THE INHERITANCE 0F SALT TOLERANCE lN BARLEY (HORDEUM VULGARE L. ) Dissertation for the Degree of Ph. D. MICHIGAN SIAIE UNIVERSITY ABUBAKER M. MADDUR 1976 b wrra MIGRIGAN S‘I’ATE UNNERSITY LIBRARIES E *4»; r. LIBRARY 5 Michigan Scans Univctity . ABSTRACT THE INHERITANCE OF SALT TOLERANCE IN BARLEY (HORDEUM VULGARE L.) by Abubaker M. Maddur The differential responses of barley varieties (Hordeum vulgare L.) to salt stress in various growth stages suggest that one of the most promising methods to overcome the problem of salt injury to plants is the use of tolerant varieties through a viable breeding program perhaps in the combination with land reclamation and de- salination of salty water. The first step in this direction requires the de- ve10pment of a method to classify large numbers of indi- vidual plant genotypes for salt tolerance in various growth stages along with a knowledge of the inheritance of salt tolerance. A method was developed to screen barley popula- tions for salt tolerance in the germination stage. Abubaker M. Maddur Thirty-three varieties were included in the screening test. On the basis of the screening results, six parental varie- ties were chosen for a diallel cross set in order to inves- tigate the genetic basis of salt tolerance. The F2 from the 6 x 6 diallel-cross were tested for salt tolerance in two growth stages; the germination stage and the early growth stage, following germination and early seedling. The data were analyzed according to the Jinks- Hayman diallel-cross analysis. Salt tolerance in barley appears to be controlled by dominant genes. Dominance seems to be partial in the germination stage and nearly complete in the early growth stage. Non-allelic gene interaction was found to be im- portant in determining salt tolerance in the early growth stage. Transgressive segregation was observed in some crosses suggesting the possibility of identifying superior crosses. THE INHERITANCE OF SALT TOLERANCE IN BARLEY (HORDEUM VULGARE L.) by (*050 Abubaker M? Maddur A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 1976 To my people ....... ACKNOWLEDGEMENTS The author wishes to express deep acknowledgement to Dr. John E. Grafius and Dr. David H. Smith for their encouragement, guidance and advice throughout this study. Acknowledgement is extended to Dr. Carter M. Harrison for reviewing this manuscript. Appreciation is expressed to Dr. William T. Magee, Dr. Charles E. Cress and Dr. Donald Penner for serving on the guidance committee. The author wishes to express further acknowledge- ment to his family in Libya for their moral support during his stay in the U.S.A. Special thanks to his wife Soufia and his son Laith for their companionship, understanding, encouragement and help during the years of graduate study. iii TABLE LIST OF TABLES O O O O O 0 LIST OF FIGURES . . . . . INTRODUCTION 0 O O O O O 0 REVIEW OF LITERATURE. . . MATERIALS AND METHODS . . RESULTS C O O O O O O O 0 DISCUSSION. 0 O O O O O 0 SUMMARY AND CONCLUSIONS . LITERATURE CITED. . . . . APPENDIX. 0 O O O O O I 0 OF CONTENTS iv Page vi ll 26 49 57 59 65 LIST OF TABLES Table Page 1. List of barley varieties screend for salt tOlerance O O O O O O O O O O O O O O O O O O 12 2. Mean squares for salt tolerance score in the germination stage of the F2 of the 6 x 6 diallel cross set. . . . . . . . . . . . . . 32 3. Salt tolerance score (% of the control) in the germination stage of the F of the 6 x 6 diallel cross set. . . .2. . . . . . . 34 4. Mean squares for coleOptile growth score of the F2 of the 6 x 6 diallel cross set. . . . 37 5. Coleoptile growth score of the F2 of the 6 x 6 diallel cross set. . . . . . . . . . . 33 6. Correlation coefficients between the salt tolerance score in the germination stage test of the offspring with coleoptile growth in the control for the non-recurring parent (d.f. 4) for each array . . . . . . . 42 7. Mean squares for plant total dry weight in the early growth stage of the F2 of the 6 x 6 diallel cross set. . . . . . . . . . . 43 8. Total dry weight score (% of the control) in the early growth stage of the F2 of the 6 x 6 diallel cross set. . . . . . . . . . . 44 Figure la. 1b. 1c. 6. 7. LIST OF FIGURES Test tubes containing germinating barley seeds in several salt concentrations. . . A view of the tray used to grow plants in the nutrient solution . . . . . . . . . . A sketch of the plastic tube used to support plants in trays . . . . . . . . . . . . . Effect of increasing NaCl concentration on the coleoptile growth of 11 barley varieties . . . . . . . . . . . . . . . . Effect of increasing NaCl concentration on the coleoptile growth of two salt tolerant and two salt susceptible barley varieties. . . . . . . . . . . . . _A view of nutrient tray in operation containing 12 barley varieties in solution with 0, 3000, 6000 and 9000 ppm NaCl . . . . . . . . . . . . . . Effect of NaCl concentration on the plant height of 12 barley varieties . . . . . . Effect of NaCl concentration on the total dry weight of 12 barley varieties . . . . W graph analysis of salt tolerance scores Iin the germination stage of the F2 of the 6 x 6 diallel cross set . . . . . . . . . vi Page 15 18 19 27 28 29 30 31 35 LIST OF FIGURES (cont'd.) Figure Page wr/Vr graph analysis of coleoptile growth of " .the F2 of the 6 x 6 diallel cross set. . 40 9. W r/Vr graph analysis of the total dry weight rin the early growth stage of the F2 of the 6 x 6 diallel cross set. . . . . . . 45 10. W r/‘Vr graph analysis of Figure 9 with rparent l eliminated. . . . . . . . . . . 47 vii INTRODUCTION Soil salinity is a major factor in determining the capacity of the land for agricultural use in the arid zones of the world where rainfall is scarce and seasonal and the underground waters, when found, are often too saline to be of use. Soil salinity becomes important in agriculture when the concentrations of soluble salts in the soil solu- tion reach levels that adversely affect plant growth and yield. These conditions are by no means restricted to arid climates. For instance, excessive use of fertilizers in intensively farmed lands, or irrigation with saline waters, or with insufficient water or under conditions of poor drainage may also bring about the accumulation of soluble salts to the point where their concentrations seriously impairs productivity. The adverse effects on plant growth of excessive concentration of soluble salts in the root medium are varied. They range from simply inhibiting the growth of some plants and reducing yield, to actually in- juring or killing the plant tissues. 1 The urgent demand to resolve the current world problems of hunger and food shortages, complicated by the rising world population, increases the need to search for means of exploiting the vast areas of saline desert lands and the need for future use of brackish and saline waters for crop irrigation. These challenges can be met by de- veloping salt tolerant plants through a viable breeding program. This should be coupled with plans for the re- clamation of desert lands and the desalination of sea water. The fact that plant species are not equally affected by salinity and the significant variation in salt tolerance encountered in varieties of some species, suggests that plant tolerance to salt stress is under gene control. The relative high salt tolerance of the barley plant, and the striking performance of some varieties, makes barley a prime candidate among all cereals for any breeding program for salt tolerance. However, adequate information on the subject of the inheritance of salt tolerance is a prerequisite to enable a program of this kind to materialize. This study is designed to investigate the genetic basis of salt tolerance in the barley plant (Hordeum vulgare L.). REVIEW OF LITERATURE A "Saline Soil" is one which contains sufficient soluble salts 1x) affect adversely the growth of plants (27). The surface inch of soil is often more saline than the soil below it. This is a result of evaporation and capillary movement of saline water to the surface. Furrow- irrigated ridges or raised beds (10, 44) enhance this con- dition. The cations, calcium, sodium and magnesium, and the anions, chloride, sulfate, bicarbonate and carbonate were generally predominant (12). But most of the salt stresses in nature are due to sodium salts, particularly sodium chloride (32). Symptoms of salt-injured plants are often recognized as stunted growth and smaller darker green leaves_(12); also as necrosis or marginal burn (41), and in severe cases final death of the plant (32). The effects on plant growth of the excessive salt concentrations in the root zone may be mediated by osmotic effects, or by specific ion effects, or both (12, 27, 32). The osmotic theory assumes equivalent effects on the growth processes of the plant by such ions as sodium, calcium and of chloride and sulphate (12). This suggests that salt injury to plants was due primarily to physio— logical scarcity of water, resulting from increased osmotic pressure of the soil solutions (4, 9, 20, 29). The theory's advocates base their ideas on the fact that some symptoms exhibited by salt—injured plants, such as stunted growth and smaller, darker green leaves, closely reSembled those symptoms caused by drought stress (20, 29). The specific ion effect theory, on the other hand, requires the deleterious effects of salts be dependent, necessarily, on the different salts. Salts may exert ad- verse effects interfering with the plant metabolism by in- ducing shifts in mineral nutrient status or by causing direct toxicity (12). The stress actions of similar concentrations of various salts on wheat germination have been reported to decrease, in the order of magnesium, potassium, and sodium (26, 46). Sodium chloride solutions were more depressive to wheat germination than isosmotic dilutions of sea water and of glucose solutions (12). Similarly, sodium chloride exceeded mannitol in decreasing alfalfa germination, when equivalent isosmotic solutions were compared (41, 43). Excessive absorption of magnesium by plants decreased the absorption of calcium and potassium (12), while sulphate ion enhanced the uptake of sodium at the expense of calcium ions (29). Chloride absorption induced nitrate deficiency in wheat (17). There is evidence of direct toxicity to various plant species by such ions as sodium (8), chloride (16), boron (21), bicarbonate (15, 23, 39) and phosphate (13). Some of the reported deleterious effects on plants induced by salt stress include: a decrease in several metabolic processes such as respiration, photosynthesis, protein and nucleic acid synthesis in several plant species (32), a suppression of chloroplast development in bean (36) and a reduction in cytokinins translocation from the roots in tobacco (30). Salinity caused tomatoes (28) and barley (22) to accumulate more carbohydrates. The difficulty in comparing tolerance in the dif- ferent stages of plant development, due to the dissimilarity of criteria employed in the evaluation of growth, prevented a single clear cut universal definition of “salt tolerance." It may be considered as the capacity of the plant to sur- vive under conditions of increasing salinity stress. It might be evaluated either on the basis of the relative performance of the plant at a given level when compared with other plants or the performance of the plant at a certain salinity level relative to its behavior under non- saline conditions (27). Salinity effects on plant development and yield might depend on such factors as: the plant species, the type of cr0p, stage of development and/or other related and interacting factors (12). Cereals (Hordeum, Avena and Triticum) are said to be less sensitive to salinity stress than legumes (Pisum.Phaseolus,but more sensitive than other species such as Medicago, Helianthus and Beta (32). In the cereal group, barley was more tolerant than oats (5) and wheat (2, 5), while corn was the least salt tolerant cereal (26). Wild relatives of tomatoes were more tolerant than the cultivated ones (42). Phosphoenolpyruvate carboxylase enzyme isolated from leaves of C4 plants appeared more sen- sitive to inorganic salts than the same enzyme extracted from C3 plants (37). Plant reaction to salt stress varied depending on the stage of its development (12). There might not be a positive correlation between salt tolerance at germination and during later phases of growth (3, 26). Sugarbeets, for instance, were more sensitive during germination than later growth phases (7, 12). On the other hand, corn (3, 12) and sesame (47) were more tolerant during germination than at later stages (3, 12). The early seedling stage of most grains was more ,affected than either germination or later stages (7, 38). 'However, the four-leaf stage was the most sensitive stage in wheat and barley (2). Selection in wheatgrass for salt tolerance in the germination stage was ineffective in increasing salt toler- ance at subsequent growth stages (19). Furthermore, while salinity decreased markedly the vegetative growth of barley, the grain yield remained essentially unchanged (2), whereas the reverse appeared to be the trend in rice (12). The adaptability mechanism may not be a capacity to function with large quantitites of salt within plant tissues (18), and it may not necessarily be the same in all species. The desert saltbush [Atriplex polycarpa_ (Torr.) S. Wats.] adapted to salinity by localizing the absorbed salt in the trichomes, essentially isolating it from the mesophyll tissues (18). The adaptability mechan- isms were reported to include passive exclusion or active extrusion, and dilution of the entering salt (18, 24, 32). Significant varietal differences in salt tolerance of agronomic value have been observed in crops such as barley (l, 2, 3, 6, 26, 34, 35, 40), wheat (17), wheatgrass (l9), rice (38), sugarcane (11), green beans (8), alfalfa (l4) and tomatoes (33). The amount of variation in barley varieties was the most striking of all. California Mariout, for instance, a barley variety of Egyptian origin is widely known for its excellent salt tolerance during its entire growth period (1, 3, 26). In one study it gave 80% germin- ation at the 0.3 percent salt level (26). Atlas (2, 35) and Regal (34) varieties were also highly salt tolerant. When irrigating with water containing 10,000 ppm salt, Atlas gave 96% grain yield of the control (2). Also, Pal- estinian, Eriterean and Ethiopian barley samples from col- lections at Rastov (USSR) were more resistant to salinity than those of European origin (40). Although the physiology of tolerance of such su- perior varieties is not yet determined, it was reported 10 in one study that salt tolerant varieties of barley, trans— located less sodium and chloride to the shoots than did a salt sensitive variety (24). MATERIALS AND METHODS The lack of a standard procedure to approach the problem of salt tolerance and to measure the various de- grees of tolerance in barley brought about the need to first develOp a practical screening technique. A satis- factory technique should be sensitive to a wide range of salinity levels and capable of measuring and identifying various levels of salt tolerance. I. Preliminary Test Thirty-three varieties of barley, native to several geographic locations and of various growth habitats, were selected for the study. There was no prior information on the salt tolerance of these varieties, except for California Mariout, a variety which was previously known to be salt tolerant. However, the varieties selected were assumed to represent a satisfactory range of germplasm adequate for the purpose of the study. The list and description of varieties are given in Table l. A technique developed by 11 12 TABLE 1.--List of barley varieties screened for salt tolerance. Variety Name Identification Source Abed Mendor Brzz not available Denmark Ackermanns Isaria Nova PI 328618 Germany Akan Mugi CI 11225 Japan Asa C1 11307 Sweden Baladi Cl 11187 Egypt Barbless C1 5105 USA Beecher C1 6566 Egypt Bonus C1 11189 Egypt Bruens Volla C1 11332 Germany Bruens Wisa C1 10089 Germany California Mariout C1 1455 Egypt Carlsberg II Cl 10114 Denmark Coho Cl 13852 USA Conquest C1 11638 USA Dickson C1 10968 USA Domen C1 11417 Britain Freja C1 7130 Britain Giza 117 C1 11190 Egypt Heines Haisa II C1 10113 Germany Ingrid C1 10083 Sweden Lajbjey Drosihezy A not available Denmark Manchuria C1 2947 Manchuria Mashu Mugi Cl 11226 Japan ND B134 not available USA Orge Martin 839 C1 9266 Algeria Orge Saida 183 not available Algeria Pallas C1 11313 Sweden Paragon C1 13649 USA Primus Cl 13109 USA Rika C1 11421 Britain Rokkaku Ozeki C1 11227 Japan Traill C1 9538 USA Wadi Majanen _C1 11211 Libya 13 Whitmore and Sparrow (45), originally designed for labor- atory malting tests, was applied with some modification to fit the purpose of this test. Six levels of salinity were chosen. Solutions of 4, 8, 12, 16, 20 and 24 thousand parts per million (ppm) sodium chloride were prepared by dissolving the equivalent amount of the salt in a proper volume of distilled water. A seventh treatment consisting only of distilled water was included as a check. From the thirty-three varieties, eleven were picked at random for a preliminary test. To identify different varieties a sample of 240-270 kernels, selected for uniformity from each variety, was placed in a petri dish and the dorsal side of the kernels was sprayed with a thin coat of colored enamel, which, when previously tested was found to have no adverse effect on germina- tion. A record was kept for the varieties along with their matching color. For each treatment, 30 seeds from each variety were mixed and placed in a 1007m1 beaker. A volume of 40 cc of treatment solution was poured into the beaker. The seeds were mixed thoroughly with the solution to prevent kernels 14 from floating. The beakers, with their contents, were placed in a growth chamber at 12°C for 48 hours. The treatment solution was changed every 12 hours. At the end of the 48-hour' period the salt solution was filtered off and the wet kernels were gently m0pped with paper towels. The seeds were then transferred to 15.0 x 2.5 cm test tubes. Corks, each with a 0.48 cm hole, were inserted and the test tubes were placed in an upright position in a growth chamber at 17°C for 6 days. On alternate days the germinating seeds were carefully removed from the test tubes, mixed to prevent rootlets from tangling together, and soaked for 3 to 5 minutes in the salt solution and replaced after removing the excess solution. The test was carried out twice. The test tubes with the germinat- ing seeds are shown in Figure 1a. The effect of salinity on seed germination was evaluated on the basis of the coleoptile growth. The coleoptile length relative to the length of the kernel was rated from zero to 8.1 A zero rating designated no coleoptile growth, "1" equals 1/4, "2" equals 1/2, "3" equals 3/4 of the kernel's length and so on up to "8" A 1A standard technique. 15 24- )6 1° l2. Control NaCl concentration in thousand ppm FIG. la.--Test tubes containing germinating barley seeds in several salt concentrations. 16 designating cole0ptile growth as twice the length of the kernel. Coleoptiles were considered as extending from the mid-point of the coleorhiza to their respective tips. II. Screening test (Germination Stage) Based on the results obtained from the preliminary investigation, the 20,000 ppm salt level was selected for the screening test. The varieties were divided at random into three equal groups of 11 varieties each. In each group the varieties were identified as previously described and 30 uniform seeds from each of the 11 varieties were thoroughly mixed and placed in a 100 ml beaker. The kernels were soaked in 40 cc of the 20,000 ppm NaCl salt solution. The rest of the experiment was pursued as described in the preliminary test. This was repeated three times with a separate random regrouping of the 33 varieties prior to each replication. III. Early Growth Stage Test The purpose of this experiment was to test whether salt tolerance in the germination stage was correlated with tolerance in the early growth stage. Based on the 17 results of the screening test (in the germination stage), twelve varieties were chosen for this test.. Of these varieties three were salt tolerant, three salt sensitive and the rest intermediate. Standard plastic trays with air tight sealed lids, about 30.5 x 19.5 x 6.5 cm (Figure lb), were adapted to this test. Seventy-two holes, slightly larger than 0.64 cm in diameter, were drilled in the tray lids. The holes were arranged in six rows of twelve holes each. The distance between holes was kept approximately equal. Trays and lids were sprayed on the outside with several coats of aluminum enamel paint to discourage algal growth. A 10.5 x 0.64 cm transparent plastic tube (straw) was adapted to support the young growing plant. At about 2.5 cm from one of the tube's ends, two holes were punched using a paper punch. A cotton cigarette filter about 2.0 x 0.64 cm was inserted into the tube to a posi— tion, so that it did not entirely cover the”holes: leaving a small opening that was big enough to allow root growth but small enough to keep the kernel from slipping through (Figure 1c). The size of the opening was increased as needed by carefully pushing down the filter to allow for y) ., . .. n _. , .c0flu5H0m ucwfinudc 9.3 c.“ mucmam 3on0 ou omma mangdfi .mo 3wa> amanmn Ha mo nu3onm maaumomaoo may no cowumnunmocoo Humz mcfimmwuocfl mo uomwwmll.~ .me d l ON oc 0m 0m 00. (IOIQUOO 8H3 30 %) qqnorfi atrqdoetoo 28 5mm mason—05 sun GOHUMHUGQUGOO HUMZ e.~ hum“ my. _~_ mu .4 q I s q 1 a q u I I I I I I /. I / I, 1 ON I I I .mowuofiuflr II hoauon oflfi—moumnfi ”Sow II 03» was unoHoHou ”new 0.5 I no gum oawumooaoo on» 0 ¢ II :owuouusoocou Huoz 9:" . Immouocw mo uuowumlld .UHh . nvmu . nvmw 3386 + ooovooéu u a llllll 335 I 3323.0 + avowooévl n n .'.l.‘.l «.me . OC- mmmoooé + mo~moo.00 n a A t anon—unmoun— momnflmn Hmoooo.o + mawmoo.0| u a 1+ Hozooom (toxeuoa an: 30 &) unnoxb atrndoatoo .Homz Ema ooom can ooom .ooom .0 Spas coausfiom ca mofluowum> moauoa NH mcflcfloucoo coflumuomo CH Nona ucflouudc mo 3oH> 0.50). The regression graph of Wr on Vr is shown in Figure 7 along with the limit- 2 ' bl W=VV. ing para o a r p r 34 mmvo.on. oamv.Hm wo.om mm.om mm.0m mm.mm mH.0m ha.mm mmH owfimm omHO w mmMH.HHH mmvo.vm mm.HN mm.NN 00.0N vv.mm mm.Nv Hmfiz Dfimmz m ommw.mm mmmm.©m Hm.mN mm.mm ma.mm hm.Hv onoo w hmmm.vh fihmm.om mm.Hm hv.mv mm.vv UHHmGH m mHNv.mv mem.am 5v.mv m®.Hv 4 wuonflmoum >0nflnmq N mmmo.mm HmNH.MH vm.m¢ USOHHMZ MHGHOMHHMU H HoQEDZ H3 H> o m w m N H mucoumm Hmucoumm N .uom mmOHU HoHHMflG oxo 0:0 mo m onu mo ommum cofluosflsuom osu CH Aaouucoo onu mo xv ouoom oosouoaou uHomII.m mamma 35 l20.0 90.0 80.0 60.0 40.0 20.0 l L l A 20.0 40.0 60.0 80.0 IOOO V' FIG. 7.--Wr/Vr graph analysis of salt tolerance scores in the germination stage of the F2 of the 6 x 6 diallel cross set. ' 36 The graphical analysis shows that Wr and Vr enjoy an almost perfect linear relationship with regression co- efficient b=l.0230*: 0.0795, significantly greater than zero and practically equal to the unit slope. The significance of the regression coefficient plus the uniformity of (Wr - Vr) over arrays satisfy the assumptions underlying the theory of the diallel-cross analysis. The graph shows that the most tolerant variety, parent 1 is at the dominant side of the regression graph, and parent 5, the least tolerant variety at the recessive side of the graph. This means that tolerance to salt was dominant, with parent 1 in this carrying most of the dom— inant genes and parent 5 carrying most of the recessive alleles. From Figure 7, the regression line is shown to in- tercept the Wr axis above the origin (80:17.4388 : 5.5134). Having found that Q0 is significantly greater than zero, it is said that dominance is partial rather than complete.v Since the data from the control were available and could be analyzed in the same way, it was thought useful to obtain information on the genetics of coleoptile growth 37 under conditions free from salt stress. This would enable us to understand the influence of the growth of coleoptile of the seed germinated under salt-free conditions on our measurements of salt tolerance in the germination stage. The analysis of variance for the coleoptile growth of the 6 x 6 diallel set when seeds were germinated in a distilled water (control) is given in Table 4. TABLE 4.—-Mean squares for coleoptile growth score of the F of the 6x6 diallel cross set. 2 Source of Degrees of Mean F Variation Freedom Square Blocks 1 0.2121 4.4103* Entries 20 1.2596 26.1844** Blocks x Entries 20 0.0481 Total 41 *,** significant at 5% and 1% levels respectively. It shows a highly significant difference between entries. Table 5 summarizes the coleoptile growth scores for the six parents and progenies. Each figureis a mean of 60 seeds. The values of the array's variances and covariances were calculated and given on the right hand side of Table 5. A look at the table shows that the score of the cross exceeded 38 owoo.o Ohoq.o mo.m mm.© Hv.m Hm.® 55.0 no.0 mmH mwflmm ovuo m OHNm.o Nth.o mm.¢ Ho.m mm.m mH.m mH.m Hos: dams: m mHmv.o omon.o om.v mw.v mH.w mm.m 0:00 v hmvb.o wmmm.H MH.v Hw.w om.v UHHmsH m mevo.0I memo.o mm.m hH.@ < wuonHmOMQ wonnnoq N mmhm.o mmmb.o mm.m DSOHHMS MHGMOMHHMU H H3 H> m m e m N H mucoumm HMMMMHMQ .uom mmouo HoHHoHp mxw oz» mo mm osu mo ouoom £u3oum oHHumooHooII.m wands 39 in every case the score of the parent with the longest coleoptile. The lowest score was associated with the 3 x 4 cross, and the cross 3 x 6 gave the highest score. Both crosses involve parent 3 which has a relatively low score. Array 2 was the least variable array and array 3, the most variable one. The graphical analysis is shown in Figure 8. The uniformity of (Wr - Vr) over arrays (t=l.5550) and a re- gression coefficient of Wr on Vr (b=0.7452*10.1591) not significantly different from unity suggests that the data satisfy the assumptions of the theory of the diallel an- alysis. The distribution of parents on the diallel graph places parents 3 at the recessive side and parent 2 at the dominant side. From Table 5, parent 2 has a longer coleop- tile than parent 3. This indicated that dominance was in- volved in determining coleoptile growth in barley and was directed towards the longer coleoptile. The regression line, however, intercepted the WI. axis below the origin suggesting over-dominance. When the Wr intercept was. tested (§o=0.1023 :0.0579) it failed to be significantly different from zero. Therefore it was concluded that dominance was complete. 40 .uom mmouo HmHHmne u x o 0:0 no No on» 00 agoum oHHumooHoo mo mHmaHmso nmmnm H>\H3I1.m .0Hm N.Ol . I... N. 0.. mo mo 6.0 «.0 N > d d d 1 d .W u 0.0 . «.0 e n . ¢.0 J. :35 H 32.5 n n o. 1 0.0 . it. e n . 0.0 . 0.. 41 The correlation between the salt tolerance score in the germination stage test for the progeny with mid- parent value for coleoptile growth in the control was r=-0.16 for 13 d.f. Correlation Coefficients were calculated for each array with 4 d.f. between the salt tolerance score in the germination stage test of the offspring with coleoptile growth in the control for the non-recurring parent (Table 6). When they were converted to z values and averaged a non-significant r=-0.12 was obtained. It is also noted that the range in r values although quite large, the chi- square test for the homogeneity of the r values was not significant (x2=l.6484), indicating no difference from an expected random sample of r values. In order to investigate the relationship between salt tolerance in the germination stages and tolerance in the early growth stage, following germination and early seedling data from the early growth stage was subjected to the diallel and graphical analysis. Table 7 gives the analysis of variance for the 6 x 6 diallel set F2 data. It shows a variation of significant value in the entries. The mean values of the plant's total dry weight of the F2 42 TABLE 6.--Correlation coefficients between the salt tolerance score in the germination stage test of the offspring with coleoptile growth in the control for the non-recurring parent (d.f. 4) for each array. ._____-7 Array r 1 —0.61 2 +0.01 3 +0.21 4 -0.22 5 +0.09 6 -0.05 2 Average -0.12 x for z values = 1.6484 ns of the 6 x 6 diallel set are summarized in Table 8. Each figure is an average performance of 18 plants. The array's variances and covariances are given on the right hand side of the table. Table 8 reveals that the performance of the crosses lies within the range of parents with few excep- tions. For instance, the cross 2 x 3 exceeded the best 43 TABLE 7.--Mean squares for plant total dry weight in the early growth stage of the F2 of the 6 x 6 diallel cross set. Source of Degrees of Mean F Variation Freedom Square Blocks . 5 19.3154 4.0611** Entries 20 34.1937 7.1893** Blocks x Entries 100 4.7562 Total 125 **significant at 1% level. parent in the diallel cross set. While the crosses, l x 2, l x 3, and 2 x 5 fell below their corresponding parent with the low score. The test for the consistency of the variable, (Wr - Vr) resulted in a non-significant t value of (t=0.7106). The graphical relationship of Wr and Vr along with the limiting parabola Wi=VbV£ is given in Figure 9. In the graph, points are scattered randomly around the regression line. This was further shown by the non- significance of the regressiOn (b=0.4991 : 0.2906). Having found b not significantly different from zero violated the 44 momm.MH mmm0.aa Hm.Hm om.nm om.~e Hm.me hm.mm es.nm my: meamm mono o oaaw.» mmam.a Ho.oe Hm.He vm.ve vo.em oa.~v ems: game: m ammo.MH mHmN.NH o~.mm oa.me em.ev mm.~q 0:00 a ammo.m Haee.m cm.me em.mv mm.vv caumaH m ommm.~H mono.oa «mate me.~v a sumrsmouo smnhnmq m HmvH.aH mmmH.0H Hm.sv unease: «easemaamo H m thm H0 2 H3 H> 0 m v m m H u m Housoumm .uwm mmDHU HTHHMHU m X m 9.3 MG m m onu mo ommum nuzoum >Humo onu CH AHouucoo onu mo wv ouoom uano3 who Hmuoall.m mqmde 45 28.0“ 24.0“ 20.0’ l6.0' '200- b = 0.4991 I 0.2906 8 .0 - 5. 3e 4.0 V} 4.0 8.0 l2.0 l6.0 20.0 FIG. 9.---Wr/Vr graph analysis of the total dry weight in the early growth stage of the F2 of the 6 x 6 diallel cross set. 46 assumptions underlying the diallel analysis theory. When analysis was done with parent 1 excluded the regression coefficient raised to (b=0.6002 i 0.1433) and became sig- nificantly different from b=0 and from b=l.0 (Figure 10). This new situation with b being between zero and one sug- gested that non-allelic gene interaction played a part in determining the control of salt tolerance at the early growth stage. The Wr interception of the regression line fell above the origin (§O=4.8926 1 2.0178). 'However, this value when tested, was not found to be significantly dif- ferent from zero at the 5% level but greater than zero at the 10% level. The examination of the distribution of arrays on the graph after removing parent 1 reveals that parent 3 and 6 occupy positions near the ends of the regression line. Parent 3 being in the dominant side and parent 6 in the recessive one. Since parent 3 had a higher total dry weight value that parent 6, it can be said that tol- erance to salt at the early growth stage was also dominant, with the degree of dominance ranging from partial dominance to complete dominance. 47 320 ‘I 280 240» 200 I ISO” IZO’ I 8.0 5 «$0 I 1 l A n A v 4.0 so l2.0 l6.0 200 240 ' FIG. lO.--Wr/Vr graph analysis of Figure 9 with parent 1 eliminated. 48 In order to examine whether there was any relation between salt tolerance in the germination stage and salt tolerance in the early growth stage the correlation between the salt tolerance scores in the two growth stages was cal- culated for parents and crosses. The correlation for par- ents was r=0.81 with 4d.f. (P < 0.1) and for crosses was r=0.28 with 13 d.f. (P > 0.2). When the two r values were averaged using the z transformation a non-significant r=0.40 was obtained. The chi-square test for the homogeneity of the r values (x2=1.867) indicated that the two coefficients are not significantly different. DISCUSSION The adverse effect of increasing sodium chloride salt concentration on coleoptile growth in the germinating stage of barley seeds appeared to be linear. NaCl had a similar effect on the height and the accumulation of dry matter in barley plants grown to the four—leaf stage in nutrient media containing various concentrations of this salt. The genetic investigation based on the Jinks-Hayman diallel cross analysis revealed that the assumptions under- lying the theory of the diallel cross analysis were satis- fied by the data of salt tolerance test in the germination stage of the F2 of the 6 x 6 diallel cross. The graphical analysis of the germination stage test (Figure 7), puts the most tolerant variety, parent 1, at the dominant side of the Wr/vr regression graph and parent 5, a salt-sensitive variety at the recessive side. (Parent 1 originated in Egypt while parent 5 is from Japan.) This indicated that barley's tolerance to salt stress in 49 50 the germination stage seems to be determined by dominant genes. The regression line intercepted the Wr axis above the origin, suggesting that dominance is partial. In the salt tests, the coleoptile growth, as a percentage of control was used in all analysis. When examining the growth rate of coleoptile per se of the controls, it too was found to be under genetic control (Table 5 and Figure 8). Dominance was complete and was in the direction of the longer coleoptile. Since the salt tolerance scores of the codeoptile growth were ex- pressed as percentages of the control it may be argued that the higher salt tolerance scores obtained in the offspring of parents 1 and 3, both of which had a low coleoptile growth in the control, were an artifact of the use of percentages. The average correlation between the salt tolerance scores in the germination stage test of the offspring with coleoptile growth in the control for the non-recurring par- ent (r=-0.12) plus the fact that the chi-square test for the homogeneity of the six r values was not significant (X2=l.6484) do not support such an argument. Although it should be pointed out that the correlation in array 1 was 51 relatively large (r=0.61), even though the average corre- lation for all arrays was near zero. The diallel analysis indicated salt tolerance is dominant and further that parent 1 was at the dominant end of the diallel graph. Dominance coupled with a genetic tendency towards a low growth could be expected to produce negative r value when the offspring values within array are compared with the coleoptile growth rates of the non- recurring parents. However, since the coefficient in ques- tion is not significant plus the fact that the population of coefficients is homogenous and tend towards zero leads one to believe that the use of percentage values is valid. There seems no other way to measure salt tolerance (using growth rates) to eliminate or reduce the general effect of genetic differences in growth rate per se between culti- vars. In retrospect it would seem that perhaps one could investigate half-life as an alternative measure, but this would be more difficult to do. The regression analysis (Figure 9), shows that the structure of the F data of the 6px 6 diallel cross of salt 2 tolerance test at the early growth stage, as distinguished from the germination test, did not quite meet the assumptions 52 demanded by the theory of the diallel cross analysis. The higher combining ability of parent 1, indicated by its high Wr value and by its location on the regression graph rela- tive to other parents, suggests that parent 1 played a major part in the data's deviation from the expected lin- earity. When the analysis was repeated, with parent 1 eliminated (Figure 10), the regression coefficient in- creased to a significant level from b=0.499l:0.2906 to b=0.6002i0.1433. The distribution of parents on the regression graph puts parent 3 on the dominant side and parent 6 on the recessive side of the graph. Since parent 3 scored higher than parent 6 in the salt tolerant scale in this growth stage, it was interpreted as barley's tolerance to salt at the early growth stage was also controlled by dominant genes. The regression line of Figure 10 intercepted the Wr axis in a point slightly above the origin suggesting that the degree of dominance was between partial and complete dominance. The regression coefficient b, being significantly different from both zero and one, indicated that non- allelic gene interaction played a role in determining the 53 control of salt tolerance in this growth stage. Parent 3 a semi-tolerant variety, ranked third in salt tolerance among the six diallel parents in both the germination stage and the early growth stage. Parent 6 was the least salt tolerant of the parents in both growth stages. Parent 3 originated in Sweden where salinity stress is probably not as agriculturally important as it is in Algeria where parent 6 originated. The most and the least salt tolerant varieties of the six diallel parents, in both growth stages are common to the North AfriCan region; where screening for salt tolerance over a long period of time is expected. In spite of the fact that parent 3 has shown an intermediate degree of salt tolerance in both tests, the most tolerant hybrid in the germination stage (1 x 3) and the most tolerant hybrid in the early growth stage (2 x 3) both involved parent 3. Although the correlations between salt tolerance scores in the two-growth stage were not significant in either parents or crosses and were not significantly dif- ferent, the r value for crosses with 13 d.f. was rela- tively small (r=0.28) compared with the r value for parents 54 with 4 d.f. (r=0.81) which is near the edge of significance at the 5% level (t=2.723). The biological conditions which led to such a small r value in crosses while a relatively high r value in parents are not at all obvious. The following hypotheses are in- troduced for discussion. If salt tolerance in the two growth stages involved two separate gene systems the correlation of the two salt tolerance scores in the partially heterozygous F2 progeny is expected to be influenced by differences in the degree of dominance in the two growth stages while the correlation of the two salt tolerance scores in the homozygous parents is free from dominance effect. The absence of strong evi- dence of separate gene systems reduces the validity of such a hypothesis. Given the facts that salt tolerance readings were not made on the same F2 genotypes in the different growth stages and since the sample size was also different, (120 F2 seeds in the germination stage vs 18 plants in the early growth stage), and the fact that heterozygosity in the F2 population was relatively high, it is not unlikely for a situation like this to stem from a sampling error. It is expected that selfing the F2 progenies for a few 55 more generations beyond the F would fix most of the genes 2 and result in a significant positive correlation between salt tolerance in the two growth stages. In completing the discussion of the results of the genetic inquiry of salt tolerance in barley some remarks are in order. Firstly, all the tests regarding salt tol- erance were confined to a single salt, namely sodium chloride. Thus the possibility that other gene systems may surface if other salts (single or combined) are used as the stress element cannot be ruled out. Secondly, plant growth is a continuous process and stages of growth often overlap with no distinctive boundaries between se- quential stages. Therefore, reference to growth stages in this manuscript is rather artificial and done so only for convenience. Thirdly, the interdependency that char— acterizes some growth stages may result in salt tolerance or salt sensitivity in one growth stage influencing and/or ' depending on tolerance or sensitivity in the proceeding growth stage or stages. Finally, there is no current evi- dence that the investigated growth stages are, necessarily, the most critical stages in determining barley's overall tolerance to salt. We do know, however, that a seed must 56 germinate before a crop can be harvested and so germina- tion is one critical stage. More related information on the subject of salt tolerance is undoubtedly needed to expand our knowledge of the genetic basis of salt tolerance in barley. SUMMARY AND CONCLUSION This study acquired information on the genetic basis of salt tolerance in barley. The germination stage test indicated that salt tolerance in barley appeared to be governed by genes possessing partial dominance. Parent 1 seems to possess most of the dominant genes and Parent 5 most of the reces- sive alleles. The best combination was the cross 1 x 3. Genetic control of salt tolerance seemed somewhat differ- ent as the plant growth stage advanced. The early growth stage test revealed that salt tolerance at this growth stage was controlled by dominant genes. Dominance appeared to be near complete. Parent 3, seemed to possess most of the dominant genes and Parent 6, most of the re- cessive alleles. The most tolerant combination in the early growth stage from the six parental diallel-cross was the cross (2 x 3), which surpassed even the best parent in the set. Non-allelic gene interaction appeared to play a role in determining the inheritance of salt tolerance in this growth stage. 57 58 The degree of tolerance was found to be somewhat consistent in the two growth stages in the parents, and not so in the crosses. This was attributed most probably to the differences in sample sizes in the two stages and to the heterozygous nature of the F2 population. LITERATURE CITED LITERATURE CITED Ayers, A. D. 1953. Germination and emergence of several varieties of barley in salinized soil cultures. Agron. J. 45:68-71. Ayers, A. D., Brown, J. W. and Wadleigh, C. H. 1952. Salt tolerance of barley and wheat in soil plots receiving several salinization regimes. Agron. J. 44:307-310. Ayers, A. D. and Hayward, H. E. 1948. A method for measuring the effects of soil salinity on seed germination with observation on several crop plants. Soil Sci. Soc. Am. Proc. 13:224-26. Ayers, A. D., Wadleigh, C. H. and Bernstein, L. 1951. Salt tolerance of six varieties of lettuce. Proc. Amer. Soc. Hort. Sci. 57:237-242. Ballantyne, A. K. 1962. Tolerance of cereal crops to saline soils in Saskatchewan. Can. J. Soil Sci. 42:61-67. Bernstein, L. 1961. Tolerance of plants to salinity. Proc. Amer. Soc. Civil Eng. 87:1-12. Bernstein, L. 1964. Salt tolerance of plants. U.S. Dept. Agr. Inf. Bull. No. 283, 23pp. Bernstein, L. and Ayers, A. D. 1951. Salt tolerance of six varieties of green beans. Proc. Amer. Bernstein, L., Ayers, A. D. and Wadleigh, C. H. 1951. The salt tolerance of white rose potatoes. Proc. Amer. Soc. Hort. Sci. 57:231—236. 59 10. 11. 12. 13. 14. 15. 16. 17. 18. 60 Bernstein, L. and Fierman, M. 1956. Laboratory studies on salt distribution in furrow- irrigated soil with special reference to the pre-emergence period. Soil Sci. 83: 249-263. Bernstein, L., Francois, L. E. and Clark, R. A. 1966. Salt tolerance of N. Co. Varieties of sugar cane. I. sprouting, growth, and yield. Agron. J. 58:489-493. Bernstein, L. and Hayward, H. E. 1958. Physiology of salt tolerance. Ann. Rev. Plant Physiol. 9:25—46. Bhatti, A. S. and Loneragan, J. F. 1970. Phosphorus concentrations in wheat leaves in relation to phosphorus toxicity. Agron. J. 62:288—290. Brown, J. W. and Hayward, H. E. 1956. Salt tolerance of alfalfa varieties. Agron. J. 48:18-20. Brown, J. W. and Wadleigh, C. H. 1955. Influence of sodium bicarbonate on the growth and clorosis of garden beets. Botan. Gaz. 116-201-209. Brown, J. W., Wadleigh, C. H. and Hayward, H. E. 1953. Foliar analysis of stone fruit and almond trees on saline substrates. Proc. Amer. Soc. Hort. Sci. 61:49-55. Carlos, T. B. and Bingham, F. T. 1973. Salt toler- ance of Mexican wheat: 1. Effect of N03 and NaCl on mineral nutrition, growth, and grain production of four wheats. Soil Sci. Soc. Amer. Proc. 37:711-715. Chatterton, N. J., McKell, C. M., Bingham, F. T. and Clawson, W. J. 1970. Absorption of Na, C1, and B by desert saltbush in relation to compo- sition of nutrient solution culture. Agron. J. 62:351-352. 19. 20. 21. 22. 23. 24. 25. 26. 27. Dewey, Eaton, Eaton, Gauch, Gauch, 61 D. R. 1962. Breeding crested wheatgrass for salt tolerance. CrOp Sci. 2:403-407. F. M. 1942. Toxicity and accumulation of chloride and sulphate salts in plants. J. Agr. Research 64: 357-399. F. M., McCallum, R. D. and Mayhugh, M. 1941. Quality of irriguation waters of the Hollister area of California with special reference to boron content and its effect on apricots and prunes. U.S. Dept. Agr. Tech. Bull. No. 746, 60 pp. H. G. and Eaton, F. M. 1942. Effect of saline substrate on hourly levels of carbohydrates and inorganic constituentS' of barley plants. Plant Physiol. 17:347-365. H. G. and Wadleigh, C. H. 1951. Salt toler- ance and chemical composition of rhodes and dallis grasses grown in sand cultures. Botan. Gaz. 112:259-271. Greenway, H. 1962. Plant response to saline sub- Hayman, sttates. I. Growth and ion uptake of several varieties of Hordeum during and after sodium chloride treatment. Aust. J. Biol. Sci. 5: 16-38 0 B. I. 1954. The theory and analysis of the diallel cross. Genetics 39:789—809. Hayward, H. E. 1954. Plant growth under saline con- ditions. In Reviews>of research on problems of utilization of saline water. Arid zone programme, UNESCO, Paris, p. 37-71. Hayward, H. E. and Bernstein, L. 1958. Plant-growth relationships of salt-affected soils. Botan. 28. 29. 30. 31. 32. 33. 34. 35. 36. 62 Hayward, H. E. and Lone, E. M. 1941. Anatomical and physiological responses of the tomato to varying concentrations of sodium chloride, sodium sulphate, and nutrient solutions. Botan. Gaz. 102:437-462. Hayward, H. E. and Wadleigh, C. H. 1949. Plant growth on saline and alkali soils. Advances in Agron. 1:1-38. Itai, C., Richmond, A. and Vaadia, Y. 1968. The role of root cytokinins during water and salinity stress. Israel J. of Botany 17: 187-195. Jinks, J. L. 1954. The analysis of quantitative inheritance in a diallel cross of Nicotiana rustica varieties. Genetics 39:767-788. Levitt, J. 1972. “Responses of plants to environ- mental stresses." Academic Press New York and London. Lyon, C. B. 1941. Responses of two species of to- matoes and the F generation to sodium sul- phate in the nutrient medium. Botan. Gaz. 103:107-122. Marshall, J. B. 1942. Some observations on the tolerance of salinity by cereal crops in Saskatchewan. Sci. Agr. Canada 22:492-502. Millington, A. J., Buruill, G. H. and Marsh, B. 1951. Soil salinity investigations, salt tolerance, germination and growth tests under controlled salinity conditions. Dept. Agr. West Australia leaflet No. 1052. Nieman, R. H. and Poulson, L. L. 1971. Plant growth suppression on saline media: Interaction with light. Botan. Gaz. 132:14-19. 37. 38. 39. 40. 41. 42. 43. 44. 45. 63 Osmond, C. B. and Greenway, H. 1972. Salt responses of carboxylation enzymes from species differ- ing in salt tolerance. Plant physiol. 49: 260-263. Pearson, G. A., Ayers, A. D. and Eberhard, D. L. 1966. Relative salt tolerance of rice during germination and early seedling develOpment. Soil Sci. 102:151—156. Porter, L. K. and Thorne, D. W. 1955. Interrelation- of carbon dioxide and bicarbonate ions in caus- ing plant chlorosis. Soil Sci. 79:373-382. Sevast'janov, V. I. and Matuhin, G. R. 1961. Salinity resistance of barleys. Trud. Rostov. Otd. Vsesojuzn. Bot. Obsc. (Proc. Rostov. Div. All-Un Tot. Soc.). 1960 No. 1:148-152 from Ref. Z ' (Ref. J.) 1961 No. 17: Abot. 17G447: p. 51 (Russian). Strogonov, B. P. 1964. "Physiological basis of salt tolerance of plants." Acad. Sci. USSR, Davy and Co. New York. Tal, M. and Gavish, U. 1973. Salt tolerance in the wild relatives of the cultivated tomato: water balance and abscisic acid in Lycopersicon escu- lentum and L. peruvianum under low and high salinity. Aust. J. Agri. Res. 24:351-361. Uhvits, R. 1946. Effect of osmotic pressure on water absorption and germination of alfalfa seeds. Amer. J. Botany 33:278-285. Wadleigh, C. H. and Fireman, M. 1948. Salt distribu- tion under furrow and basin irrigated cotton and its effect on water removal. Soil.Sci. Soc. Amer. Proc. 13:527-530. Whitmore, E. T. and Sparrow, D. H. B. 1957. Labora- tory micromalting technique. J. Inst. Brew. 63:397—398. 64 46. Younis, A. F. and Hatata, M. 1971. Effects of chlorides and sulphates of sodium, potassium and magnesium on germination of wheat grains. Plant and Soil 34:183-200. 47. Yousif, Y. H., Bingham, F. T. and Yermanos, D. M. 1972. Growth, mineral composition, and seed oil of sesame (Sesamum indiCum L.) as affected by NaCl. Soil Sci. Soc. Amer. Proc. 36:450- 453. APPENDIX APPENDIX 1. The Hoagland's No. l nutrient Solution used as the basic nutrient media to grow barley plants to the four-leaf stage. Six stock solutions were prepared separately as described below by dissolving in distilled water the designated amount of the chemical compound and then bringing up the volume of the solution to one liter. Stock Solution No. Concentration per Liter l 1 M KH2P04 2 1 M KNO3 3 l M Ca(NO3)2 4 l M MgSO4 5 26.3 9 Fe chelate f 2.86 g H3BO3 1.81 g MnC12.4H20 6 a 0.22 g ZnSO4.7H20 0.08 g CuSO .5H 4 2O \0'016 g M003 66 To prepare one liter of the basic nutrient solution the specified amounts from each stock solution were added to 985 ml of distilled water as follows: Stock Solution No. ml/liter l l 2 5 3 5 4 2 5 l "IAIAHIAIHAT