THE ERECT OF THE OAK HILT FUNGUS UPON
OAK HOOD
DISSERTATION
Presented in Partial Fulfillment of the Requirements
for the Degree Doctor of Philosophy In the
Graduate School of the Ohio State
University
WILLIAM tf.BRANDT, BJU, M.Sc.
The Ohio State University
1954
Approved ty:
'f e w i
ACKN0HLEDGEMENT5
The writer wishes to express his deepest gratitude to Or*
W.D. Gr^r for counsel and adrioe during this investigation, to Or*
H*C* Young for suggesting the problem, to the Ohio Agricultural Exp eriment Station for sponsoring it, to the Statistics Laborstoxy at the Ohio State University for guidance, to the Department of Mechanics of the Ohio State University for adapting a machine to the purposes of the investigation and for use of the machine, to the Department of
Electrical Engineering for use of an analytical balance and space in an air-conditioned laboratory, to the Forest Products Laboratory at
Madison, Wisconsin, for advice and for the isolates of wood-rotting fungi used in Part U of the study, to George J* Bart and Oren W*
Spilker for isolating associated fungi and procuring boards for blocks, and to maqy others who gave assistance and material aid*
\ If) 107 ii
CONTENTS
Introduction ------1
Part I. The Effect of the Oak Wilt Fungus on Oak Wood and on Sates of Decay ty sone Fungi Associated with It*
Materials and Methods ------5
Results ------16
Part II* Durability of Oak Wilt-Killed Oak*
Materials and Methods ------24
Results ------26
Summary —————————————————————————— 26
Literature Cited ------29
Autobiograpty ------30 -1-
THE EFFECT OF THE OAK >.TLT FUNGUS UPON OAK HOOD
INTRODUCTION
During the course of oak wilt investigations, majy workers ob served what appeared to be a rapid loosening of bark on trees which hud died of the disease* This loosening and sloughing of the bark seemed much more rapid than on trees killed ty other agents such as lightning and girdling* These observations gave rise to suspicions that the wood of oak trees killed ty the oak wilt fungus decayed more rapidly than the wood of heal tty oak trees* If accelerated decay actually did occur, it would have been necessary to remove oak wilt- killed trees from the forest proportionately sooner than other timber; the manufacture of lumber from diseased trees would have to take place sooner, and there would possibly be serious doubts regarding the dura bility of the lumber thus obtained.
This stu Breti) upon oak wood, the direct effects being possible decay of the wood by the oak wilt fungus; the indirect effects a possible oak wilt- induced acceleration of the decay caused by other organisms* There are two distinct aspects to this wood decay problem: (1) the decay of standing timber and (2) the rotting of structural timber and lumber. These are due to the interaction of two factors: envir onment, and the fungus inoculum. The forest environ— nt includes widely—ranging physical conditions such as temperature, light, and moisture, which influence the kinds of fungi growing in the forest and -2- the amount of vegetative and reproductive growth of these fungi. These in turn determine the relative amounts of fungus inoculum to which a tree might be exposed and how fast the tree will decay if conditions permit it to become infected. The internal environment of the standing tree is usually much different from that of lumber; for instance, the moisture content of a standing tree is generally greater than that of seasoned lumber. In addition( the external environment of much structural wood is more uniform than the external environment of the wood of a tree in the forest. The effect of the interaction of ail these factors, as well as others, is that the fungi which usual ay inlect standing timber do not commonly cause decay of structural wood. No doubt a field study involving periodic testing of diseased oaks which were left in the woods would have given the most direct information with reference to standing timber. Such a study was not feasible in Ohio since it would have meant purposely leaving centers of infection scattered throughout the forest areas, a practice not deemed to be in the public interest. Instead, a laboratory investiga tion was undertaken in which an effort was made to subject oak wood to preliminary treatment with the oak wilt fungus or its products and then to subject it to infection ty fungi found to inhabit oak wilt- killed trees. Part I deal» with this work. The resistance of wood to decsy is termed durability. Therefore, insofar as structural timber is concerned, the primary interest is the relative durability of 1 timber from healthy trees and from oak wilt- -3- killed trees* The Forest Products Laboratory at Madison, Wisconsin, has developed certain methods of ascertaining the durability of oak wood ty me«tns of fUngi commonly found to be causal agents of the rot ting of structural wood* (Scheffer, et al* 1949) These methods, slightly modified, were employed in the experiments described in Part II of this paper* Little work appears to have been done upon the effect of non wood -destroying organisms on the durability of wood* Here are some comments from an early bulletin about chestnut timber: "Strength and Durability of Piscase-Killed Timber*" "Preliminary strength tests conducted by the Forest Service upon chostrmt killed by the bark disease indicated that sound wood from dead trees is fully as strong as wood fron healthy trees. Discase-kiLied chestnut seasoned on the stump is probably more durable than green-cut chestnut used unseasoned, owing to the former's comparative dryness* In 1910 a railroad in Pennsylvania rejected a shipment of chestnut ties from Long Island because they were cut from diseased trees. The ties were left piled on the right of way for three years, but upon reinapection in 1913 were found to be so sound that they were finally accepted." (Nellis, 1913). Baxter and Gill (1931) studied the rate of decay of chestnut blight-killed chestnut but not the rate of decay of non-diseaso-killed timber so a cougari son is not available from their work* Scheffer and Lindgren (1940) state, "The evidence indicates that whatever differences there nmy be are not of practical importance," in referring to the relative durability of fungus-stained sapwood and non stained sapirood* However, they mentioned some investigators who felt, on the basis of their own research, that stained wood was less durable, Prescott and Dunn (1949) state that, "Saprophytic types (bacteria) occassionally may increase the rate of decay ty fungi," but fail to -4- cite ary experimental evidence* There is some data concerning the effect of preliminary rotting by one fungus upon the rate of subsequent rotting ty another fungus. Findlay (193*;) found that subjecting wood to the preliminary decay of a brown rot fungus had no effect upon the rate of decay caused by a second brown rot fungus* Likewise, a preliminary rotting Ty a brown rot fungus had no effect upon the rate of subsequent rotting Ty a white rot fungus; but if a white rot fungus partially rotted the wood first, the rate of subsequent rotting ty a bro-n rot fungus was in creased. (Findlay, 1940)* Lehman and Scheible (1924) showed that a preliminary inoculation of wood with Coniophora cerebolla Pers* in creased the rate of decay caused Ty some other fungi. In 1943, Roth reported that incipient decsy in oak crossties and posts enhances the rate of decay as compared to the rate of decay of sound wood* - 5- PUtT I THE EFFECT OF THE OAK WILT FUNGUS ON OAK VOOD AND ON KATBS OP DECAY FT SOME FUNGI ASSOCIATED WITH IT MA1EBXALS AND METHODS: Teat Blocks The primary interest in this investigation was the rate of decay of dead oak trees in the forest* As stated in the introduction, a dir ect inrestigation was not desirable for reasons of sanitation* The obvious alternative to this, the comparison of the rate of decomposi tion of pieces of healthy* oak wood with the rate of decomposition of diseased** oak wood in the laboratory was thought not feasible because of the great amount of variation in decay resistance among individual trees* Unless this variation were eliminated, it would not be possi ble to infer whether axy differences resulted frou the effect of oak wilt or fToa the inherent variation present before testing was started* This variation oould hsve been overcome ty taking a large enough sample of oak trees; however, since snail blocks had to be used due to certain naterial considerations, a large enough sample would have in volved the cutting of dosens of oak trees in various parts of Ohio, a procedure which would have exoeeded the financial limitations of the project* Another possible method of overooming the variation is to take the test blocks from the same location in the same tree, extensive studies having indicated that such a procedure is valid* (Scheffer, et al* 1949) • Therefore this method was adopted* The major difficulty • heal tty — refers to aiy oak tree not infected ty the oak wilt fungus* ** diseased — aigr oak tree having mrmptoaut of oak wilt* - 6- iabartnt in this procedure is that ths blocks are sithsr ftos healthy oak trass or all frou diseased oak trass* To attempt to obviate this probloa, ths blocks vers out from a singla board of heart ty- oak wood, raadoaissd sad half wars inooulatsd with ths oak wilt fungus or trsatsd with its astabolio products* It was hoped that any possible deoqr-aooeleratiag substaaos produced in whole oak trass would also be produoed sithsr in liquid culture or on ths blocks whoa ths oak wilt fungus was cultured thereon* Of oourss, it was realised that such would not necessarily be the case* Exoept where noted, hsartwood was used as test aatsrial* The oak wilt fungus is confined to ths outer portions of the sapwood, but the sapwood of oak wood is usually of little importance as limber* Unless sons effect were exerted oa the hsartwood, it would not likely be of interest froa an eoonoaic point of view* Therefore, hsartwood was used for all exoept the early portions of this work* Associated Fungi Sinoe the rate of decay of oak trees in ths forest was of interest, it was felt necessary to use fungi rnwnonly present in forest oaks* Although considerable information exists oonceming the fungi which are the most important destroyers of forest oak, it was not known whether these fUagi were also important destroyers of forest oak which has had oak wilt* Therefore, about fifty fungi were isolated from oak trees which had typical oak wilt mrwptons and from which the oak wilt fungus had been isolated* Seven of these were tested on oak blocks and four were fount to be wood-rotting fUngi* It is not known whether these fUngl m>st frequently deoonpose diseased forest oak, taut they W i n frequently observed during routine isolations of ths oak wilt fungus from disoasod oaks. Two of the four wore identified, Isolate 9 being Xylaris multi- flex Kse. and Isolate 47 being Aral liar ia asllea (Tahl.) Fr. Isolate 19 was a Basidixmyoete but otherwise unidentified. These three were tested extensively. The other wood-rotting fungus, Isolate 28, was used In one experiment only. Index of Deoar Several indices of amount of rotting nay be used. Most of them involve weight loss or loss of resistance to various stresses and strains such as compression and static-bending. Only a modified static-bending or weight loss test was practicable with the equipment which was available and since a decrease in static-bending is apparent before a decrease in weight, the former was chosen as the basis upon which rate of decay would be ascertained. A Scott fabric-tea ting machine was adapted for this purpose ty the Department of Mechanics at the Ohio State University (fig. 1). The rate of descent of the lower jaw of the machine was about 2 7/8 inches per minute, considerably faster than that of conventional static-bending machines. Three hundred eighty-four untreated blocks were broken on this wfi*»hint in order to test it and to obtain infor mation about variation in block strength which would enable estimates to be made of the sample sise neoessaxy to obtain the desired accuracy. At first, a frequency distribution of bending strengths having four modes was obtained; however, after fitting a slower motor (which gave the 2 7 / 8 " rate) to the machine this was overcome and a normal 1* A Scott fabric-testing Machine adapted for static-bending tests of small oak blocks* distribution of static-bending strength values was obtained* Static-bending strength was measured in machine units which usually corresponed to about three foot-pounds* The machine was never calibrated exactly since there was no necessity for determin ing absolute strengths* Of course, for a given experiment the machine had the same setting throughout, but settings were varied from experi ment to experiment* Procedure Fig. 2 represents the treatments given the blocks in this study* Boards about 4 feet in length, 3/4 inch in thickness and 4 inches in width were sawn from green logs* Identification of oak species was made ty the author, experienced woodsmen or reliable saw mill operators* Blocks about 1/4 x 3/4 x 3 inches were carefully cut on a power saw from a single board* After discarding blocks with crooked grain or other blemishes such as stains, offsize blocks, detected with the aid of a vernier caliper, were either thrown away or sanded to size* These operations were intended to reduce variation in initial bending- strength which of course would afford more accurate results* Randomization consisted of a thorough mixing of the blocks during the processes mentioned above plus mixing in a wire basket ty a standard rolling and inverting procedure* The testing of groups of the blocks on the static-bending machine and subsequent statistical analysis showed that randomization had actually been achieved* Three sets of blocks were analyzed; each set was made up of four groups of blocks; each group was made up of ten blocks* No statistically significant 3.. .:*s f«OW * SEl i ' T S 3 0 OAR 31 .1 I . AK V, N S C I* V" IS . -I* I', = T/r: E*cu s « FuTRACT PfiOOcCTS N rr * 1 L - 4 - c ‘ - A : c J- *-,: ' \ .s .. PE f-l E P E a I h* O h I . : N . ■ . J r r N ^ :c N C ■ : c ^ I ’’’BfA ’ A'S CTHf AT ASGiX N '. B A T ~N "'PEA' ASSOC-TREAT ASSCC N^ SLANT N3 5LANT , c Nu NC SLANT MC SLAN' 1 t r e a ta s s o ct r e a t ass; : h . l e o treatassoctreat a s s o- h r i r r SCRAPING c ~ WASHING ‘ BREAKING * a n a l'Z in c ’ REATASSOG ENDC A . t DRYING Fig. 2. flow diagram of block-testing procedure. -11- differences bttwtes the M a n bending-strengths of these blocks within a givan set was found at the I f * level ty means of a ono-vay analysis of variance. To gat some conoept of the amount of non-significant variation in one such set of blocks saa fig, 3, This represants the approximate extremes of variation commonly encountered among groups of untreated blocks. The blocks vere soaked under reduced pressure over night to try to insure uniform moisture in the blocks as vail as to provide an in creased supply of water for the long incubation period. It was felt that the addition of water at intervals could have introduced undesir able variation in the rates of decay and increased the chances for the occurrence of contamination. In methods III and IV the blocks were soaked in potato extract (200 g, potato boiled 1 hour in 1 1, of water) or the filtrate from 15-45 day old, liquid, potato extract cultures of the oak wilt fungus, Ihe oak wilt fungus used in this investigation was procured from a single endoconidium of an isolate obtained from a diseased tree in Ohio, For use as inoculum the oak wilt fungus was usually cultured in liquid potato extract at about 77* F, for 9— 12 days, but successful inoculations vere made with cultures as old as 25 days. Associated fhngi when applied as liquid inoculum vere also cul tured on potato extract for about the same length of time. Inoculations were made after cooling, Where liquid inoculum was used, 1 ml* was sterilely pipetted directly onto the blocks, A reason ably uniform inoculum was obtained ty placing the liquid culture of the desired fungus in a sterile Waring Blendor for a period of 10— 12 i. . vrg edn tegho orgop f1 lcs each blocks 10of groups four of strength bending Average 3. Fig. before aiy fungus action. fungus aiy before MEAN STRENGTH - MACHINE UNITS RAE BLOCKS TREATED N U -12- second** This method resulted in a acre rapid and uniform mycelial coverage of ths blocks than the placing of a piece of mycelium on them. Contamination was rare when due care was exercised. The placing of blocks directly upon slants of an actively growing fungus in 25 x 200 mi. test tubes was also a satisfactory method of inoculation and in addition maintained moisture at a higher level for a longer period. A possible source of error inherent in this method is that the wood could absorb various nitrogen compounds from the agar which might influence the rate of decay. Where E. fagacearua was allowed to grow on the blocks, an incu bation period of about 15— 30 days was sufficient to allow an easily visible growth of mrcelium to appear. A four month incubation period was arbitrarily selected for decay ty the associated organism. The humidity was not controlled but the temperature was maintained at approximately 77* F. The fungi on the blocks were killed in all experiments except one ty placing a few drops of fonnaldetyde in each tube and allowing several days for death to occur. Autoclaving , which was employed in the exception, was felt undesirable since different pH's on dif ferent groups of blocks could have resulted in differential tydroly- sis and consequently differential weakening of the blocks; of course other non-controlled variables s y have been introduced. The results in this one experiment indicate that something of this nature may have happened, (fig. 4; table 1 9 «xp. 9). The ^rcelium was then carefully scraped off, care being taken not to reawve any wood. The blocks were rinsed and allowed to become air dry. Since moisture content strongly affects bending-strength of wood when present in quantities below the fiber saturation point, care was taken to give all the blocks in a given experiment an opportunity to dry equally and evenly* Bending-strength was ascertained in arbitrary machine units on the machine mentioned above* The blocks were broken as indicated in Fig* 5. Where rotting was apparent, a multiple analysis of variance, two-way classification was employed to find whether the oak wilt fungus or the associated fungus had axy significant effect on the wood or whether there was axy interaction between them in their effect* Where little or no weakening appeared to have occured, a one-way analysis of variance was administered to detect axy possible differ ences between treatment means* -15- i oo ;> m ccK BRASS HEMTCTI TNDfR — BAIJ BHARINT, HOOK WHICH IS PULI EH DOWNWARD BNTII THE BIOCK BREAIS Fig. 5. Diagram of oak block in position for breaking. -16- KESULTSi In this portion of ths study in which ths main interest was ths effect of ths oak wilt fungus upon ths rates of decay caused ty assoc iated fungi and also the effect of ths oak wilt fungus on ths oak wood itself, thirty-seven experiments were attempted. Each experiment was composed of thirty-six to eighty blocks divided equally into four groups; one group received no treatment; a second received either a direct inoculation of the oak wilt fungus or was soaked in metabolic products of the oak wilt fungus; the third group was inoculated with an associated fungus; the fourth group received a combined oak wilt fungus (or products) and associated fungus treatment* Of the thirty-seven experiments, only twenty-two were completed, the remainder being unsuccessful usually because of contamination or failure of the oak wilt fungus to grow on the wood. These twenty-two completed experiments contained 1152 blocks* Interaction between the oak wilt fungus or its products and the associated wood-rotting fungus could have been detected in only fifteen of the twenty-two experiments* In some experiments no interaction could have been revealed because a non-rotting fungus was used, in other experiments because the length of the experiment was too great or too little* However, the results of twenty-two experiments are summarized in table 1, since the effect of the oak wilt fungus itself could be shown in all but one of these non-interaction-revealing experiments* The Effect of the Oak Wilt Fungus In twelve experiments the oak wilt fungus grew directly on the blocks (methods I, H , and V) • Only in one experiment, no* 9 (fig* 4) Fig. 4. H u n bending-strengths of groups of blocks from an experi an from blocks of groups of bending-strengths n u H 4. Fig. et nwihtefniwr ildb autoclaving. by killed were fungi the which in ment MEAN STRENGTH - MACHINE UNITS OTO ENOO CONTROL -17- DO ND E 13 * !Iean l I i » w h i t e 9 12 119.2 7 4 . G 80.5 69.4 75.5 (one way: F - 2.05) 4.26 4 bla ck 19 io 8 8 . 7 70 .O 70.0 7.6- 10.2- _ 420.8 — 7.39 5 red 19 io 72.5 82.2 7 9.7 6.9- 6.2- 726.1 7.39 6 c he st n ut 28 IO 149.5 91.2 89.4 61.4 59.8 _ 59.2 7.39 T scarlet 34 IO 1 2 . 4 68.4 69.3 69.3 66.7 (one w a y : F — 1.2 2) 4.38 n 8 w h i t e 13 1C 80.6 90.7 85.6 92.6 90.5 (one w a y : F — 1.11) 4.38 9 c he s t n u t 13 IO 63.5 104.8 92.6 99.9 94.4 1 2. 34 - 1.77 7.39 i n 11 black 38 IO 3 8 . 6 87. 6 8 3. 5 88. 0 90.9 3.95 3.18 7.39 13 w h i t e 19 IO 75.7 94.6 94.3 83 .O 92.4 2.73 6.02 3.10 7.39 XV 15 r e d 19 9 102.9 70.2 72.6 42.2 36.6 8.86 1.52 7.50 20 s ca r l e t 47 IO 93. 6 52.4 51.9 35.6 37.8 — 25.5 _ 7.39 2 8 b la ck 9 20 30.5 61.5 66.6 36. 8 33.8 1.24 554.3 9.34 7 .0 1 29 b l a c k 47 20 7 2 . 1 91.4 9 1 . 4 72.1 74.7 89.8 _ 7 .01 30 white 47 19 133.3 82.2 80.2 4 8 . a 57.3 1.64 114 .5 4.23 7.0 1 3 1 w h i t e 9 18 35.8 95.8 98.3 65.2 65.2 — 51.0 — 7 . 0 4 32 r e d 9 18 67.8 70.4 70.1 47.6 45.1 — 151.9 _ 7 .0 4 33 scarlet 47 20 98.2 85.9 83.3 60.3 67.4 - 8.81 5.80 7.01 V 18 b l a c k 9 9 36.2 47.6 50.3 25. 8 2C .4 _ 164.7 3.87 7.50 21 b l a c k 47 IO 188.1 76.9 78.2 66.1 S O.4 _ 10.87 _ 7.39 25 b l a c k 19 IO 128.3 73.1 67.3 56.8 43.8 6.89 30.87 1 . C l 7 .39 26 b l a c k 47 15 15.6 71.8 69.8 40.5 37.6 5.85 2 69.7 — 7.12 2 7 b la ck 9 18 66 .2 88.8 88.3 65.0 60.5 1 . 7 4 181.1 1.06 7.04 * Blocks were exposed to attack of associated Fungus For only six weeks. - “ an F value oF less than l.OG. 0 0 1 Fow — variance ratio relative to the effect oF the oak wilt Fungus or its products on the strength oF oak blocks. Fa — variance ratio relative to the effect oF some Fungi Found associated with the oak wilt Fungus on the strength oF oak blocks. Fov + a — variance ratio relative to the eFfect oF the oak wilt Fungus combined with an associated Fungus on the strength oF oak blocks. Any F value larger than, the F q ^ value For a given experiment indicates significance at the 1 percent level. Error fie an Square « an estimate oF the variance oF strength values oF a samp] e o F oak blocks. TABLE 1. Summary oF Block Testing Data. -19- was there aiy significant indication (at the 1 per cent level) that this fungus might have a direct wood-destroying action; however this was the one experiment in which formaldehyde was not used as the killing agent, autoclaving being substituted instead. As implied above, it cannot be ascertained whether this wood-weakening effect was caused Ty the oak wilt fungus itself or Ty some effect of auto claving such as differential hydrolysis possibly caused Ty a pH difference » With the exception of this experiment the oak wilt fungus caused no detectable decjy even under conditions which permitted a true wood- rotting fungus to decrease the static-bending strength of the wood Ty more than 90 percent, (exps. 4 & 5). In addition, other workers (Beckman, ejt al. 1953) have found little ability on the part of the oak wilt fungus to utilize cellulose. Therefore, the evidence avail able at this time indicates that the oak wilt fungus is not capable of weakening oak wood Ty any direct action. The Effect of Pre-Treatment of Oak Blocks with the Oak Wilt Fungus or Its Metabolic Products upon the Bate of Destruction Ty Associated Fungi. In thirteen of the fifteen experiments capable of showing inter action between the oak wilt fungus (or its products) and the associated fungus there was no indication of its occurrence. However, interaction appears to have taken place in experiment 25 (fig. 6) since the difference between the means of the Isolate 19 treatment and the oak wilt fUngus plus Isolate 19 treatment is about twelve machine units. This difference is more than would be expected -20- 90 CO 80' CONTROL ENDO ENDO + 19 TREATMENTS Fig. 6. Mean bending-strengths of groups of oak blocks where pre- treatrasnt with the oak wilt fun-us appears to have caused accelerated decty an associated wood-rotting fungus. -21- as pre-treatment varianoe. However9 multiple analysis of variance of the four means and a t-test of these two means did not confirm this hypothesis. It was felt that this experiment should be repeated in view of the unexpectedly large difference but five attempts to repeat it using wood from two different black oak trees failed. experiment no. 28 (fig. 7) resembled other experiments upon cursory examination but analysis of variance indicated a significant interaction at the 1 percent level although the disparity between the means of the Isolate 9 treatment and the oak wilt fungus products plus Isolate 9 treatment was only three machine units, ordinarily well within the range of experimental variation. A possible explanation of this result lies in the method of analysis. When testing, y means of multiple analysis of variance, for interaction, the separate effect of each treatment sust be statis tically estimated on the basis of experimental results. These effects are added; then an allowance is made for variability. Next, the sum of the effects plus variability is compared to the experimentally es timated effect of the two treatments acting together. If the combined treatment causes a significantly greater effect at the chosen level than the sum of the separate treatments plus variability allowance, interaction is assumed to have occured. In this particular case, the mean strength of the oak-wilt-fungus- products-treated blocks was greater than that of the blocks not so treated. In analysis of varianoe this must be assumed to be a real effect unless information to the contrary is available. Such an -22- ENDOENDO TREATMENTS Fig* 7, Mean bonding-strengths of groups of o&k blocks* >fultiple analysis of varianoe indicated interaction between the oak wilt fungus products and Isolate 9, but a t-test of the two groups on the right showed they were not different* -23- assumption applied her* leads to the inference that oak wilt fungus products not only affected the strength of the wood, but that thqy affected it in a positive manneri This in turn would make a — n negative difference appear larger than it actually is, with the result that interaction is indicated at lie 1 percent level. If there is evidence that the apparent strengthening effect of the oak wilt ftmgus products is not an effect at all, but merely the result of sampling error, its influence can be eliminated ty comparing only the associated fungus treatment with the associated fungus plus oak wilt fungus products treatment* Such evidence is apparent in table 1 (all exps* listed under me thods III and IT), and permits such action* Further it seems highly unlikely that the action of ai$r organism, except a tree, could increase the strength of wood ty means of axy metabolic products* A t-test for cony arisen of the means of two groups reveals no significant difference at the 5 percent level between the means of the associated fungus treatment and the associated fungus plus oak wilt fungus products treatment, so the apparent interaction seems to have been due to sampling error* TABLE 2* t-test of experiment no* 28* Mean Value of Treatment Strength t Associated Ftmgus 36*8 • 2618 Associated Fungus + Oak Wilt Fungus Products 33*8 t must exceed 2*74 in order for the difference between means to be "" significant at the 1 percent level or 2*030 to be significant at the 5 peroent level* - 24- TVS DURABILITY OP OAK WOOD FROM OAK WILT-KILLED TKEES HATBBTAI5 AND METHODS: The aethode described ty S chef far, et al» (1949) were used in this portion of the investigation with sons Modification* Therefore, the discussion of Materials and Methods will be similar to the dis cussion of the sirn topic in their paper* Source and Kinds of Test Trees The species of oak included in the test were: Red Oak Group: Northern Red Oak lercus borealis Michx Scarlet Oak ooccinea Ruench* Black Oak velutina Lam. White Oak Group: White Oak £• alba L The trees were located in or near Pike and Scioto Trail State Forests in southern Ohio and in the Flack Woods in Knox County, Ohio* Trees of various sixes were used* Samples were collected during the stumers of 1952 and 1953 from twenty-six trees exhibiting typical oak wilt symptoms, and from twenty-six comparable trees not showing ax$r symptoms of disease* For exaaple, whenever a set of samples was taken from a diseased oak tree, a set of samples was also taken froa a healtty oak tree of the same species with a disaster breast high which deviated no norm than 1 inch from that of the diseased tree* In addi tion, this healtty tree was selected froa a site of similar slope, drainage, exposure and biotic surroundings* An effort was made to find the nearest healtty tree which fit these qualifications, although saaples were never oolleoted from a heal tty tree nearer than 50 feet to the diseased tree, since there is a possibility of root grafting -25- within that distance* The relative numbers of species sampled reflects the proportions of species having oak wilt in these areas* Test Specimens Increment borer c o n s were collected and treated ty procedures similar to those outlined ty Scheffer, et el* Six cores were taken from each tree breast high fro® six points approximately equidistant fro® each other around the trunk* The sapwood was identified ty means of a coloronetrie pH test according to Forest Products Laboratory Tech nical Note 253, and was removed, the outer two inches of heartvood being retained for testing* Testing Procedure Two cores from opposite sides of each tree were subjected to rotting ty each of three ftangi: Poria monticola Hnrr, (Madison 698); Steremn frustulosum Fr* (R*P* 56461-R) ; and Polyporus versicolor Fr* (Hash* 72074, or Madison 697); which were a brown rot, white pocket rot and white rot fungus respectively* The cores were partially sterilized ty steam heating the air-dry cores in air tight bottles at 100* C* for twenty minutes and were then placed in 6 ox* prescription bottles which contained 20 ml* malt agar oovered ty mats of the fungi listed above* Y-shaped glass rods were used to support the cores which came from the first seven diseased and seven healtty trees but were omitted for the remainder* Dr* Scheffer in a personal oomaunication stated that the limiting factor, exoessive moisture resulting in deficient oxygen was probably not of importance when '11 samples such as cores were used, an inference borne out ty -26- the remainder of the cores tested* The amount of decay was measured ty percentage loss in dry weight* The cores were weighed after allowing them to equilibrate to certain mild conditions of temperature and humidity both before and after a 4 month period of fUngus infection. In some instances the conditions were 30 percent relative huaddity and 80* F. and in others 45 percent relative humidity and 75* F. but they were the same for both weighings. RESULTS: Table 3 is a summary of the results. Regardless of the combin ation of fungus and tree species, the weight loss of the samples from healthy trees did not differ ly more than 1.8 percent fromthe weight loss of samples from diseased trees; this is true for oak species of the red oak group. There appeared to be nO tendency for the differ ences to be in one direction. The average weight loss for all cores from diseased trees in the red oak group for all three fungi was 0.1 percent less than that for cores from healthy trees. Although the conclusions which may be drawn from a small sample are very limited, it is felt that a statement should be made regarding the white oak group. The six core samples from the diseased white oak tree were no less decay resistant than the six from the healthy con trol. The heartwood of oak trees of the red oak group which have died or are (tying of oak wilt is apparently just as durable as the heart- wood of trees of the red oak group whdch are healthy. -27- Average Poria Sterna Polyporua s All Fungi Oak Species H D U D H D HD Black Oak 36*9 38.7 5.6 5.3 19.3 19.2 20.8 21.0 (11 dis.) (11 healthy) Scarlet Oak 39*7 38.7 8.0 6.9 39.3 39.0 29.1 28.2 (9 dis*) (9 healthy) Red Oak 35*8 35*7 8.6 7.5 23.9 25.7 22.7 23.1 (5 dis*) (5 healthy) White Oak j -2.2 -1.8 7.9 -0.3 33.0 28.7 12.9 9.0 (1 dis*) (1 healthy) Average Red : 37.7 38.1 7.1 6.3 27.7 27.9 24.2 24.1 Oak Group ; Average ; 36.5 7.2 6.0 27.9 27.9 23.7 23.5 All Oaks : 36.0 H “ healthy D - diseased TABLE 3. Average percent loss in weight of core samples of oak* Two cores from each tree were exposed to each fungus* The figures in parentheses refer to the number of trees sampled* -28- SUHKAiflT 1. The oak wilt fungus displayed no capacity to decay oak wood or weaken it in any direct manner, even under conditions which permitted associated wood-rotting ftongi to decrease the static-bending strength of oak ty amounts ranging up to 90 percent* 2* The rates of decay caused ty fungi isolated from oak wilt-killed trees of black oak blocks upon which the oak wilt fungus had been growing were not significantly different (at the 5 percent or 1 percent level) from the rates of decay of blocks which had no contact with the oak wilt fungus* 3. The rates of decay of white, red, scarlet and black oak blocks which had been soaked under reduced pressure in a filtrate from liquid cultures of the oak wilt fungus were not significantly greater at the 5 percent or 1 percent level than the rates of decay of similar blocks soaked in sterile liquid culture medium (potato extract)* 4* The durability of increment borer samples of twenty-five red, black and scarlet oak trees which had typical oak wilt syaqjtoms was fully as good as similar samples from twenty-five comparable healtty trees* The average weight loss of the two groups differed ty only 0*1 percent* 5* The evidence obtained in this investigation indicates that the oak wilt fungus has no direct or indirect effect upon the rate of decay of oak trees infected ty the oak wilt fungus* -2?- LITERATbEE CITED Baxter, D.V. and L.S. Gill. .1931. Deterioration of chestnut in the southern Appalachians. U.S. Dept. Agr. Tech. Bull. 257: 1-27. Beckman, C.H., J.E. Kuntz and A.J. Riker. 1953. The growth of the oak wilt fungus with various vitamins and carbon and nitrogen sources. Ptytopath. 43: 441-447. Findlay, V.P.K. 1922. A study of Paxilljus panuoides FT. and its effect upon wood. Ann. Appl. Biol. 19: 33i-350. 1940. Studies in the physiology of wood-destroying fUngi. Ann. Bot• n.s. 4: 701-712* Lehnan, K.B. and E. Scheible. 1923. Quantitativ Untersuchung uber Holzzerstorung durch Pilze. Arch. hyg. Berl. 42: 39-108. Nellis, J.C. 1914. Uses for chestnut timber killed ty the bark dis ease. U.S.D.A. Farmers' Bull. 582: 1-24. Prescott, S.C. and C.G. Dunn. 1949. Industrial microbiology. 2nd. Ed. McGraw-Hill. p. 838. Roth, S.R* 1943. Effects of invisible decay on deterioration of un treated oak ties and posts. J. Forestry. 41: 117-121. Scheffer, T.C., G.H. Englerth and C.G. Duncan. 1949. Decay resist ance of seven native oaks. J. Agr. Research. 78: 129-152. Scheffer, T.C. and R.M. Lindgren. 1940. Stains of sapwood and sapwood products and their control. U.S.D.A. Tech. Bull. 714. U.S. Bureau of Plant Industry, Soils, and Agricultural Engineering. 1948. Color teats for differentiating heartwood and sapwood of certain oaks, pines, and douglas-fir. U.S. Forest Prod. Lab. Tech. Note 253. -30- AUTCBIOGRAFHT I, Milliaa H. Brandty was b o m in Groat Falla, Montana, Hty 25, 1927. I rocoivod ty secondary school education in the public schools of the sane city. ty undergraduate training was obtained at Montana State University, from which 1 received the Bachelor of Arts in 1950* From the Ohio State University, 1 reoeived the degree Master of Science in 1951, I acted in the capacity of graduate assistant in general botaty during the school year 1950-51. During the summer 1951 1 was assistant to Mr* Harmon Runnels, specialist in diseases of ornamental plants at the Ohio Agricultural Experiment Station* In 1952 I received an appointment as research assistant alternating with the rank of assistant from the Ohio Agricultural Experiment Station on the oak wilt project. I held this position more than two and one-half years while completing the requirements for the degree Doctor of Philosophy*