THE RELATIONSHIP OF HESSIAN FLYV INFESTATION T0 SOME CHARACTERS ‘ OF THE WHEAT CULM “9122333 FER THE mm m: M. s] ' Dean C; Anderson » ; .9}934Z: 2‘," ‘qqnvcu/l b} m RELATIONSHIP or means mr INEBTATIOII 'ro son CHARACTERS or mm mm aULu AOUOMDGm T8 The writer is grateful to Professor 1. 1. Down and Mr. H. M. Brovn for guidance and aid in conducting this problem and for the review of this thesis. THE RELATIONSHIP OF H3881“ FLY INEBTATION TC CON! CHARACTERS OF THE WHEAT CULM THESIS RESPECTFULLY SMITTED IN PARTIAL FULFILLMENT FOR THE DEGREE OF MASTER OF SCIENCE AT MICHIGAN STATE COLLEGE OF AGRICULTURE AND APPLIED SCIENCE DEAN C. ANDERSON . 1934 'C 0 MI I THE I. II. III. IV. V. TABLE OF CONTENTS INTRODUOTIOR OBJECT REVIEW or LITERATURE A. nuns AND VARIETIES 0F GRAIN 111.1%!) E. means In LARVA m: we METHOD OF TAKING moo O. BIOLOGICAL STRAINS or HESSIAN FLY D. RESISTANCE OUT: TO MORPI'DLOGICAL CHARACTERS E. RESISTANCE DUE To PHYSIOLOGICAL CHARACTERS r. LODGING mam INVESTIGATION A. MATERIAL B. muons 1. LABORATORY a. 1932 CROP b. 1933 CROP 2. mm RESULTS AND OIaouesIon A. INTER—STRAIN DIFFERENCES 1. menu In INFESTATION z. BREAKING STRENGTH OF OULMS 3. WEIGHT or Guns 4. DIAMETER or onus 93802 moomsiqmasuhacn NNI—‘H H'Hl-‘I—‘HHHH ooqoacnmmmo VI. VII. VIII. B. DIFFERENCES BETWEEN FLY-FREE AND INFESTED GROWS C. RELATIONSHIPS 1. INTRA-ANNUAL CORRELATIONS 2.. COEFFICIENTS OF CONTINGENCY 3. INTER-ANNUAL CORRELATIONS SUMMARY CONCLUSIONS LITERATURE CITED 18 19 19 22 23 25 26 37 INTRODUCTION Hessian fly’, Phytophaga destructor, damage to win- ter wheat in.Michigan is generally agreed to be a source of great loss when.wheat is planted early, and of minor loss when it is planted late. The damage may be recognized in.the field by single culms breaking over where the flax. seeds, the puparia, develop. The Object of this study was to determine Whether there were any inter-strain'* differences in the extent of fly injury on the wheat grown on the Experiment station farm at East Lansing. If there were any differences, it was desired to determine the relationship of susceptibility to breaking strength, weight, and diameter of mature culms, and lodging as observed in the field. Such a relationship would.be of value in selecting resistant strains. A second phase of the study involves the inter—relationship of these characters. ’ Hereafter fly will be used to mean Hessian fly. “ The word strain is used to include strains and varieties. -2... REVIEW OF LITERATURE winter wheat is infested by fly in the fall or spring. If in the fall, the infested culm usually dies during the winter; if in the spring, the larva attack the culm and develop there as a flaxseed, under which condition the culm usually breaks over where the flax— seed is developed. The spring infestation is considered in this study. Painter, et a1. (14) after comparing several methods of recording infestation for eight check varieties show that the probable errors are least for the per cents of infested tillers, and conclude that for medium or heavy infestations the per cent of infested tillers is consid- ered to be the best measure of infestation. KINDS AND VARIETIES OF GRAIN INJURED McCalloch and Salmon (10) after counting the total number of eggs on 100 plants from each of 87 kinds and varieties of wheat, oats, barley, rye, emmer, einkorn, and spelt, conclude that the fly appears to show a pre— ference for common wheat compared with barley, oats, einkorn, spring emmer, spelt, or durum wheat. Black winter emmer, Poulard wheat, and Polish wheat were as heavily infested with eggs as many of the common wheats, and rye showed a very heavy infestation. However, rye, barley, oats, durum wheat, Poulard wheat, Polish wheat, spelt, emmer, and einkorn all showed a low or no in- festation with flaxseeds. HESSIAN FLY LARVA AND ITS METHOD OF TAKING FOOD Easeman (7) says 'the Hessian fly larva appears to be well equipped physically to take food from the host wheat plants and a study of its digestive tract fails to explain why it is able to mature more successfully on one strain of wheat than another". BIOLOGICAL STRAINS OF’HESSIAN ELY painter (12) concludes that ”the Hessian fly popu- lation of any one locality consists of a mixture of two or more genetically distinct strains which differ in their ability to infest various wheat varieties. In general, the soft wheats frequently are more or less resistant in the hard wheat belt, while in the soft wheat bolt the va- rieties which are resistant are more frequently the hard wheats. These field Observations give some indication that we may expect varieties to be most heavily infested where they are grown in greatest abundance“. RESISTANCE DUE TO MORPHOLOGICAL CHARACTERS ROberts, et al. (16) noted that those varieties of wheat with large, coarse, strong straw were less liable to injury from fly than weak strawed and slow growing varieties. MOOolloch and Salmon (ll) found that in some of the resistant strains of wheat the ligule was slightly higher than in the susceptible strains and, therefore, offered a barrier which prevented the hatching larva from reaching their normal feeding place at the base of the plant. RESISTANCE DUE TO PHYSIOLOGICAL CHARACTERS Painter, et al. (14) point out that I'resistance pro- bably is due to physiological rather than gross morpho— logical characters and that in young plants it resides at the base of the leaf sheaths, that is, at the crown of the plant. A large per cent of the larva are able to reach this region but are unable to develop on resistant varieties”. Easeman (6) showed that the extent of infestation in three varieties of wheat varied directly with the ash content. McOolloch (9) concluded that preliminary work at the Kansas Station indicated that in some way silica -5- is a factor in the resistance of wheat to fly. Painter (13) found that 'by following the individual histories of eggs of Hessian fly laid on different leaves, it is shown that no flaxseed survives from eggs laid on the outer (lst. leaf); 6.35% survive from eggs laid on 2nd. leaf; and 45.4% eur- vive from those on the 3rd. (central leaf). This decrease in survival on the different leaves is paralleled by the increase in deposition of cellulose or perhaps some condi- tion.arising with it". Parker and Painter (15) suggest that ”not much is known as to the basis of fly resistance, but it is believed to be protoplasmic; that is, related to the nature of the cell sap rather than to any readily observable morphological characters of the wheat plant“. LODGING LOdged grain is generally referred to as grain whose culms are bent over, and in severe cases may lie on the ground. Many references could be cited which deal with lodging, but only those dealing with the particular factors involved in this study are here given. -6- Ielton and.Morris (18) say that lodged grain may result from two sets of causes; first the interaction of those hereditary and environmental factors which produce weak stems, and second, the external factors such as violent wind, rain, or hail storms. Albrecht (l) weighed one centimeter sections of the straw and concluded that resistance to cross bending was positively correlated to the weight of the straw. Gerber and Olson (5) made a study of the correlation between lodging behavior and.the average size of culm, average number of bundles, average area of sclerenchyma, thickness of culm'wall, length of lignified cells, and thickness of lignified cell wall. They concluded that, in general, no one factor studied seemed to be correlated closely enough with lodging to be of much value as a selec- tion index. A number of investigators have studied the breaking strength of straw in.relation to lodging. Helmick (8), ‘Willis (19), Salmon (1?), and Clark and Wilson (2) have described apparatus for testing the breaking strength of culms. Salmon (1?) in studying the breaking strength of winter wheat varieties found that the soft winter wheats which are noted for their lodging resistance, generally showed the highest breaking strength, while the hard winter -7- wheats, which lodge easily, showed the lowest breaking strength. Clark and.lilson (2) working with fifteen com— mon.wheat varieties found no correlation between breaking strength and lodging. Davis and Stanton (3) in cat stud- ies reported that ”in general the reputed stiff-strawed varieties as determined by field observations when sub- jected to a mechanical straw strength test showed the highest resistance to breaking'. PRESENT INVESTIGATION MATERIAL Samples were obtained by taking a sevenpfoot section from.one of the outside rows of the five-row yield.plots from 111 strains of soft winter wheat in.1933 and 112 strains in 1933. The material was out wdth a sickle in 1932 and with a lawn grass edger approximately one4half inch below the ground line in 1933. In 1933 only the lower two-feet of the plant were saved for examination, while the upper portion was discarded. This method greatly facili- tated handling and minimized crushing the straw. -8- METHODS LABORATORY 1932 Orgy Forty to forty-five culms from each strain were examined.by cutting the leaf sheath to the node with a razor blade to determine whether or not the 0ulm.was infested with fly. With a simple recording device, the diameter of the culm was taken in the center of the second internode above the ground. The breaking strength tests were made on the second internode with special equipment used with a Toledo scale reading in grams. The apparatus (Pig. 1) was designed and made by I. H. Sheldon of the Agricultural Engineering Department. Inch culm was tested individually and was laid in a horizontal position across two supports placed 5 cm. apart and a force was applied to the middle of the culm through a screw and lover arm. The force was applied in a vertical direction, at a right angle to the length of the culm and the amount required at the instant of break- ing was recorded in grams. -9- Fig. 1. Apparatus used for testing the breaking strength of the culms. -10- A five centimeter section of the culm, taken where it had.been tested.for breaking strength, was then.cut out and.weighed on an analytical balance. W Between eighty and ninety culms from each strain were used in the determinations made on this crop. Each culm'was examined for flaxseeds, as in 1932, and those ‘with flaxseeds present were placed in one group and those free from flaxseeds were placed in another. The diameter was not taken when the larger number of culms was used because the time factor did not allow for all measurements. The fact that some of the culms were off-round, making it difficult to get an exact measure- ment, was another reason for not taking diameters this year. Instead of one,culm two culms were placed on the scale together and the breaking strength determined. A five centimeter section from.each culm was then cut out for‘weighing. As two culms were broken at a time, individual culm weights were not taken, but the weights of the two groups, those with flaxseeds and those free from flaxseeds, were taken separately. The number of culms weighed was record- ed, thus making possible the calculation of the average 'weight of two culms. -11- AmH swam no moanHaaoov he. Se oe m Hm S one rm. _ $2 2E. ._ we mmm 0m e m.m on mOm on meH monm = we mmm cm H m.m mm pom 0N mHmH onNm mmm ea mam Hm o m.» we Hmm mm eHmH moamm r mm has Om m m.~ me «on «a mHmH «Hmmm : me «mm mm o m.» me Hoe Hm homH HHmmm r am new mm H H.m me can mm momH mommHm how an Nam mm m m.m mm «mm mm mamH Omommm OmH om mom mm m m.m on 0mm mm mmmH memom e as «0m mm a m.~ mm mom me «mmH ammom r «a Nmm mm # m.m mm mmm he mmmH mHmaw a 2. see oe e Hm en Em mm SR 08:. ... 2. ear 3 e H.» me «an 3 SR H85 .. 2. S» me e Hm 3 3m 3 82 82a ... Nb mmm as m o.m He man 0» ummH eone = 2. 3h 5 SA: 3 SJ pee 238 .. ma mHm mm m o.m am mow mm mOmH mmbmHo = so mad am e 0.» am mmm on mOmH Nummm mmH ma mom cm H H.» He mmm bu mmmH mommH¢ r as mam me N Hem we mmm 0m mmmH bemde a ma ¢Hm mm m m.m mm pom mm mmmH moomHe mmH ma mom mm H m.m em mmm 0m .Mo mooao mm mm emm Hm H N.m em mom mm NOmH Hoon 0 macaw nsHao usaamHHHa owv no mashw nsHao sawsouph uuopoaHHda naspmHHHa. .aHao no cap no wcH peso sod .sHao no .aHso 9o Apmcouvm pace and astow. nxsoup soHosvoom mcH novoasHu uwaow_ wcHMSoup aoHpsvmou .02 .oz .02 omenopd amassed ucH hfih Tween smegmha owesoea owsao>a ucH th ¢OHm ansHm soHneooo< nmmH mmmH romances was usoHascHaaooou anuvm .H ages -J£&- mm mm mm om mm mm Nb mm mm mm hm mm mm mm mm Nb hp Hm ¢> Hm Hb mm mm hm Om mm mm mm 05 we we on wmfi mnm ”Hm wmb Odm mom 5mm mum 0mm Nmb mow bum mam «mm Nww OH» mum mom HHm wmm dam mam 55¢ mom wo¢ mmm boo mom mm mm mm mm mm mm mm mm mm mm mm mm mm cw mm mm nfi Hm H¢ om Om we 0d mm pm 00 mm 6» mNmH mmmH HmmH mHmH meH kaH mHmH ¢HmH wmmm comm mwNN mmmm mmNN mmmm ¢mNN #wwm mmmm Hmmm omNN mwmm .mbmH dbmH meN mme nbmH HbmH nme NmmH ObmH wmmm mbmH bbmH «raidTFTQImicIcIcIcIricIcIcId'd'«>1'w'd'd'd'd‘Nirdrioirtrarqrtra 0.000.000.0000 00...... QBO’OQOHOONd'QKOCDQOSBOOOI-INQQG MNNNNNI’JMMI’JI’DMNWNNMBOMMSONNN «>b-w3xPCD so... NNNN (h e N wow wow hem mm hm mm mm mm mm mm om an on Om mm mm mm om mm H0 H0 on N¢ mfi HG Nd an on o» H» mm 0* h» an mm flew ”Hm 5mm OH» mmm we» won mm» #0» won mHe flaw mmm mmm nwn Own mHn NH» 009 0mm dmm up» so» mmw mam mam mum mmH 5H» van on» Hm» AmH swam so me OmHH mm meeH mm >¢¢H om mva mm meeH mm medH mm meeH mm H¢¢H om mmeH mm mmmH he mmeH mm mmmH me NmmH mm HmmH om mmmH mm ava 0m mmmH rm hmmH mm mmmH mm «mmH om memH me memH me memH ow HNmH mm mde Hm HemH mm mHmH me mHmH b mmmH mH mmmH mm HmmH mm 0mmH 3.5383 memm ”Hme HHmNm onmmm mmbmm smvmw memm pomm vOm¢m memon QOHOHw momma nowmo Nomwm HOmwm dmbmm acmbm membm mommo ”comm ONHeflv mom>H¢ Non>H¢ Hommfim deHQ NHmH¢ mNNHHN NOHHHN momma HHHQm Nommo Hommm 3 8'3': 8 3 :I' 3 e5 2.» 2,.» a 8 = 8 hm» I own 8 mm» e 3 mm» m¢n B 8 mm» mm» mom 3 wow A Nantes .H .2: -13- mm Nb mp 05 ms mm as as as >5 mm Hp am Oh Hr up on oh oh oh «5 mo om me Oh Nb H» n» was «we mam ppm bow mmm mmm vac «mm own «am Can moo mum mac omo ¢¢m ham #Gb flab mmm bum web cam mam bmm com am mm mm m0 on om me mm H» mm w» on mfi an H» H0 0» m» mfi hm wm monm 50mm 00mm momm menu bmmu momH >mmm mmmm «mNN menu Honu nmmm Nmmm HmNN meN nflmH NHmH H¢mH aan homH oomH mOmH nomH NomH HOmH mNmH bmmH wNmH mmmH «Numeric-ontonunnnnmmmmnnnnnmnnnnw HHHHHHOOHHNNHHQNNNQ‘IOOHOONQQ‘IOQ‘IO OOQOHNNNNHNDNQ‘OQQQboanfl'b-OHHOG an an ~¢ N0 «0 mm 6v an an an m0 an an on m» N» mfl aw m0 mm an on b» an on mm 00 an mun Hun an» no» mum nan «on «on 0H» #0» mHn mHn. 0mm mom new Han 6nd mum Ham «H» Haw ham mom ”Hm 06» #H» Nb» mm am >0 mm mm fim «6 mm mm an an bu ON Hm m¢ mi m» an an 0H on km N» H0 mm 00 «N mm mbmH dsmH nbmH HbmH ObmH memH omnH mw¢H nm¢H HowH bmmH cmmH mmfiH wde bde HawH oadH ao¢H hofiH omvH mnnH ian nnMH HnnH OnnH amnH mmmH emflH nm¢H HmQH 3H ewes ee eaeaeeeov mHHmNm «HHmNm NHHmNm monNm monNm HObH¢ Homemm HNHHHQ mnoHHm GHOHHw Howme nowHHo mHmdw HHaew Home» Hemap ¢o¢wb ¢o¢wb «owns Howe» madam owfiem 0H¢¢a bHvfim HH¢dm wov¢m «mawm mamww mama» «Nam» 8 2.» 3.80 AvonnuvnOOV eH oHadB ~14- ee Nee »» »H»~ e H.» oe em» a» aHeH re» »s Hoe h» AH»~ e e.~ o» e»» em hHeH. so» as ch» em hH»~ H e.~ h» ohm on «HeH me» e» es» en eH»H e ».H a» »HH »» »HeH «or a» Has H» HH»H » ».» he he» s» eehH as» e» oHe e» eeeH H e.» «e eee He. HahH me» am oee e» »HeH » H.» he He» He eoeH he» re eee «e seam e H.» »e o»» »e eeeH e»h we she »» »e- » e.» he ooe Hm oHeH «oeeH» . as can as comm » H.» e» »»H eH eoeH HoeeH» »»» H» Hee ea e»hH H e.» »e em» em eseH HoNeH» . H» e»s 0» menu. a e.» »e ea» »» aeeH HomeH» H»» mm eHe e» ream e he as» H» heeH eoHeHe a ma am» He e»hH H »e ea» em «seH eoHeH» r om has He »»»H H as o»¢ «H »aeH »oHeH» . H» Hse we e»hH H he see a» oeeH HoHeH» o»» «a He» »» »»mH H we aHe »¢ eeeH eooeHe m »» see a» H»»H o eh eee e» HeeH eooeH» e 8 an. S. 82 o e» on. H me: 83H» 9..» ea or» he HHmH H or see o» es»H Hoeeee «He e» ems we sehH H «e are e» eeeH »H» he was »e HH»H » He em» »» aeeH so» es «a» a» loH H e» Ha» H“ »»»H eoHeee ewe es ewe e» eeeH o as «ea a" eeeH «HeeH» m as »ee ee hehH o as see he aeeH hoeeHe e H» are e« when H «a e»» e» emeH eoaeHe «»e Ha ea» ha eeeH N He he» »» «NeH roe HeasHfieev .H .23. FIELD The lodging notes taken in 1932 were strictly field observations. Five ratings were used, grading from no lodging, indicated by O, to complete lodging, indicated by 4. There was no lodging in.1933. RESULTS AND DISCUSSION In Table l are presented, for 1932: the per cent fly infestation for the strain, the average breaking strength, weight, and diameter of single culms, and lodging for the plot; and for 1933: the per cent fly infestation for the strain, and.the average breaking strength and average weight for two culms. INTER-STRAIN DIFFERENCES HESSIAN ELY INFESTATION The standard error of a single determination (8.3,) ‘was calculated from the check plots, of which there were 55 in 1932 and 52 in 1933, to determine whether there were any significant differences in the per cent of fly infestation. Since single determinations of fly infesta- tion were used in all cases, except with the check, it -16- was felt that as an indicator of possible significance of inter-strain differences the standard error of a difference between single determinations as obtained from the check plots could be used as a measure of dif- ferences between strains. The check strain (Red.Rock) had an average infesta— tion in 1932 of 30.3% with a s.r., of 9.9; and in 1933 of 39.3%‘with a SJ.a of 8.8%. The SJ.a was multiplied by the square root of two, which gives the standard error of a difference, and then multiplied by 2.2 because odds of 35:1 were deemed to show significance. It was consid- ered that those strains which had a difference as great as or greater than this amount were significantly differ- ent from each other. In 1932 the per cent fly infestation varied from 7 to 78%, and in 1933 from 5 to 75%. There was a significant difference in the per cent of fly infes- tation for the strains of winter wheat studied. BREAKING STRENGTH OF CULHS The standard error of the mean breaking strength for each of several strains was calculated to determine whether or not the strains differed significantly in breaking strength. A difference between.the means for two strains of 50 grams in 1932, when single culms were used, and about 80 grams in 1933, when two culms were used was found to be significant. In 1932 the average breaking strength for the strains varied from 199 to 530 grams, and in 1933 from 408 to 855 grams. There were signifi- cant differences in the average breaking strength be- tween many of the strains in spite of the fact that individual values of these strains overlapped. llIGHT‘OF’CULMS The standard error of the mean for the weight of single culms for each of several strains was calculated in 1932 to determine the significance of the difference. The means varied from 29 to 58 mg. and a difference between the means of about 6 mg. was found to be signi- ficant. In 1933, individual 0ulm.weights were not made so the standard error of a single determination of the means for the weights of two culms from the 52 check plots was calculated and that standard error used to determine dif- ferences between strains. If strains differed in the weight of culms‘by approximately 12 mg. they were found to be significantly different from each other. The aver- age strain weight for two culms varied from 52 to 96 mg. -18- DIAMETER.OF CULHS The standard error of the mean diameter of the culms for each of several strains was calculated and the stand- ard error of the difference of these strains was deter- mined. When.the means of the various strains differed from.each other by about .3 mm., they were significant. The mean diameters for the strains varied from 2.3 to 3.9 DIFFERENCES BETWEEN FLY-EREE AND INEESTED GROUPS Student's method of interpreting paired experiments was used to determine whether there were any significant differences between the breaking strength, weight, and diameter in 1932, and breaking strength and weight in 1933, of the culms that had flaxseeds present and those free from flaxseeds for all of the strains. In nearly every strain.the fly-free group average was higher in breaking strength, weight and diameter than the corre- sponding fly-present group average. The consistency of the results is reflected in the large values of '2' which are shown in Table 2. -19.. Although the group averages within a strain were significantly different from.ea0h other, yet the indi- vidual or two culm determinations show a considerable overlapping and not a clear cut demarcation. Table 2. '2' values for comparisons between fly- free and infested groups within strains. 1932 1933 Breaking strength ..86 2.79 Height ‘ .85 2.85 Diameter .72 RELATIONSHIPS INTRA-ANNUAL CORRELATIONS Simple correlation coefficients, Table 3, were cal- culated between.all characters, except field lodging, to determine the relationship of the several characters to one another. Coefficients of contingency, Table 4, were calculated to show the relationship of field lodg- ing to the other characters. -30- Table 3. Correlations between the various characters 1932 1933 Breaking strength to per cent fly infestation -.O5 -.08 Weight of culm ' ' ' ' ' -.08 -.18 Diameter of 0ulm ' ' ' ' ' .01 Breaking strength to weight .92 .91 Breaking strength to diameter .59 Weight to diameter .74 Breaking strength to weight and diameter .95 The significance of the correlation coefficients was determined by Fisher's (4, Table v.11.) 't" test, using the level of significance which gives odds of 100:1. This test shows that there was no significant relationship between per cent fly infestation and the three culm char- actors. The coefficients showing the inter-relationship of the culm characters to one another are all significant. Although the coefficients between fly infestation and the three 0u1m characters are not significant it is noteworthy that those strains which had the lower per- centages of infestation tended to have stronger or heavier -21.. 0u1ms than those with the higher percentages of in. festation, and that the diameter of the calm showed almost no relationship to infestation. ‘ The degree of relationship between the culm char- actors, as shown by the significant coefficients of Table 3, was examined by Nright's (20) coefficient of determination r3. considering that values of r3 above .50 indicate a strong and those between .50 and .30 a a medium degree of relationship, it was found that breaking strength is very strongly correlated to weight and to weight and diameter taken together. From one viewpoint this strong relationship might be expected because an increase in the weight of the culm may be due to an increase in the thickness of the calm walls, the individual cell walls, or of the sclerenchyma with a resultant increase in breaking strength. On the other hand, an increase in culm weight might be due merely to an increase in diameter without a corresponding increase in thickness of culm wall, cell wall, or sclerenchyma ’ tissue. Such a condition would result in a weakening of the calm and hence a lowering of the breaking strength. Though the material was not examined for these characters, -22- and hence nothing definite can be said about them, yet whatever the cause of the increased weight, the heavier the straw the greater was its ability, on the average to resist breaking. The low value of the coefficient (r H .59) for the relationship between breaking strength and diameter might have been due to the fact that the larger straws did not always have the thicker culm walls, and due to the fact that not all of the 0ulms were round. That the first condition existed is somewhat shown by the value of the coefficient (r = .74) when weight and diameter are com- pared o 00E EPIC IENTS O F CONT INGENCY The field lodging notes were purely observational and arranged in only five different classes, therefore, it was felt that the best coefficient to use in deter- mining the relationship of this character to the other. four would be the coefficient of contingency. The values of this coefficient are given in Table 4. -33— Table«4. Coefficients of contingency between lodging and the other characters studied in 1932. Lodging to per cent fly infestation .46 ' ' breaking strength of culm -.50 ' ' weight of culm -.50 ' ' diameter of culm -.57 The significance of the values was determined by Fisher's (4) 13 method. This test shows that all four determinations were highly significant. The nature of this coefficient is such that it cannot be interpreted by Iright's (20) test, the coefficient of determination, but can be interpreted as a measure of the degree of association between the two characters. The coefficients, in Table 4, indicate that the greater the fly infestation the greater the amount of lodging, and that the strains which tended to have the stronger, heavier, or larger culms lodged less than those with the weaker, lighter, or smaller 0ulms. INTER—ANNUAL CORRELATIONS The ability of a quantitative character to express itself from year to year, in spite of varying environp mental factors, may be measured in some degree by employ- -34.. ing inter-annual correlations. A high value of the coefficient would indicate that environmental factors have not materially affected the expression of the par- ticular character or, if they‘have, they have treated all strains in the same manner and in such a way to allow strain differences to show up. If the inter- annual coefficients are small, environmental factors have tended to cover up possible inherited tendencies. Table 5 gives the inter-annual coefficients for the three possible comparisons. Table 5. 1932 to 1933 inter-annual correlation coefficients. Fly infestation .30 Breaking strength of culm .47 Height of 0ulm. .54 The weight of culm (r = .54) was the character showing the greatest stability, while per cent of inp festation (r . .30) was most variable and most affected by environment. The low coefficient for infestation might have been thought due to an unequal infestation over the nursery, were it not for the fact that the checks distributed throughout the nursery showed no such gradation. ~25- The lack of a stronger relationship in.breaking strength (r g .47) can doubtless be accounted for by the fact that such factors as weathering, handling, and diameter and weight of culm.affect this determination. The weight of culm would.naturally not be affected so much by weathering and handling as would.breaking strength, and this is reflected in the value of the coef- ficient (r H .54). SUMMARI Per cent fly infestation and field lodging for the strain, and breaking strength, weight, and diameter of the 0ulms were determined for 111 strains of soft winter wheat in 1932 and 112 strains in 1933. The differences between strains and the differences between fly-free and infested groups within the same strains, and the relation» ship of these characters to each other were studied in this investigation. The results show: (1) that there was significant inter—strain differences in fly infesta- tion, and breaking strength, weight, and diameter of the culm; and (2) that there were significant differences in.breaking strength, weight and diameter of the oulm between the fly-free and infested groups within the same Strains '-26- No significant correlation was found between fly infestation and breaking strength, weight, or diameter of the culm. The coefficient of contingency showed that there was a significant association.between fly infestation and lodging. Significant positive correlations were found be- tween.breaking strength and weight, breaking strength and diameter, weight and diameter, and between breaking strength and weight and diameter taken together. The coefficients of contingency showed that there were sig- nificant negative associations between lodging and average breaking strength, weight, and diameter of culm. The inter-annual correlations for fly infestation, breaking strength, and weight of 0ulms were positive and significant though none of them strong. CONCLUSIONS Though the strains of wheat differed significantly in fly infestation, and breaking strength, weight and diameter of 0ulms, the culm characters were not corre- lated with fly infestation, nor associated with lodging close enough to aid in the selection of fly resistant strains. The 0ulmtcharacters were positively correlated with each other but negatively with lodging. 1. 9. 10. -27- LITERATURE CITED Albrecht, Dr. Kuno. Untersuchungen fiber Korrela- tionen im Aufbau des Heisenhalmes, welche fur die Logerfestigkeit des Getreides von Bedeutung sind. Laudw Jahrbucher 37:617-672. 1908. Clark, H. R., and‘lilson, H. K. Lodging in Small Grains. Jour. Amer. Soc. Agron., 25:561-572. 1933. Davis, L. L., and Stanton, T. R. Studies on the Breaking Strength of Cat varieties at Aberdeen, Idaho. Jour. Amer. Soc. Agron., 24:290—300. 1932. fisher, Rt A. Statistical Methods for Research Ibrkers. Edinburgh: Oliver and Boyd. Ed. 3. 1930. Gerber, R. J., and olson, P. J. A Study of the Relation of Some Horphological Characters to Lodging in Cereals. Jour. Amer. Soc. Agron., 11:173—186. 1919. Haseman, Leonard. An Investigation of the Supposed Immunity of Some Varieties of Wheat to the Attack of Hessian Fly. Jour. Icon. Ent., 9:291-294. 1916. Haseman, Leonard. The Hessian Ply Larva and its ggggod of Taking Food. Jour. Econ. Ent., 23:316-321. Helmick, B. C. A Method for Testing Breaking Strength of Straw. Jour. Amer. Soc. Agron., 7:118-120. 1915. McColloch, J. N. The Resistance of Plants to Insect Injury. Kan. State Hert. Soc. Bien. Rpt., 37:196-208. 1924. HeColloch. J. I., and Salmon, S. C. Relation of Kinds and Varieties of Grain to Hessian Fly Injury. Jour. Ag. Res., 12:519-527. 1918. 11. 12. 18. 14. 15. 16. 17. 18. 19. 80. -33- McColloch. J. H., and Salmon, S. C. The Resistance of wheat to Hessian Fly Injury. Jour. Econ. Bht., 16:293-298. 1923. Painter, R. H. The Biological Strains of Hessian Fly. Jour. Econ. Ent., 23:322-326. 930. Painter, R. H. Observations on the Biology of the Hessian Fly. Jour. Econ. Ent., 23: 326-328. 1930. Painter, Reginald H., Salmon, 8. 0., and Parker, John.H. Resistance of varieties of Winter Wheat to Hessian Fly. Kansas Agr. Exp. Sta. Tech. Bul. 27:1-58. 1931. Parker, John H., and Painter, R. H. Insect Resistance in Crop Plants. Proc. Sixth Int. Congress of Genetics, 2:150—152. 1932. Roberts, I. P., Slingerland, H. V., and Stone, J. L. The Hessian Ply. Cornell Univ. Exp. Sta. Bul. 194:239-260. 1901. Salmon, S. C. An Instrument for Determining the Breaking Strength of Straw, and a Prelinunp ary Report on the Relation Between Breaking Strength and Lodging. Jour. Ag. Res., 43: 73-82. 1931. Nelton, P..A., and Herris, V. H. Lodging in Cats and‘lheat. Ohio. Agr. Exp. Sta., Bul. 471:1-88. 1931. 111119, N. A. An.Apparatus for Testing the Breaking Strength of Straw. Jour. Amer. Soc. Agron., 17:334—335. 1925. . wright, Sewall. Correlation and Causation. Jour. Ag. Res., 20:557-586. 1920-21. ‘ w .. ¢ 5 r . g . . a . n ‘ f V ' ' V ‘4' n * ,.| "f v "a ' .. . " ‘ _ ,..}f a, ‘ . -',_. ‘ ,,.‘- r. (up . j V U , XY‘ 1‘ Fe ‘. . ‘ . ’ a w .,'. , D ’ ‘ I min»anmrmwmmmr ”WWW 3 12