crop Sci. 12:35—37. Blum, P. H., Panos, G., and Smith, R. J. 1960. Measurement of hardness of barley and malt with the Brabender Hardness Tester.II. Modification during malting. Am. Soc. Brew. Chem. Proc. 18:95. Briarty, L. G. 1978. The mechanisms of protein body deposition in legumes and cereals. Ln: Plant Proteins. G. Norton, ed. Butterworths: London, Boston. pp. 81—106. Briarty, L. G” Hughes, C. E” and A.IL Evers. 1979. The developing endosperm of wheat—-A stereological analysis. Ann. Bot. 44:641-658. Briggle, L. w. 1963. Heterosis in wheat—-A review. Crop Sci. 3:407-412. Brown, C. M., Weibel, R. O., and R. D. Seif. 1966. Heterosis and combining ability in common winter wheat. Crop Sci. 6:382—383. Buttrose. M. S. 1960. Submicroscopic development and structure of starch granules in cereal endosperm. J. Ultrastruct. Res. 4:231-257. Chaudhary, V. K” Yamazaki, W.'Tq and W.IL Gould. 1981. Relation of cultivar and flour particle size distribution to cake volume. Cereal Chem. 58(4):314—317. Cockerham,(L C.andEL S.Weir. 1977. Quadratic Analyses of Reciprocal Crosses. Biometrics 33(3):187—203. Cockerham, C. C. 1980. Random and fixed effects in plant genetics. Theor. Appl. Genet. 56:119—131. Dickinson, A.(L, and J.I“ Jinks. 1956. A gen— eralized analysis of diallel crosses. Genetics 41:65—78. Donelson, D.}L and W.TL Yamazaki. 1972. Soft wheat flour particle—size analysis by integrated sieve and coulter counter procedures. Cereal Chem. 49(6):641—648. 91 24. Evers, A. D. 1969. Scanning electron microscope observations on the gross structure of wheat starch granules. Chem. Ind. Berlin. 1969:200- 201. 25. Evers. A. D. 1970. Development of the endosperm of wheat. Ann. Bot. 34:547-555. 26. Evers, A. D. 1971. Scanning electron microscopy of wheat starch. III. Granule development in the endosperm. Staerke 23:157—162. 27. Evers, A. D. 1973. The size distribution among starch granules in wheat endosperm. Staerke 25:303-304. 28. Evers, A. D. and J. Lindley. 1977. The particle size distribution in wheat endosperm starch. J. Sci. Food agric. 28:98—102. 29. Falconer, D. S. 1981. I d ' n n ' ' Genetics. 2nd Ed. Longman Grp. Lmt., p. 251. 30. Farrand, E.Am and J.LL Hinton. 1974. Study of relationships between wheat protein contents of two U.K. varieties and derived flour protein contents at varying extraction rates. II. Studies by hand dissection of individual grains. Cereal Chem. 51:66. 31. Favret,l& A” Manghers, Li,Solari, R” Avila,AJ,and Monesiglio, J. C. 1970. Gene control of protein production in cereal seeds. In: Imprgying_glagt Preteiné by Nuclear Techniques. IAEA Proc. Series STI/PUB/258, Vienna. pp. 87-95. 32. Gilbert, N. E. G. 1958. Diallel crosses in plant breeding. Heredity 12:477—492. 33. Griffing, B. 1956a. A generalized treatment of the use of diallel crosses in quantitative inheritance. Heredity 10:31-50. 34. Griffing, B. 1956b. Concept of general and specific combining ability in relation to diallel crossing systems. Austr. J. Biol. Sci. 9:463-493. 35. Grosh, G. U. 1959. Water penetration and internal cracking in tempered wheat grains. Cereal Chem. 36:260. 36. Gyawali, K.IL, Qualset, C.CL, and W.fih Yamazaki. 1968. Estimates of heterosis and combining ability in winter wheat. Crop Sci. 8:322-324. IIIIIIIIIIIIIII[:::T________________——fi 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 92 'c ‘ M ' B d' . Iowa State U. Press. Chapter 4, Hereditary variance: mating designs. Harrington, J. B. 1940. Yielding capacity of wheat crosses as indicated by bulk hybrid tests. Canad. J. Res. 18:578—584. Hayman, B. I. 1954. The theory and analysis of diallel crosses. Genetics 39:789—809. Hinton, J..J.C. 1955. Resistance of the testa to entry of water into the wheat kernel. Cereal Chem. 32:296. Hughes, C. E.and L.(L Briarty. 1976. A quantitative analysis of some ultrastructural aspects of seed development. Natl. Bar. Stand. 0.8. Special Publ. 431. Proc. 4th Int. Congr. Stereol. Natl. Bur. Stand. Gaithersburg, MD. pp. 385-388. Jennings, A.(L, Morton, R.K. and B.Au Palk. 1963. Cytological studies of protein bodies of developing wheat endosperm. Aust. J. Biol. Sci. 16:366—374. Jones, R.W.and R.J.Dim1er. 1962. Electrophoretic composition of glutens from air—classified flours. Cereal Chem. 39—336. Kasarda, D.IL, Bernardin, J. E” and Nimmo, C. C. 1976. Wheat proteins. In: Advances in Cereal Science and Technology. V01. 1. Y. Pomeranz, ed. Am.Assoc.Cereal Chem” St.Paul, MN. pp.158- 236. Katz, R., Collins, N. D., and A. B. Cardwell. 1961. Hardness and moisture content of wheat kernels. Cereal Chem. 38:364—368. Kempthorne, O. 1956. The theory of the diallel cross. Genetics 41:451—459. Konzek, C. F. 1977. Genetic control of the content, amino acid composition, and processing properties of proteins in wheat. In: Adyances in Genetics, Vol. 19. E.7L Caspari, ed. Academic Press, Inc., New York. pp. 407-482. Kosmolak, F. G. 1978. Grinding time-—screening test for kernel hardness in wheat. Can. J. Plant Sci. 58:415. —:— 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 93 Kronstad, W.EL and W.EL Foote. 1964. General and specific combining ability estimates in winter wheat. Crop Sci. 4:616-619. Law, C. N., Young, C. F., Brown, J. W. S., Shape, J. W3, and A.;L Worland. 1978. The study of grain protein control in wheat using whole chromosome substitution lines. IQ: Seed Protein Improye- men; by Nuclear Techniques. IAEA Proc. Series STI/PUB/479. Vienna. pp. 483-502. Lineback, D.IL, Cashman, W. E., Hoseney, R. C., and A. B. Ward. 1978. Note on measuring thickness of wheat bran by scanning electron microscopy. Cereal Chem. 55(3):4lS—419. Lupton, F. G. 1961. Studies in the breeding of self— pollinated cereals. 3. Further studies in cross predictions. Euphytica 10:209—224. MacRitchie, F. 1980. Physiochemical aspects of some problems in wheat research. From: Ad c ' C Sc' nc T c n 93y, Vol. III. AACC St. Paul, Minn. Y. Pomeranz ed. 271-325. Mayo, 0. 1983. Problems of significance tests in diallel analysis. Sabrao Journal 15(2):147—149. Mecham, D. K” Kasarda, D.IL, and C.CL Qualset. 1978. Genetic aspects of wheat gliadin proteins. Biochem. Genet. 16:831—853. Mirjana, I., Vulic, B., and Virginia Djolai. 1979. Expression of heterosis and combining ability for grain protein in a diallel wheat cross. Wheat Info. Serv. 49:1—4. Miller, B. S” Trimbo, H. B” and K.IL Powell. 1967. Effects of flour granulation and starch damage on the cake—making quality of soft wheat flour. Cereal Sci. Today 12:245. Miller, B. S" Hughes, J.VL, Afework, S” and Y. Pomeranz. 1981a. A method to determine hardness and work of grinding wheat. J.Food Sci. 46(1):851. Miller, B.S.,Afework,S.,Pomeranz,Y.,and L.Bolte. 1981b. Wheat hardness: time required to grind wheat with the Brabender automatic microhardness tester. J. Food Sci. 46(1):863. f 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 94 Miller, B. S., Pomeranz, Y. and S. Afework. 1984. Hardness (texture) of hard red winter wheat grown in a soft wheat area and of soft red winter wheat grown in a hard wheat area. Cereal Chem. 61(2):201—203. :flillet, E. and M. J. Pinthus. 1980. Genotypic effects of the maternal tissues of wheat on its grain weight. Theor. Appl. Genet. 58:247-252. Millington, A. J. and E. Remilton. 1954. The flour strength of Australian wheat varieties. J. Aust. Inst. Agric. Sci. 20:24-35. Milner, M. and J. A. Shellenberger. 1953. Physical properties of weathered wheat in relation to internal fissuring detected radiographically. Cereal Chem. 30:202-212. M011, R.EL Lonnquist, J.}L, Fortuno, J.\h and E.CL Johnson. 1965. The relationship of heterosis and genetic divergence in maize. Genetics 52:139-144. Moss, H. J. 1978. Factors determining the optimum hardness of‘wheat. Aust.;L Agric.Res.29:lll7- 1126. Nakagawa, N” Soga, Yuywatanabe, S” Gocho,fL, Nishio, K., and Kurata, H. 1959. Genetical studies on kernel texture of wheat and barley. Bull. First Agron. Div., Tokai—Kinki Nat. Agric. Exp. Sta. N0. 6. Newton, R., Cook, W. H., and Malloch, J. G. 1927. The hardness of the wheat kernel in relation to protein content. Sci. Agric. 8:205. Obuchowski, w., and Bushuk, W. 1980. Wheat hardness: effects of debranning and protein content. Cereal Chem. 57:426. Orth, R. A. 1977. Determination of kernel hardness of Australian wheats by a rapid grinding procedure. Aust. J. Exp. Agric. Anim. Husb. 17:462. Parish,¢L A.,ahd N.£L Halse. .1968. Effects of light, temperature, and rate of desiccation on translucency in wheat grain. Aust. J. Agric. Res. 19:365. Pearson, D. C., Rosielle, A. A., and W. J. R. Boyd. 1981. Heritabilities of five wheat quality traits for early generation selection. Aust. J. Exp. Agric. Anim. Husb. 21:512-515. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 95 Percival, J. 1921. The Wheat Plant 1974. A mono— graph. Duckworth, London. Purdy, L. H., W. Q. Loegering, C. F. Konzak, C. J. Peterson, and R.E. Allan, 1968. A proposed standard method for illustrating pedigrees in small grain varieties. Crop Sci. 8:405—406. Qualset, C.O.,and(L W.Wrigley. 1977. Chromosomal control of gliadin proteins: The practical impli- cations. (AbstrJ. Cereal Foods World 22:475. Sampson, D. R. 1982. Lack of correlation between kernel hardness and protein content in a winter wheat progeny. Can. J. Plant Sci. 62:797—799. Simmonds, D. H. 1972. The ultrastructure of the mature wheat endsperm. Cereal Chem. 49:212—222. Simmonds, D.1L, Barlow, K.K.,and C.VL Wrigley. 1973. The biochemical basis of grain hardness in wheat. Cereal Chem. 50:553-562. Simmonds, D. H. 1974. Chemical basis of hardness and vitreosity in the wheat kernel. Bakers Dig. 10:16—29. Simmonds,IL H.and T.P.(TBrien. 1981. Morphologi- cal and biochemical development of the wheat endo- sperm. From Adyances in Cereal Science Yend Tech— anqu. Vol. IV. AACC St. Paul, MN. Y. Pomeranz, Ed., pp. 5— 55. Sokol, M. J.and R..J.Baker. 1977. Evaluation of the assumptions required for the genetic interpreta- tion of diallel experiments in self—pollinated crops. Canada J. Plant Sci. 57:1185—1191. Stenvert, N. L. 1974. Grinding resistance. A simple measure of wheat hardness. Flour Anim. Feed Milling. 7:24. Stenvert, N. L. and Kingswood, K. 1977. The influence of the physical structure of the protein matrix on wheat hardness. J. Sci. Food Agric. 28:11—19. Symes, K. J. 1961. Classification of Australian wheat varieties based on the granularity of their whole meal. Aust. J. Exp. Agric. Anim. Husb. 1:18. Symes, K. J. 1965. The inheritance of grain hardness in wheat as measured by the particle size index. Aust. J. Agric. Res. 16:113. —i——’ 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 96 Symes, K. J. 1969. Influence of a gene causing hardness on the milling and baking quality of two wheats. Aust. J. Agr. Res. 20:971. Tandon, J. P., Joshi, A. 8., and K. B. L. Jain. 1970. Comparison of graphic and combining ability analyses of diallel crosses in wheat. Indian J. of Genetics and Plant Breeding. 30(1):91~103. Ternovskaya, T. K. and E. G. Zhirov. 1979. Herita— bility of chromosomes of the D genome in the progeny of pentaploid wheats. II. Analysis of endosperm density in seeds with different chromo- some numbers. Genetika, USSR. 15(4):701-707. Thompson, J.EL and R. N.PL Whitehouse. 1962. Studies on the breeding of self—pollinating cereals. 4. Environment and the inheritance of quality in spring wheats. Euphytica 11:181-196. Trupp, C. R. 1976. Particle size index - breeding behavior and association with protein percentage in wheat. Crop Sci. 16:618. Verma, P.1L and O.Ih Luthra. 1983. Heterosis and combining ability analysis for yield and its attributes in macaroni wheat. Indian J. Agric. Sci. 53(6):385-389. Weiss, E.I“ and H.1L Frock. 1976u Rapid analysis of particle size distribution by laser light scattering. Powder Technology 14:287—293. Wertheimer,ZL.L.,and W.Iu Wilcock. 1976. Light scattering measurements of particle distributions. Applied Optics 15(6):l616—l620. Williams, P. C. 1979. Screening wheat for protein and hardness by near infrared reflectance spectro— scopy. Cereal Chem. 7:24. ‘Williams, P. C. 1983. Physical methods for grain testing. Cereal Foods World 28(9):593. Worzella, w. W. 1942. Inheritance and interrelation- ship of components of quality, cold resistance, and morphological characters in wheat hybrids. J. Agric. Res. 65:501. Wrigley, C. W. 1972. The biochemistry of the wheat protein complex and its genetic control. Cereal Sci. Today 17(12):370—375. 97. 98. 99. 100. 101. 102. 97 Yamada,lh U.and K.Olden. 1978. Fibronectins-— adhesive glycoproteins of cell surface and blood. Nature 275:179-184. Yamazaki, W. T. 1972a. A modified particle-size index test for kernel texture in soft wheat. Crop Sci. 12:116. Yamazaki, W. T. and D. H. Donelson. 1972b. The relationship between flour particle size and cake- volume potential among eastern soft wheats. Cereal Chem. 49(6):649—653. Yamazaki, W.TL andIg C.Andrews. 1982. Small.scale milling to estimate the milling quality of soft wheat cultivars and breeding lines. Cereal Chem. 59(4):270—272. Yamazaki,7L T.and J.IL Donelson. 1983. Kernel hardness of some 0.8. wheats. Cereal Chem. 60(5):344—350. Yates, F. 1947. Analysis of data from all possible reciprocal crosses between a set of parental lines. Heredity 1:287-301.