A 37qu OF THE HOST-PARASITE RELATIONSHIPS or scam-us mumsmmws MARSHAM AND METHODS or PROPAGATION FOR THE INTRODUCED PARASITE DENDROSOTER moruasmns (NEES) WES‘M. Thesis fer the Degree of M. S. MICHiGAN STATE UNIVERSITY DOUGLAS A. VALEK 1967 A __...., U: LIBRARY v III IIIII II III III IIIII III III III IIII III II IIII III IIIIII w Michsé: 9:313 293 10199 ABSTRACT A STUDY OF THE HOST-PARASITE RELATIONSHIPS 0F SCOLYTUS MULTISTRIATUS MARSHAM AND METHODS OF PROPAGATION FOR TEE INTRODUCED PARASITE DENDROSOTER PROTUBERANS (NEES) WESM. by Douglas A. Valek Body of Abstract Mature parasites and predators of Scolytus multi- striatus were collected from elm logs by using a standard trap-log method. 0f the numerous parasitic forms collected only Entedon leucogramma (Ratz.), Cheiropachus colon (L.), and §pathuis canadensis Ashm. have previously been recorded in the literature as attacking g. multistriatus. It was not determined if the other forms were parasitic on §. multistriatus. The clerid Enoclerus nigripes (Say) was observed to be very numerous at times and may be a significant predator of s. multistriatus. Dissections of adult g. multistriatus were made to identify the species of parasitic nematodes which may affect the beetle. The one species found, tentatively identified as the larva of Parasitaphelenchus oldhami Ruhm, was found in 84 percent of the beetles dissected. As many as 129 nematode larvae were found in one beetle. Larvae were found in the haemocoel of the abdomen near the repro- ductive organs. However, no resulting reduction in size of these organs was noticed and the effect of these nematodes on the beetle is not known. A rearing and release program was initiated with Dendrosoter protuberans (Nees) Wesm., a European parasite of g. multistriatus. One hundred female 2. protuberans were released on September 1, 1965 in East Lansing, Michigan and approximately eight thousand females and one thousand males were released in the same area during the Summer of 1966. Laboratory studies on the biology of Q. protuberans showed that one complete generation could be reared in 24 to 27 days. Beetle larvae were not susceptible to para- sitization ble. protuberans until two weeks after initial beetle infestation. The peak of susceptibility occurred between the fourth and sixth week. The log bark thick- ness was found to have definite influence on the ability of 2. protuberans to parasitize beetle larvae. Larvae under the thickest bark were the least parasitized. Adult mortality of 2. protuberans was significantly reduced by allowing them to feed on honey, cane sugar, cane molasses, 0r raisin extract. A STUDY OF THE HOST-PARASITE RELATIONSHIPS OF SCOLYTUS MULTISTRIATUS MARSHAM AND METHODS OF PROPAGATION FOR THE INTRODUCED PARASITE DENDROSOTER PROTUBERANS (NEES) WESM. By Douglas A. Valek A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for a degree of MASTER OF SCIENCE Department of Entomology 1967 ACKNOWLEDGEMENTS I wish to express my appreciation to my major professor, Dr. James W. Butcher for the Opportunity to work on this project. His encouragement and enthusiasm will always be remembered. Special thanks go to Dr. Dean L. Haynes, who became my major professor in Dr. Butcher's absence, for advice and help in the statistical analysis of data. My sincere gratitude goes to Drs. Gordon E. Guyer, James E. Bath, Orlo Jantz, and Victor J. Rudolph for serving on my guidance committee. I am also grateful to Dr. John Knierim and Mrs. Natalie Knobloch for their help, advice, and facilities concerning hematological techniques. I wish to thank the United States Forest Service for granting financial support for this study and to Dr. W. H. Anderson and his colleagues, Insect Identification and Parasite Introduction Research Branch, United States Department of Agriculture for the identification of insects. Appreciation is extended to Mr. James G. Truchan for assistance in the statistical analysis of data. ii TABLE OF CONTENTS PAGE LIST OF TABLES........................................ iv LIST OF FIGURES....................................... v LIST OF APPENDICES.................................... vi INTRODUCTION.......................................... 1 REVIEW OF LITERATURE.................................. 4 METHODS AND'EQUIPMENT....................:............ 9 RESULTS Parasites and Predators of Scontus multistriatus.................................... 23 EndoparaSitfc NematOdeSOOOOOOOOOOOOOOCOOOOOOOOOOOOO 31 Miscellaneous Observations on the - Biology of Dendrosoter rotuberans............... 33 Influence of Host Size ans Bark THickness _ on the Parasites Capacity to Reproduce........... 36 Mass Rearing and Field Release of Dendrosbter protuberans.......................... 46 LongeVify OTAdult Parafites....O..............‘O.’ [l7 DISCUSSIONAND SUMMARYOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 51 LITERATURE CITEDOOOOOOOOOOOOO0.0.000...0......00...... 56 APPENDICESOOOOOOOOOOOOOOOOO0......OOOOOOOOOOOOOOOOOOO. 59 iii LIST OF TABLES PAGE Species and number of individuals COIIeCted in trap-logsOOOOOOOOOOOOOOOOOOOOOOO 24 The number of nematodes from naturally and artificially reared bark beetles............. 32 Reproduction achievement of 2. protuberans under varying conditions of bark thickness, host density, and host maturity..................................... 37 Source of variation from analysis of variance of feeding data from three replications of female 2. protuberans fed diets of honey, sugar, molasses, raisin extract and water..................... 48 iv LIST OF FIGURES FIGURE 1. 2. 30 10. 11. Fiber emergence drum used in collecting inSeCtS from trap-logsoooooooooooooooooooooo Method of log bark thickness measurement...... Small lightproof emergence cages for the collection of bark beetles and parasites from IOgSOOOOOOOOOO0.OOOOOOOOOOOOOOOOOOOOOOO Feeding cages constructed from ice cream containerSOOOOOOOOOO..OOOOOOOOOOOOOOOOOOOOOO The collection record of Entedon leucogramma.. The emergence rate of Scolytus multistriatus and Entedon 1eucogramma..................... The relationship between host development and amount of parasitization by‘ Dendrosoter protuberans..................... The relationship between bark thickness and the number of Dendrosoter protuberans which emergedTFEm each log.................................... The relationship between bark thickness and the number of Scolytus multistriatus larvae per IOgOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO The relationship between the number of Scolytus multistriatus larvae per log and the number of Dendrosoter protuberans which emerged per log........... The relationship between diet and rate of death of Dendrosoter protuberans femaleSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO PAGE 11 15 17 21 27 29 38 41 42 43 49 LIST OF APPENDICES APPENDIX PAGE I. The collection record of all insects collected from trap-logs.................. 59 II. The feeding and longevity of Dendrosoter protuberans eXPer—iment data-O...OOOOOOOOOOOOOOOOOOOO... 7O vi INTRODUCTION The smaller European elm bark beetle, Scolytus multi- striatus Marsham is the principal vector of Ceratocystis 2121 (Buism.) C. Moreau, the causal fungus of the Dutch elm disease. The Dutch elm disease has spread throughout most of the range of native elms in Eastern North America. This has produced large numbers of diseased elm and correspond- ingly high populations of elm bark beetles in these weakened trees. Chemical control has been attempted in most cities affected by the disease, but has not proven entirely successful. Even when spraying is accompanied by sani- tation programs to remove breeding material, the number of bark beetles generally remains high due to immigration from rural areas. The prohibitive cost of spraying extensive rural areas and the undesirable effects which accompany the use of chemicals prompted the exploration of biological control as a possible means of reducing the number of S. multistriatus. In the early Spring of 1965, through the cooperation of the European Parasite Laboratory, Agricultural Research Service, United States Department of Agriculture, the braconid Dendrosoter protuberans (Nees) Wesm. was obtained from Nanterre, Seine, France for rearing and release. This insect is an important parasite of S. multistriatus and other bark beetles in various areas of Europe. If 2. pro- 2 tuberans could be established in this country, it was hoped that it might reduce the smaller European elm bark beetle population to a more tolerable level. The choice of Q. protuberans for release, rather than some other species, was made on the basis of its avail- ability and its performance as a parasite in Europe. If in time this parasite fails to establish itself or other factors inhibit its effectiveness as a biological control agent, other parasites will probably be imported. However, there is much to recommend the importation and release of a single species at a time until an effective biological control is established rather than the release of several parasites which might compete with each other and never produce the desired pest suppression. The initial stock of 2. protuberans were used for laboratory rearing. This insured fairly high numbers for release and additional numbers for observing their habits and biology. Before an effective rearing program could be initiated it was necessary to determine the Optimum degree of develOp- ment of the S. multistriatus larvae for maximum parasite production. 2. protuberans oviposit on host larvae through the bark. Therefore bark thickness was an important con- sideration in the production of parasites. Adult longevity and fecundity were also of concern. It was known from previous experience that 2'.E£2' tuberans could be reared through several generations without adult feeding. Rearing which took place during the spring 3 and summer of 1966 was done entirely without adult feeding. However, an experiment was conducted to find if feeding would produce any difference in longevity of the adult para- sites. No attempt was made to ascertain the effect of diet on the fecundity of the females. The first release 0f.2' protuberans was made from laboratory rearings on September 1, 1965 when one hundred females were released in a woodlot in East Lansing, Michigan. In May, June, and July of 1966 approximately five hundred parasites (mostly females) were released from similar small scale laboratory rearings. After large scale rearing facilities and techniques had been developed, approximately eight thousand females and one thousand males were released in the period from the end of July to the middle of October, 1966. In addition to the release of S.protuberans, a study was made to determine the indigenous insect and internal nematode parasites which may play a role in the population dynamics of S. multistriatus in Central Michigan. The acquisition of more knowledge on the biotic agents affecting the beetle was desired since no study of a similar nature has been undertaken in the area. REVIEW OF LITERATURE The first step in developing the ancient concept of biological control probably began when early man first noticed that insects fell victim to disease and attack by other animals just as he himself suffered from illness and was preyed upon by wild beasts. Steinhaus (1956), in an interesting history of insect pathology, attributes the first recorded account of a disease of an insect to Aristotle (584-322 B.C.) in a description of the maladies of the honeybee. Steinhaus speculated the diseases of silkworms were probably known more than two thousand years before Christ. The first truly successful use of organisms by man to control an insect pest was the introduction of Rodolia (Vedalia) cardinalis (Mulsant) into California in 1888 to control the cottony-cushion scale, Icerya purchasi (Maskell). This success removed all doubt that biological control was a valid method of pest control (Doutt, 1958). Many entomophagous insects have been introduced into North America to control forest pests. Clausen (1956) summarized the biological control attempts undertaken in this country through 1950. Dowden (1962) brought part of Clausen's list up to date by summarizing the biological control attempts against forest insect pests in the United States through 1960. 5 Turnbull and Chant (1961) reviewed the biological con- trol work that had been done in Canada. This same area was more completely and extensively reviewed by McLeod, McGugan and Coppel (1962). There have been only two attempts to control bark beetles in North America by biological means. Hopkins, in 1892-93, introduced the clerid Thanasimus formicarius (L.) into the forests of West Virginia as a predator of the Southern pine beetle Dendroctonus frontalis Zimm. (HOpkins,I 1897). No recovery of this clerid beetle has been made. This was the first known attempt to introduce a natural enemy of any forest insect to this country. The second attempt, was the release of Rhizophagus sp. and Rhopalicus tutela (Walk.) in Quebec (1955-1934) to control the Eastern spruce beetle, Dendroctonus piceaperda Hopk. (McLeod, 23.21., 1962). Again there is no indication that there was any survival of these parasites. There is no known record of a biological control pro- gram to control Scolytus multistriatus; however, there has been some research on the natural control of the beetle. Doane (1959) found that the fungus Beauveria bassiana (Belsamo) Verillemin killed 97 percent of S. multistriatus larvae in an epizootic he observed and suggested that the spores may be dispersed through the larval galleries by small insects. Pesson g: El° (1955) described two new bacteria (Aerobacter scolyti Pesson 23.2l- and Escherichia klebsiellaeformis Pesson gt_al.) which would infect the 6 larvae of S. multistriatus. Doane (1960), working with A. scolyti, S. klebsiellaeformis and Serratis marcescens Bizio, found that these bacteria were transmitted from larva to larva through bite wounds in the skin inflicted by the continuous chewing motion of the mandibles of nearby larvae. Crowding increased the number of bites and greatly enhanced the ability of the disease organisms to Spread. Many species of nematodes have been found in association with bark beetles, but most are strictly free-living forms. However some parasitic forms have been observed. Oldham (1930) observed the larval stage of Parasitaphelenchus oldhami Ruhm’in the body of S. multistriatus in England. Later Saunders and Norris (1961) found that in Wisconsin, ‘2. oldhami was the only parasitic species of the twelve nematode genera associated with S. multistriatus. Another nematode species, Cylindrocorpus erectus Massey, was found in association with S. multistriatus and believed by Massey (1960b) to be parasitic. The exact nature of the effects the nematodes exert on S. multistriatus has not been demonstrated. Oldham (1930) thought that there was a reduction in the size of gonads in beetles infected with g. oldhami and therefore induced sterility. I have not been able to verify his conclusion. Nematode parasitization of other bark beetle species has resulted in reduced egg production and shorter egg galleries. Massey (1960a) observed these symptoms when Aphelenchulus elongatus Massey parasitized the California five-spined engraver, Ips confusus (Lec.). Nickle (1963) found that 7 '1. confusus parasitized by Contortylenchus elongatus (Massey) had fewer fat cells than unparasitized beetles. He also observed that an unidentified internal nematode of the sub— genus Sulphuretylenchus Ruhm reduced the size of the gonads to the extent that some beetles were sterile. Chapman (1911) was the first author to consider the European parasites of S, multistriatus in North American literature. No North American parasites were known at the time. Included in Chapman's list was Entedon leucogramma (Ratz.) which has since been introduced to the United States (probably unintentionally) and now is recognized as an important domestic parasite of S. multistriatus. Pechuman (1937), published an annotated list of insects collected from elm in New York State. He speculated that Spathius canadensis Ashmead and Cheiropachus colon (L.) were para- sites of S. multistriatus. Hoffman (1942) reviewed the literature and compiled an annotated list of elm insects found on elm in the United States. In addition to those parasites above, Hoffman added Eupelmus cyaniceps var. amicus Gir., and Eubadizon magdali (Cress.). Hoffman did not indicate the sources of these additions. Burks (1959) indicated that Trigonura elegans (Provancher) and T. ulmi Burks may also be parasites of S. multistriatus. More recently Bushing (1965) reviewed the North American literature in a synoptic list of parasites of North American Scolytidae. Bushing's list contains only those suspected parasites of S. multistriatus which appear above. 8 The braconid, Dendrosoter protuberans (Nees) Wesm. is a European parasite of Scolytus multistriatus (Dalla Torre, 1898 and‘Chapman, 1911). Its geographic range appears to include France, Italy (including Sicily), Russia, and Hungary (Dalla Torre, 1898 and Russo, 1938). The morphology of the male, female, egg, larva, and pupa of S. protuberans was described by Russo (1938). Russo described the biology of S. protuberans as it exists in the warm climates of Italy where it is a parasite of species of beetles attacking olive trees. METHODS AND EQUIPMENT TRAP LOGS Parasites and predators of S. multistriatus were collected by using trap-logs. Fifteen logs from American elm (Ulmus americana L.) were cut weekly from June 9 through September 21, 1965 and exposed to natural infesta- tion by S. multistriatus and associated insects. A11 logs were 30 inches long and 2 to 8 inches in diameter. Each log was exposed in the field on the same day as the tree was cut. The logs were exposed in an East Lansing woodlot which had a large population of S. multistriatus and many elms dying from Dutch elm disease. Martin (1936) showed that the best location for opti- mum attack by S. multistriatus was in an area of partial shade. Using this criterion as a guide, the trap-logs were placed along the edge of a road through the woodlot just under the overhanging branches. The only direct sun- light on the logs was in early morning. The trap-logs were held approximately 4 to 6 inches above the ground on base logs. The fifteen trap-logs cut weekly were divided into three groups and exposed to attacks as follows: Group I 5 logs exposed for 1 week, then caged Group II 5 logs exposed for 3 weeks, then caged Group III 5 logs exposed for 5 weeks, then caged Logs were group assigned on the day of cutting but in such a way as to avoid putting all large or small diameter 9 10 logs together. Two to three trees were used each week and evenly distributed to log groups. This method was used to reduce the effect of a preference that the insects may have for logs of a certain size or tree source. After field exposure, each five log group was brought into an insectary and placed into a 41 gallon fiber drum for emergence (Figure 1.). Insects showing photopositive behavior were collected in a glass attached to the bottom of the drum. As the insects emerged they were labeled and stored in alcohol. Experience has shows that a large number of insects (particularly S, multistriatus) injured themselves and were not able to fly through the opening if the drum was on its side. Therefore, the trap jars were mounted on the bottom edge of each drum, allowing the in- sects to walk to the opening. Insects from the logs exposed during the first part of the summer (June 9 through July 21, 1965) began to emerge during the late summer and fall of the same year. However, emergence was halted by cold weather before all insects had emerged. These logs were moved indoors during the winter to complete emergence. The emergence from the remaining logs (exposed July 28 through September 21) did not begin in 1965 but took place the following spring and summer in the unheated insectary. 11 or ....... OOOOOOOOO .......... oooooooo r- 0- A u I .I , “ ‘Lf'TT.t " «'fi l §'-*‘r : .fl ' FIGURE 1. FIBER EMERGENCE DRUM USED IN COLLECTING INSECTS FROM TRAP-LOGS l2 DISSECTION 0F ADULT BEETLES Fifty adult beetles were dissected and examined for nematodes. The first twenty-five beetles dissected had emerged from elm logs which were entirely naturally in- fested near East Lansing and had remained exposed outdoors until the beetles began to emerge. The second twenty-five were reared entirely indoors and were the direct progeny of field collected beetles. Dissection procedures were similar to those used by Saunders and Norris (1961). Dissections were made in 2-3 drops of physiological saline solution to prevent dis- tortion of the nematodes. Beetles were alive and often hard to handle during dissection so a drop 0f detergent was added to the saline solution to make the beetles less buoyant. After removing the elytra, the abdominal notum was cut using an ocular scalpel and the haemocael was checked for nematodes. The alimentary tract was extended and examined and the condition of the gonads noted. The thorax and head were also dissected and examined. The location of the nematodes within the body was recorded as they were encountered. The nematodes were removed from the dissecting dish with a dental root canal reamer, placed into 2-3 drops of the physiological saline solution on a glass slide, and killed by gently heating the slide over an alcohol burner. After killing, the nematodes were washed from the slide with 2-3 ml. of double strength F.A.A. into a small test tube. The formula for double strength F.A.A. was: 20 ml. 13 95% ethanol, 6 ml. formalin (40% formaldehyde), 1 m1. glacial acetic acid, and 20 ml. distilled water. After 24 hours in the fixative the excess solution was aspirated from the test tube leaving only enough to cover the nema- todes. The method described by Seinhorst (1959) was used to transfer the nematodes into glycerin. The nematodes were washed from the test tube into a 27 x 8 mm. watch glass with a solution consisting of 20 parts 95% ethanol, 1 part glycerin, and 79 parts distilled water. Enough solution was added to fill the watch glass to the top. The watch glass was placed into a tightly closed glass dessicator containing an excess of 95% ethanol and kept at 40°C. After 24 hours the watch glass was refilled with another solution containing 95 parts of 95% ethanol and 5 parts glycerin and placed into a partially covered petridish. It was held at 40°C for several hours to evaporate the ethanol and then was refilled with the same solution and allowed to evaporate at 40°C. This last step insured the proper coverage by the glycerin. Five to six nematodes were arranged in a small drop of glycerin on a glass slide. Three small lengths of finely drawn glass rods of approximately the same diameter as the nematodes were placed around them. A coverslip was centered over the nematodes and sealed to the slide with ringing compound. 14 BEARING DENDROSOTER PROTUBERANS Host DevelOpment and Bark Thickness This experiment was designed to determine the effect of size of S. multistriatus larvae and influence of bark thickness on rate of parasitization. Logs 4 to 6 inches in diameter and 12 inches long were divided into three replications. Each replicate was com- posed of sixteen logs. Eight logs in each replicate were randomly selected from those which had bark thickness greater than the median thickness of all logs (4.4mm.) and eight from those which had bark thickness less than the median. The experiment was conducted indoors at 75-800F. and 70-90% R.H. After removing a 1/4 inch wide strip of bark from each end of the log, the bark thickness was measured with a clear plastic scale (Figure 2.). Five measurements were taken at one centimeter intervals on each end of the log. Bark thickness per log was an average of ten measurements cal- culated to the nearest tenth of a millimeter. The ends of the logs were dipped into melted paraffin to retard dessi- cation and placed in a single large cage with 600 Scolytus multistriatus adults. Two weeks after introducing the beetles the sixteen logs in the first replication were removed from the cage. These logs were placed into another cage with 32 female and 12 male parasites. After one week of exposure the logs were removed and the female parasites remaining alive were 15 METHOD OF LOG BARK THICKNESS MEASUREMENT 2. FIGURE l6 counted. It was assumed that the dead parasites had died at even intervals through the week so that an average number of live parasites for the week could be calculated. A similar pattern was followed with the remaining logs but at four week and six week intervals after beetle in- festation. It was assumed that the female parasites used in the three groups were equally capable of attacking host larvae. After exposure the logs were enclosed individually in small lightproof emergence cages (Figure 3.). These cages were constructed by nailing a lxlelO inch pine board on each end of the logs. The boards acted as a frame so that the entire structure could be wrapped with heavy alumi- num foil and secured with masking tape. A pint size glass jar was screwed to the bottom board over a 2 inch hole and the cage was positioned on a table with the jar hanging down over the edge. Enough light was provided to attract the emerging S. protuberans and S. multistriatus into the glass jars. The bark beetles and parasites were counted as they were collected in the jars. 17 mwoq Zomm wZHcmMZW mmBHm52 WEB azs mmmZMons mm /- 0. ON on on 00 Ch 901 83:! SNVUBEfllOUd '0 L06 0F 8.NULTISTRIATUS LARVAE 41 2 v - 7.792+(-.544*I x rs-a342 3.0 3. 5 4.0 4.5 5.0 5.5 6.0 6.5 “BK THICKNESS 1m.) ------ 951: CONFIDENCE INTERVALS * SLOPE NOT SIGNIFICANTLY DIFFERENT FROM ZERO AT 5% LEVEL FIGURE 9. THE RELATIONSHIP BETWEEN BARK THICKNESS AND THE NUMBER OF SCOLYTUS MULTISTRIATUS LARVAE PER LOG 42 wow— mmm Qmwmafim EUHE mzmmukz_ wozwoion $00 IIIIII 00... run u<>¢<4 00: 000. 00m 000 02. 000 000 00¢ DOM OON OO. o 0;. a; \\\I\ XONO. +wm~6 u> \\\\\ 0. ON On 0w OO‘I 83d SNVUSEI'IIOUEI 'O 43 ness held constant did not have a significant influence on' parasite success. Therefore host larval density has little or no influence on the lepe of the regression line in Figure 8. However, bark thickness with host density held constant did have a significant influence on parasite success at the 5% probability level. The demonstrated influence of bark thickness on para- sitization was not unexpected. Since host density, at the levels used in the experiment, had little effect on the amount of parasitism, bark thickness probably became the more important influence. In nature, where parasites would have the Opportunity to choose oviposition sites on the basis of both host density and bark thickness, host density may at times become a more important consideration. The partial regression coefficient by x .x equalled -22.2. 2 1 An increase of one millimeter in bark thickness is associated with a decrease of 22.2 parasites. This slope is in the same direction but of greater magnitude than the lepe in Figure 8. Even though the relationship between the number of beetle larvae per log and the number of S. protuberans which emerged was insignificant, it must have had some , TI“; influence. This is indicated by the change produced in slope Of the regression line when its influence is removed. In terms of propagation, both host density and bark thickness should be considered. However no information 44 was obtained on the Optimum ratio of host to adult para- sites. In this experiment no attempt was made to control host densities. If this had been done, the searching efficiency of the parasites would have strongly influenced the results. 45 MASS REARING AND FIELD RELEASE OF DENDROSOTER PROTUBERANS Approximately twenty-six days after introducing the parasites into the rearing chambers, the first generation of S. protuberans began to emerge. These parasites were collected and released daily in various woodlots around East Lansing, Michigan. The mass rearing program with this particular host- parasite system was extremely efficient. ‘2. protuberans emerged continuously in overlapping generations and attacked beetle larvae. Those bark beetles which escaped parasitization, reinfested old logs and the fresh ones which were supplied to maintain a continuous supply of host material. The second generation of parasites was extremely large. Only a slight decline in the rate of emergence marked the transition between the emergence of the first and second generations. Approximately eight thousand females and one thousand males from the rearing chambers were released near East Lansing in the period from the end of July to the middle of October, 1966. 46 LONGEVITY OF ADULT PARASITES Male 2. protuberans used in the first experiment lived significantly longer than the female parasites. The data concerning the females in the first test was combined with that of the second experiment for purposes of analysis in Table 4. This is how the third replication was obtained for the analysis. The original data has been included in the appendix II. In each replication the five insects were ranked accord- ing to the number of days the individual insects lived. This was called the "order of death". A two-way analysis of variance (Li, 1964) was used to test if there was a signifi- cant difference in longevity of the parasites receiving different treatments (foods) and between replications. There was no significant difference between the three replications (Table 4.) despite one replication coming from the first experiment and the next two replications from a different parasite generation. The significant difference between the "order of death" was expected but has no biological meaning. However, there was a significant difference in longevity attributed to the six treatments. 47 TABLE 4 SOURCE OF VARIATION FROM ANALYSIS OF VARIANCE OF FEEDING DATA FROM THREE REPLICATIONS OF FEMALE 2. PROTUBERANS FED DIETS OF HONEY, SUGAR, MOLASSES, RAISIN AND WATER Source df. SS . MSS F Treatments 5 2138.42 427.68 4.70** Replications 2 338.94 169.47 1.86 Order of death 4 4030.29 1007.57 11.07** Error 78 7099.60 91.02 Total 89 13607.25 ** Significant at 1% TO find which of the six treatments were most beneficial in promoting longevity, a "Duncan's New Multiple Range Test" was conducted (Li, 1964). The results of this test indicate that the insects which were fed cane sugar solution were the longest lived. Those insects fed raisin extract, cane molasses and honey were significantly shorter lived than those fed cane sugar, but not significantly different among themselves. Insects fed water and those in the check were significantly shorter lived than the previous group, but not different from each other. All comparisons were made at the 5% probability level. Figure 11. graphically shows the results of the feeding experiment. The plotted points are the averages Of the (DAYS) TIME 50 45 40 35 SO 25 20 I0 48 + SUGAR D RAISIN O HONEY /Mousses 4—3 CHECK fr—wATEa ORDER OF DEATH EACH SYMBOL REPRESENTS THE MEAN TIME OF DEATH OF THREE INSECTS (ONE INSECT FROM EACH REPLICATIONI FIGURE 11. THE RELATIONSHIP BETWEEN DIET AND RATE OF DEATH OF DENDROSSTER PROTUBERANS FEMALES. THE ABSCISSA IS THE ORDER IN WHICH DEATHS OCCURRED. 49 three replications. The curves support the results of the statistical test. The cane sugar diet is superior to all others in maintaining life, followed by raisin, honey, and molasses which are somewhat grouped, and finally the check and water whose curves are nearly identical. DISCUSSION AND SUMMARY The three known insect parasites of S. multistriatus collected were Entedon leucogramma (Ratz.), Spathius canadensis Ashm., and Cheiropachus colon (L.). The eulophid S. leucogramma was by far the most frequently collected. The three predator species collected were the clerids Enoclerus nigripes (Say), Phlogistosternus dislocatus (Say) and Phyllobaenus sp. Large numbers of larvae (probably the common S. nigripes) infested trap-logs in emergence drums during the late Summer and Fall of 1965. These larvae apparently were produced from eggs laid on these logs during the spring and summer. NO clerids were produced from logs exposed during the late summer and fall. There can be no parallel drawn between the number of clerid larvae in the unique environment of the emergence drums and the natural pOpulation. However, from observations of clerid activity in the field, it appears that significant numbers- of bark beetles may be killed by these predators. Other parasitic forms collected were Trigonura ulmi Burks, Cenocoelius saperdae (Ashm.), Atanycolus ulmicola (Vier.), Eubadizon sp. probably rotundiceps (Cresson), Xorides (Xorides) albOpictus (Cresson), and Patasson (Paranaphoidae) sp. These, except T. ulmi (possible para- site) have not been recorded in the literature as parasites 50 51 0f.§° multistriatus. Some of these parasitic forms may later prove to be parasites of S. multistriatus. The one Specimen of Patasson (Paranaphiodea) sp. (Mymaridae) collected may be a misleading clue as to the natural numbers and importance of this egg parasite. The extremely small size Of this insect suggests that perhaps many such small forms were overlooked in the collecting jars. The diverse forms and low densities of most of the in- sects collected results in a major problem in determining the exact relationship between them and S. multistriatus. Some of the parasitic forms may be true parasites of the bark beetle while others may be hyperparasites or parasites of associated boring beetles. It is possible that some parasites are not entirely dependent on a single host species. The schedule used to expose trap-logs in the field was keyed to the known emergence pattern of S. multistriatus. As a result parasitic forms which may have been active prior to or after the time of beetle activity would not have been collected in the trap-logs. Two successive summers with trap-logs containing host material in various stages of develOpment exposed at all times would have been a more desirable method of trapping parasites. The emergence drums used to collect S. multistriatus and parasites should be modified to prevent high humidity from develOping inside. Perhaps slowly circulating air piped into the drums from compressed air tanks via plastic 52 tubes would be helpful. A less complicated solution may be to replace the Mason jars used to collect insects from the drums with screen or cloth containers. Small adult parasites such as S. leucogramma, may be prevented from contaminating the emergence drums, as was suspected in this study, by briefly immersing each log into water. This may wash most of the insects from the scales and fissures of the bark. Twenty-five bark beetles collected from the field were found to be one hundred percent infected with the larvae of the nematode Parasitaphelenchus oldhami Ruhm. Twenty-five bark beetles reared from these field collected beetles were 32 percent infected. This suggests the nematodes transfer efficiently from parents to progeny. The nema- todes were located in the abdominal haemocoel of the bark beetle, closely associated with the reproductive organs. As many as one hundred twenty-nine nematodes were found in a single beetle. However, there was no noticeable reduction in size of the gonads or any positive information on the effects of these nematodes on the beetles. The apparent high incidence of infection leads to considerable doubt as to the effect of the nematodes as a mortality factor in the pOpulation dynamics of the EurOpean elm bark beetle. The beetle is not killed, but more subtle influences such as lower fecundity or vigor may result. 53 S. protuberans was not able to parasitize S. multi- striatus larvae smaller than those available two weeks after beetle infestation. The highest rate of parasitization was achieved between the fourth and sixth week after beetle infestation. These observations were verified by actual rearing experience. While mass rearing S. protuberans previous to the experiment, maximum parasite production was obtained when beetle larvae were exposed four weeks after initial beetle infestation. At this time the S. multistriatus larvae are mature enough to be parasitized but still small enough to allow two generations of para- sites to be produced before the host larvae pupate. Log bark thickness had measurable influence on the ability of S. protuberans to parasitize S. multistriatus larvae. Thick bark proved to be a greater barrier to parasitization than thin bark. The exact nature of the barrier was not found. It may be simply physical and prevent the parasite's ovipositor from reaching host larvae or act as a shield preventing detection. Bark thickness does not pose a great problem in rearing S. protuberans‘ since bark thickness is related to log diameter. Logs larger than eight or ten inches in diameter are usually not used because of the difficulty in handling. Smaller logs will not have bark thick enough to cause great concern. However, the efficiency of S. protuberans as a biological control agent may be severely limited because 54 of the influence of bark thickness. The bark thickness on very large elm boles approaches one inch and may greatly restrict the parasite's ability to attack larvae. More information is needed on this problem. The large scale propagation attempt was considered mildly successful considering the late start in the rearing program. Four generations of S. protuberans emerged from the logs in the rearing chambers while two generations of S. multistriatus were produced. The total release of S. protuberans consisted of one hundred females released on September 1, 1965, approximately five hundred parasites (mostly females) were released in May, June and July of 1966, and approximately eight thousand females and one thousand males were released in the period from the end of July to the middle of October of 1966. In addition, several thousand parasites emerged too late in the fall to be released. A higher rate of production could probably have been attained if prOper food and water had been fed to the para- sites during rearing periods. Feeding parasites diets of honey, cane sugar, cane molasses, and raisin extract more than doubled their life span. Of these foods cane sugar was superior to all others tested. No attempt was made to determine what effect feeding had on fecundity although some increased fecundity could logically be expected. LITERATURE CITED Burks, B. D. 1959. The North American species of Trigonura (Hygenoptera-Chalcididae. Ann. Ent. Soc. Am. 52: " lo Bushing, R. W. 1965. A synOptic list of the parasites of Scolytidae (ColeOptera) in North America North of Mexico. Can. Ent. 97:449-492. Chapman, J. W. 1911. The 1eOpard moth and other insects injurious to shade trees in the vicinity of Boston. Part II. The elm bark beetle. (Eccoptogaster multistriatus Marsh.). Harvard Univ., Contr. Ent. Bussey Inst. 2:30-40. Clausen, C. P. 1956. Biological control of insect pests in the continental United States. U. S. D. A. Tech. Bull. 1139. 151pp. Dalla Torre, C. G. 1898. Catalogue HymenOpterum. Vol. IV: Braconidae. Lipsiae. De Leon, D. 1935. The biology of Coeloides dendroctoni Cushman (HymenOptera-Braconidae)Fan important parasite of the mountain pine beetle (Dendroctonus monticolae Hopk.). Ann. Ent. Soc. Am. 28:411-424. Doane, C. C. 1959. Beauveria bassiana as a pathogen of Scolytus multistriatus. Ann.fiEnt. Soc. Am. 52:109-111. Doane, C. C. ' 1960. Bacterial pathogens Of Scolytus multistriatus Marsham, as related to crowding. J. Insect Pathol. 2:24-29. Doutt, R. L. 1958. Vice, virtue and the vedalia. Bull. Ent. Soc. Am. 4:119-123. Dowden, P. B. 1962. Parasites and predators of forest insects liberated in the United States through 1960. U. S. D. A. Handbook 226:1-70. 55 56 Hoffman, C. H. 1942. Annotated list of elm insects in the United States. U. S. D. A. Misc. Publ. 446. 20pp. Hopkins, A. D. 1897. Control of bark beetles by the introduction of Clerus formicatius. 6th Ann. Report West Virginia Agr. Exp. Sta. Kaston, B. J. 1937. Notes on hymenOpterous parasites of elm insects in Connecticut. SS Britton, W. E. 1937. Conn. St. Entomologist. 36th Report. Bull. Conn. Agric. Exp. Sta. 396:289-415. Li, J. C. R. 1964. Statistical Inference I. Edwards Brothers, Inc. Ann Arbor, Michigan. 658pp. Martin, C. H. 1936. Preliminary report of trap-log studies on elm bark beetles. J. Econ. Ent. 29:297-306. Massey, C. L. 1960a. Nematode parasites and associates of the California five-spined engraver, £23 confusus (Lec.). Proc. Helminthol. Soc. Wash. 27:14-22. Massey, C. L. 1960b. A new species of Nematoda, Cylindrocorpus erectus, associated with S. multistriatungarsH. in American Elm. Proc. HeIminthol. SOc. Wash. 27:42-44. McLeod, S. H., B. M. McGugan, and H. C. COppel. 1962. A review of the biological control attempts against insects and weeds in Canada. Technical Communication No. 2., Commonwealth Agr. Bur. Farnham Royal, Bucks, England. 216pp. Muesebeck, C. F. W., K. V. Krombein, and H. K. Townes. 1951. HymenOptera of America North of Mexico, SynOptic Catalogue. U. S. D. A. Agr. Monogr. 2. 1420pp. Nickle, W. R. 1963. Observation Of the effect of nematodes on IRE. confusus (Le Conte) and other bark beetles. J. Insect Pathol. 5:386-389. Oldham, J. N. 1930. On the infestation of elm bark beetles (Scolytidae) by a nematode, Pargsitylenchus scolyti n. Sp. J. Helminthol. 8(4):239-248. 57 Pechuman, L. L. 1937. An annotated list Of insects found in bark and wood of Ulmus amerigana L. in New York State. Bull. Brooklyn Ent. Soc. 32:8-21. Pesson, P., C. Toumanoff, and C. Hararas. 1955. Etude des epizooties observees bacteriennes dans les elevages d insectes xylophages. Annales De'l I. N. R. A. (Ser. 0) Ann. De Epiphyties 6(4):315-328. Russo, G. 1938. Contributo alla conoscenza dei coleotteri scolitidi fleotribo: Phloeotribus Scarabaeoides (Bern.) Fauv. Portici Lab. Ent. Agr. B01. 2:244-255. Ryan, R. B., and J. A. Rudinsky. 1962. Biology and habits Of the Don las-fir beetle parasite Coeloides brunneri Viereck HymenOptera: Braconidae),71n western Oregon. Can. Ent. 94:748-763. Saunders, S. L.,_and D. M. Norris, Jr. 1961. Nematode parasites and associates of the smaller EurOpean elm bark beetle, Scol tus multistriatus (Marsh.). Ann. Ent. Soc. Am. 54:792-798. Seinhorst, J. W. 1959. A rapid method for the transfer of nematodes from fixative to anhydrous glycerine, Nematologica 4:67-69. Steel, R. G. D., and J. H. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book 00., Inc., New York. 48lpp. Steinhaus, E. A. 1956. Microbial control-the emergence of an idea. Hilgardia 26:107-157. Turnbull, A. L., and D. A. Chant. 1961. Practice and theory of biological control of insects in Canada. Can. J. Zool. 39:697-753. Wallace, P. 1939. Notes on the smaller EurOpean elm bark beetle Scol tus multistriatus Marsham. Conn. Agr. Exp. Bull. 434:293-311. watt, K. E. F. 1963. Mathematical pOpulation models for five agri- cultural crop pests. Mem. Ent. Soc. Can. 32:83-91. A P P E N D I X I THE COLLECTION RECORD OF ALL INSECTS COLLECTED FROM TRAP-LOGS 58 APPENDIX I. 59 THE COLLECTION RECORD OF ALL INSECTS FROM TRAP-LOGS Scolytus mult istriatus Marsham (Scolytidae) Initial date of Date of Number trap-log insect collected** exposure collection Duration of trap-log exposure 1 wk 3 wks 5 wks 7-28-65 6-12-66 23 0 0 6-18-66 0 0 7 6-20-66 0 O 52 6-22-66 26 233 1152 6-25-66 89 1478 3135 6-27-66 345 2628 2128 6-29-66 214 1348 461 7-1-66 198 1136 281 7-7-66 166 1564 202 7-13-66 36 339 21 7-18-66 24 ‘ 78 12 7-29-66 17 132 ll 8-9-66 80 144 16 8-4-65 6-20-66 37 0 0 6-22-66 885 333 60 6-25-66 1610 1506 621 6-27-66 1650 1543 1345 6-29-66 843 912 969 7-1-66 1282 1280 0 7-7-66 1321 1588 1412 7-13-66 582 142 304 7-18-66 396 156 272 7-29-66 473 207 382 8-9-66 66 76 251 8-11-65 6-18-66 16 0 0 6-20-66 56 0 0 6-22-66 417 130 174 6-25-66 973 462 646 6-27-66 595 1107 541 6-29-66 582 533 410 ** No insects collected on dates not listed 60 APPENDIX I. Continued Scolytus multistriatus Marsham Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks 8-11-65 7-1-66 643 692 591 7-7-66 1377 955 749 7-13-66 178 ' 332 0 7-18-66 376 401 254 7-29—66 40 524 0 8—9-66 106 352 0 8-18-65 6-25-66 0 2 35 6-27-66 0 0 98 6-29-66 0 11 176 7-1-66 0 13 598 7-7-66 0 142 752 7-13-66 0 19 142 7-18-66 226 76 145 7-29-66 384 187 270 8-9-66 254 131 195 8-18-66 86 0 0 8-25-66 7-7-66 0 40 0 7-13-66 0 47 0 7-18-66 0 26 22 7-29-66 21 20 4 9-1-65 7-18-66 0 17 16 7-29-66 19 34 42 8-9-66 0 44 0 Saperda tridentata Olivier (Cerambycidae) 7-28-65 6-22-66 0 O 2 6-25-66 0 O 3 ** No insects collected on dates not listed APPENDIX I. Continued 61 Physocnemum brevilineum (Say) (Cerambycidae) Initial Date of date of insect Number trap-log collection Collected** exposure Duration of trap-log exposure ' 1 wk 3 wks 5 wks 6—9-65 5-19—66 one specimen (trap-log exposure unknown) Magdalis spp. (Curculionidae) 7-28-65 6-10-66 29 0 0 6-12-66 0 0 3 6-16-66 32 0 0 6-18-66 21 0 0 6-20-66 6 1 1 6-22-66 4 1 1 6—25-66 1 0 0 6-27-66 0 1 0 6-29—66 1 O 0 8-4-65 6-18-66 0 0 1 6-22-66 0 0 1 6-25-66 1 0 2 7-13-66 0 0 1 7-18-66 0 0 1 7-29-66 0 2 0 7-31-66 1 0 0 Entedon leucogramma (Ratzeburg) (Eulophidae) own parasite of §. multistriatus 6-9-64 10—6-65 0 0 1 *5—19—66 0 5 0 6-16-65 8-17-65 0 0 14 9-14-65 0 1 0 9-20-65 0 0 2 10-1—65 0 0 2 *4-29-66 0 0 8 ** Emerged indoors No insects collected on dates not listed APPENDIX I. * 62 Continued Entedon 1euCogramma (Ratzeburg) Initial date of trap—log exposure 6-23-65 6-30-65 7-7-65 Date of insect collection 9-15-65 9-20-65 9-28-65 10-1-65 10-6-65 10-14-65 10-18-65 *4-28-66 9-8-65 9-13-65 9-14-65 9-20—65 9-23-65 9-28-65 10-1—65 10—6-65 10-11—65 10—14-65 10—1s-65 *4—24—66 *4—28-66 *4—29—66 9-14-65 9-20-65 9-23-65 9-28-65 10-1-65 10-6-65 10-11—65 10-14-65 10-18-65 *4-19-66 *4-21-66 *4-24-66 Emerged indoors é OOOOOl-‘Ol-‘OOOO OOOOOOOOOOOOOO OOOOOOOO Number Collected** Duration of trap-log exposure 3 wks H H OOOQNOHO’WNUIOO \OWHOOOOOOOONWO OUIOL-li-‘ONONQ ** No insects collected on dates not listed 5 wk HOHOONv-P’Q OOOHWQWHN-PWOOH Hra PNONKOCDNIIOUIH 63 APPENDIX I. Continued Entedon leucogramma (Ratzeburg) Initial Date of date of Insect Number trap-log Collection Collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks 7-14-65 9-20-65 0 1 0 9-23-65 0 1 2 10-6-65 0 1 3 10-11-65 0 3 2 10-14-65 0 2 0 10-18-65 0 5 2 *4-24-66 0 0 4 *4-28-66 0 0 6 *4-29-66 0 0 7 7-21-65 10-1-65 0 0 7 10—6-65 3 6 9 10-14-65 0 1 3 10-18-65 1 1 11 *4-16-66 0 0 1 *4-21-66 0 O 12 7-28-65 6-12-66 0 0 5 6-16—66 0 16 49 6-18-66 0 19 81 6-20-66 0 32 42 6-22-66 3 157 205 6-25-66 7 378 174 6-27-66 11 335 131 6-29-66 5 147 55 7-1-66 4 59 10 7-7—66 0 13 10 7-13-66 1 2 3 7-18-66 0 1 1 7-29-66 0 3 2 * Emerged indoors ** No insects collected on dates not listed 64 APPENDIX I. Continued Entedon leucogramma (Ratzeburg) Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks 8-4-65 6-16-66 5 7 0 6-18-66 13 24 3 6-20-66 10 11 0 6-22-66 33 58 56 6-25-66 111 103 147 6-27-66 67 39 157 6-29-66 32 34 131 7-1-66 13 10 31 7-7-66 1? 10 68 7-13-66 1 0 8 7-18-66 2 0 6 7-29-66 1 4 8 8-9-66 0 0 2 8—11-65 6-12-66 14 0 0 6-16-66 25 11 15 6-18-66 62 22 22 6-20-66 32 28 30 6-22-66 87 155 251 6-25-66 135 296 400 6-27-66 175 119 248 6-29—66 28 113 104 7-1-66 9 34 ' 33 7-7-66 8 34 70 7-13-66 2 11 0 7-18-66 0 12 19 7-29—66 5 11 7 8-9-66 0 2 0 8-18-65 6-20-66 0 0 3 6-22-66 0 0 26 6-25-66 0 10 108 6-27-66 0 0 123 6—29-66 0 30 101 7-1-66 0 7 27 7-7-66 0 40 87 7-13-66 0 22 21 7-18-66 0 6 6 7-29—66 0 6 5 8-9-66 0 1 0 ** No insects collected on dates not listed 65 APPENDIX I. Continued Entedon leucogramma (Ratzeburg) Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks ' 5 wks 8-25-65 7-7-66 0 8 0 7—13-66 0 3 0 7-18-66 0 0 8 7—29-66 0 1 1 9-1-65 7-18—66 1 19 0 7-29-66 0 12 1 8-9-66 0 5 0 Spathius canadensis Ashmead (Braconidae) Known parasite of §. multistriatus 6-16.65 8-17-65 0 o 5 6-30-65 9-8—65 0 o 1 9—10—65 0 o 1 9-15-65 0 1 0 9-20-65 0 0 1 7-7-65 9—20-65 0 o 2 10-1—65 0 1 0 10-6-65 0 0 1 10-14—65 0 o 1 10-18—65 0 o 2 7-14-65 10-18-65 0 o 1 7-21-65 *4-16—66 0 o 1 8-18-65 7-13—66 0 o 1 8-9-66 0 0 3 * Emerged indoors ** No insects collected on dates not listed 66 APPENDIX I. Continued Cheiropachus colon (L.) (Pteromalidae) Known ' parasite of §. multistriatus Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks 7-28-65 8-4-65 2* 0 0 8—6-65 1 0 0 Trigonura ulmi Burks (Chalcididae) Possible ‘ parasite of g. multistriatus 7-28-65 7-7-66 6 0 1 7-13-66 1 o 0 7-18-66 6 0 0 Cenocoelius saperdae Ashmead (Braconidae) 6—16-65 *4-21-66 0 0 1 *4—29-66 0 O 1 7-7-65 *4-17-66 0 O 2 *4-19—66 0 0 2 Atanycolus ulmicola (Vier.) (Braconidae) Dates unknown 2 * Emerged indoors ** No insects collected on dates not listed 67 APPENDIX 1. Continued Eubadizon gp, (probably rotundiceps) (Cresson) (Braconidae) exposure unknown Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks 7-28-65 6-10-66 1 O 0 6-12-66 1 0 0 6-25-66 1 O 0 6-27-66 1 O 0 7-29-66 1 o 2 8-9-66 1 0 0 8-4-65 6-22-66 0 0 1 6-25-66 0 0 1 8-11—65 6-22-66 0 0 1 Xorides (Xorides) albopictus (Cresson) (Ichneumonidae) Dates unknown 8 Patasson (paranaphoidea) sp. (Mymaridae) Dates unknown 1 Enoclerus nigripes (Say) (Cleridae) Date unknown 9-20-65 two specimens (trap-log ** No insects collected on dates not listed 68 APPENDIX I. Continued Phlogistosternus dislocatus (Say) (Cleridae) Initial Date of date of insect Number trap-log collection collected** exposure Duration of trap-log exposure 1 wk 3 wks 5 wks Dates unknown one specimen ( trap-log exposure unknown Phyllobaenus sp. (Cleridae) 7-28-65 6-22-66 0 O 1 8-18-65 6-25—66 0 0 1 ** No insects collected on dates not listed APPENDIX 11‘ THE FEEDING AND LONGEVITY OF DENDROSOTER PROTUBERANS EXPERIMENT DATA 69 APPENDIX II. 70 LONGEVITY OF LIFE EXPERIMENT DATA Honey TREATMENTS Sugar Molasses Days of Life Days of Life Days of Life Reps. Reps. Reps. 1 2 3 1 2 3 l 2 3 1 3.5 0.5 3.5 1.0 3.5 4.5 1.0 0.5 3.5 2 5.5 4.5 3.5 1.0 4.5 7.5 16.0 3.5 3.5 3 3.3 4.3 3.3 6.0 3.0 14.3 19.3 3.3 . 3.3 4 6.5 29.0 4.5 16.0 7.5 38.0 19.5 3.5 8.5 5 16.0 38.0 5.0 65.0 18.0 62.0 30.5 10.5 8.5 Rep. Total 37.0‘ 76.5 20.0 89.0 38.5 126.3 86.5 21.5 27.5 Treatment Means 8.9 16.9 900 TREATMENTS Raisin Water Check Days of Life Days of Life Days of Life Reps. Reps. Reps. 1 1.0 0.5 5.5 1.0 0.5 0.5 1.0 0.5 5.5 2 4.0 5.5 5.5 1.0 0.5 0.5 2.5 0.5 5.5 5 10.5 5.5 5.5 2.5 5.5 5.5 2.5 5.5 5.5 4 52.5 8.5 4.5 6.0 5.5 5.5 5.5 5.5 5.5 5 40.0 11.5 56.0 6.0 5.5 5.5 4.5 5.0 4.5 Rep. Total 88.0 27.5 51.0 16.5 11.5 11.5 14.0 13.0 18.5 Treatment Means 11.1 2.6 5.0 HICHIGQN STQTE UNIV. LIBRQRIES HHIIII 31 HIIIIU l||||||||| 1 “HI” "NH! “ll 9 1 2 3 01994378