UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Variation in resistance and virulence in the disease interaction between Melampsora rust (M. occidentalis).. Hsiang, Tom 1984

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1984_A6_7 H75.pdf [ 3.53MB ]
JSON: 1.0075412.json
JSON-LD: 1.0075412+ld.json
RDF/XML (Pretty): 1.0075412.xml
RDF/JSON: 1.0075412+rdf.json
Turtle: 1.0075412+rdf-turtle.txt
N-Triples: 1.0075412+rdf-ntriples.txt
Original Record: 1.0075412 +original-record.json
Full Text

Full Text

VARIATION IN RESISTANCE AND VIRULENCE IN THE DISEASE INTERACTION BETWEEN MELAMPSORA RUST (KL OCCIDENTAL!S) AND BLACK COTTONWOOD (POPULUS TRICHOCARPA) By TOM HSIANG B.Sc, University of British Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of forestry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1984 © Tom Hsiang, 1984 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Forestry The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date: JUNE 22, 1984 Variation in resistance and virulence in the disease interaction between Melampsora rust (M. occidentalis) and black cottonwood (Populus trichocarpa). -ABSTRACT Disease severity, as expressed by spore production rate was compared in a test of fourteen clones of black cottonwood (Populus tr ichocarpa) by ten isolates of Melampsora rust (M. occidentalis), all collected from their natural pathosystem. Spore production rate was measured by average daily production on leaf disks during twice the latent period in days. The overall average uredospore production during the time from inoculation to twice the latent period was 650 spores/disk/day. Latent period ranged from 6 to 12 days with a median at 8 days. Clones as well as isolates differed significantly in their contributions to spore production rates, while there was no indication of specific differential interaction. The lack of qualitative resistance and virulence indicates that qualitative interactions do not play a major role in disease in this natural pathosystem. This finding holds the promise that black cottonwood resistance will not be devastatingly overcome when used in plantations. TABLE OF CONTENTS INTRODUCTION hostrparasite genetics 4 LITERATURE REVIEW qualitative resistance 9 quantitative resistance 10 METHODS sample size 11 host selection and telia collection 12 uredospore inoculum preparation 4 virulence and resistance trials 6 rating system 2RESULTS AND DISCUSSION analysis of variance 25 physiological specialization 28 effects of disease 30 natural pathosystems 2 CONCLUSIONS 36 LITERATURE CITED 7 APPENDIX A: terms used to classify resistance 43 APPENDIX B: listing of all clones and isolates with isolate collection and spore formation dates 44 APPENDIX C: leaf age susceptibility study 45 APPENDIX D: graph: mean average daily spore production of each petri dish vs. inoculum concentration applied to each petri dish 46 APPENDIX E: graph of haemocytometer spore count vs. light absorbance of the same spore suspension 47 APPENDIX F: analysis of variance for the disease severity parameters: average daily spore production, average daily uredia production, total spores, and latent period 48 APPENDIX G: analysis of variance for significant interaction..49 APPENDIX H: calculations for the components of variance 50 i v LIST OF FIGURES FIGURE 1: life cycle of Melampsora occidental is 3 FIGURE 2: a single Melampsora rust isolate was applied to each of the fourteen clones of black cottonwood represented by 17 mm leaf disks in each dish 17 V LIST OF TABLES TABLE I: analysis of variance for average spore production during twice the latent period 26 TABLE II: average daily spore production during twice the latent period for all isolates on all clones 28 TABLE III: components of variance for the sources: isolates, clones, blocks, interaction, and error 30 ACKNOWLEDGEMENTS I sincerely thank Dr. Bart van der Kamp for his guidance and unending patience. Thanks also go to Dr. Jeanette Leach and Dr. Raoul Robinson for their critical examination of this thesis, and to Dr. Clayton Person for the discussions on hostrparasite genetics. This research was conducted while I was under an NSERC fellowship. The photos of aecia and pycnia on page 3 were taken by Richard McBride. 1 INTRODUCTION As a rule, a parasite species has a very limited range of host species on which it can survive; but sometimes a parasite species is so specialized on a host species that only certain parasite races can parasitize a particular host variety. Knowledge of this specificity or physiological specialization may be very important to the plant breeder who seeks to produce varieties with resistance to disease, and to the plant pathologist who wishes to understand the source of disease stability. In agricultural crops, one commonly observed phenomenon is that resistance incorporated into cultivars is overcome within a few seasons through physiological specialization by the pathogen. But, there are cases in which resistance has not been lost or even decreased over decades. In such cases, the loss of resistance through physiological specialization of the parasite apparently cannot or does not occur. Thus, the recognition of the type of resistance (Appendix A) in certain varieties stems from the history of their behaviour through periods of disease. With poplars and forest trees in general, such records are mostly non-existent. This sort of data may be most quickly gathered in the laboratory through artificial inoculations in order to assess the type of disease interaction involved. The purpose of this study is to determine the type and extent of variation in resistance of Populus trichocarpa Torr. & Gray, and in virulence of Melampsora occidentalis Jacks, toward each other, when sampled from their natural pathosystem. Answers to two questions are sought: firstly, is there 2 physiological specialization (variation in virulence or resistance specific to particular host clones or pathogen isolates) such that rust isolates can be distinguished by specific differential reactions on a series of host clones, and vice versa? Secondly, does this degree of physiological specialization play a major role in this wild pathosystem? I selected this pathosystem for two reasons. Firstly, I wanted to study hostrparasite interactions in a natural system where the host had not been under cultivation and put into plantations. Naturally growing black cottonwood ( P.  trichocarpa ) was easily accessible and had the highly visible Melampsora leaf rust. Secondly, I believe that black cottonwood will become an important commercial species in western North America in the near future. Poplars are excellent as short rotation intensive culture plantation trees (Hansen et al. 1983, McNabb et al. 1983), and black cottonwood is the largest of the North American poplars (Fowells 1965). Black Cottonwood belongs to the poplar section Tacamahaca (FAO/IPC 1980) and is easily propagated both sexually and asexually (Muhle-Larsen 1970). Its natural range extends along the Pacific coast from Alaska to northern California and east to Montana. It is mostly found on bottom land, river bars, and forest meadows and streambanks, but it occurs throughout the plateau lands in north-central British Columbia (Fowells 1965). Melampsora occidentalis (Melampsora rust) is a heteroecious eucyclic rust. Its life cycle (figure 1) is as follows: In spring, dikaryotic teliospores, which overwintered on dead 3 PYCNIAL DROPLETS exuded by monokaryotic mycelia from basidiospore Infections contain pycniospores that are transferred between pycnia of different mating types in spring BASIDIOSPORES (X 750) in spring Infect conifer needles TELIAL CRUSTS overwinter on fallen leaves. Teliospores germinate in spring to give basidia which germinate to give monokaryotic basidiospore* TELIA formed by dikaryotic mycelia in early fall A EC IA WITH LOOSE AECIOSPORES dikaryotic aeciospores formed from mycelia after pycnial transfer infect black cottonwood leaves in late spring UREDIA of heavily infected leaf 1n late summe"-UREDIA formed by dikaryotic mycelia from aeciospore infections or later from uredospore re-1nfect1ons produce uredospores throughout summer UREDOSPORE (X 200) BESIDE ST0MATES will form germ tube to penetrate leaf through open stomate on abaxial surface FIGURE 1: Life-cycle of occidentalis (read counter-clockwise) 4 leaves of black cottonwood, undergo karyogamy and meiosis, and then germinate to produce basidia. These basidia produce haploid basidiospores which can infect the needles of Douglas-fir ( Pseudotsuga menziesii (Mirb.) Franco) and other conifers to grow monokaryotic mycelia which form pycnia. After fertilization by pycnial transfer, dikaryotic aecia develop on the same needles, and aeciospores are produced. These dikaryotic aeciospores infect black cottonwood leaves beginning in late spring. The infections give rise to dikaryotic mycelia which produce uredia and uredospores to reinfect black cottonwood leaves. In late fall, the dikaryotic mycelia form dikaryotic telia which overwinter on the fallen black cottonwood leaves. This rust species can be manipulated sexually (pycnial transfer) and asexually (repeating uredospores). (Ziller 1974). Host:Parasite Genetics The terms used in this paper to describe host:parasite relationships are defined here. Resistance is the host component of disease interaction; it represents the ability of a host to hinder its parasite. Virulence is the parasite component of disease interaction; it denotes the ability of a parasite to reproduce on its host. Resistance, virulence, and the disease interaction may appear to vary qualitatively or quantitatively, depending on their genetic basis and the ease with which they may be observed (Scott et al. 1978 p.29). The genetic and epidemiological attributes of qualitative and quantitative interactions also vary greatly by the authority (Appendix A). 5 Qualitative interactions occur in systems where disease interactions fall into distinct non-overlapping classes. These are commonly termed compatible or incompatible, and may represent 100% or 0% disease classes respectively; or in cases where the disease interaction is not an all or nothing situation, the reaction can still be grouped into two classes, for example above 60% and below 40% disease. In any case, the different disease severities should be repeatably and significantly detectable. This is simply illustrated by a table: HOST CLONE + compatible interaction ABC (disease occurs) - incompatible interaction a + - - (little or no disease present) The differential interaction occurs PATHOGEN b - + - here since the clones are differen-ISOLATE tially resistant and the isolates c - - + are differentially virulent. The qualitative interaction forms the basis of recognition of physiological specialization in pathogen races and host varieties. Races are thus distinguished by differential reactions on host varieties as in the preceding table. Qualitative interactions have been conclusively demonstrated to arise from gene-for-gene interactions in a few pathosystems (Sidhu 1980). According to the gene-for-gene relationship between a pathogen race and a host variety as postulated by Flor (1955), a race will be pathogenic to a variety if it possesses virulence genes which correspond to all the resistance genes in the variety. A race will be non-pathogenic or avirulent on a 6 variety if the variety possesses at least one resistance gene against which the race has no corresponding virulence gene. Thus a testing of various specified races against a host variety will give information on the existence of particular genes for resistance in the variety. The other major interaction type is quantitative. Disease severity must be rated or quantitatively measured as it does not fall into distinct classes. The commonly ascribed features of this quantitative interaction are: polygenes, constant ranking, non-specificity, and stability in the interaction (Fleming & Person 1982). By constant ranking, a series of pathogenic races can be ranked for virulence, and this ranking remains consistent on any host variety; as well, host varieties can be ranked and will maintain the same ranking when tested against any pathogen race. It is quite possible to find minor qualitative effects super-imposed over a basic quantitative system. This is well presented in a table by Scott et al. (1978 p.33): VARIETY ISOLATES WYR 69/10 WYR 71/2 Hybrid 46 130 111 Joss Cambier 44 172 Quantitative interaction with reversed ranking, mg spores per 100 square cm leaf produced on seedlings of two wheat varieties by two isolates of Puccinia leaf rust. However the qualitative differential interaction may not always be statistically significant, such as was found in the example above (Scott et al. 1978); and even when statistically significant, the differential interaction may have little 7 epidemiological impact, as host-specificity here gives only a slight advantage to the adapted race. The relative importance of qualitative vs. quantitative effects can be assessed by analyzing the components of variance. It is easy to imagine either possibility, qualitative or quantitative occuring in this Melampsora-Populus pathosystem. If a qualitative system exists, such that cottonwood consists of a number of distinct host varieties which can be defined by their differential resistance to a series of pathogen races, then there would be in effect a natural multiline; each host tree or clonal group would be susceptible to some races but resistant to others. Selection and testing for resistance could be difficult since certain clones might appear resistant simply because the test did not expose them to parasite races to which they were highly susceptible. If it were not known that qualitative interactions could occur in this pathosystem, and if these apparently resistant clones were grown in plantations, then virulent races would likely appear sooner or later and the qualitative resistance would be lost, possibly before the plantation reached maturity. (Kiyosawa 1982 p. 112). On the other hand, the system might be quantitative with typical constant ranking. Experience with agricultural pathosystems has shown that such resistance, though incomplete, is quite stable and long-lasting. (Vanderplank 1982 p.77, Fry 1982 p. 200) . 8 For this study, black cottonwood host clones will be tested against Melampsora pathogen isolates, and analysis of variance plus analysis of the components of variance should show whether the disease interactions differ significantly and whether there are specific differential interactions. But no matter the outcome, it is not possible to conclude that there are no qualitative genes for resistance and virulence in this pathosystem. Possibly only a few trees carry specific powerful resistance genes and only a few pathogen races the corresponding virulence genes. However, if an analysis of variance shows no significant interaction, or if the components of variance show that the isolate-clone interaction term is small relative to the main effects, isolates and clones, then it may be concluded that such genes do not play a major role in this natural pathosystem. The null hypothesis can then be stated as follows: qualitative differential interactions play a major role in this natural M;_ occidentalis - P^ tr ichocarpa pathosystem. LITERATURE REVIEW Many investigators have shown the variability of the Populus-Melampsora complex. The severity of Melampsora rust is a function of a complex of genetic (Heather & Sharma 1977), ontogenetic (Sharma, Heather & Winer 1980), and various environmental factors such as temperature and light intensity (Heather & Chandrashekar 1982). These Australian studies were 9 done with larici-populina on hybrid poplars. Shain (1976) with M^_ medusae on plantation deltoides in eastern North America, and Spiers (1978) with M^ medusae and M_;_ larici- populina on hybrid poplars in New Zealand reached similar conclusions. However, no extensive work has been done with the M. occidentalis - P_j_ tr ichocarpa pathosystem. Jokela (1966), Wilcox & Farmer (1967), Farmer & Wilcox (1968), and Farmer (1970) report that the heritability of rust resistance in the native Populus deltoides Marsh, population toward Melampsora medusae Thuem. is high. However Thielges & Adams (1975) rightly point out that heritability estimates are of no value where dominance between genes occurs. There are reports of both qualitative and quantitative resistance in the Populus-Melampsora complex. Qualitative Resistance Physiologic races in Melampsora larici-populina Klebahn. were first reported by van Vloten (1949). Then Muhle-Larsen (1963) reported that progeny resulting from many crosses of resistant and susceptible phenotypes of deltoides and P. nigra L. showed Mendelian segregation with respect to resistance toward Melampsora leaf rust. Chandrashekar & Heather (1980) demonstrate racial specialization through the reaction of several clones of P.  deltoides to mono-uredospore isolates of M_^ larici-populina . Several pathogen races could be recognized by qualitative distinct reactions on several clones of deltoides . For Melampsora populnea (Pers.) Karst., Gremmen (1980) 10 reports that differential reactions occur for aeciospores from different aecial host species, so that there is at least species specificity if not race specificity. Finally, a 1980 FAO/IPC report states that the results of crossing deltoides with P^ nigra and trichocarpa a few generations removed from their wild ancestors indicate that resistance to Melampsora rust is a dominant character which depends on a relatively small number of genes (FAO/IPC 1980). Quantitative Resistance Thielges & Adams (1975) report much variation in resistance to Melampsora rust among and between families of open grown half-sib seedlings of P_^ deltoides collected from several eastern North American states. Pronounced differences in rust damage and rate of progression were related to geographic origin of the seed parent, thus there was a general constant ranking by the local wild rust population, which is indicative of quantitative resistance. Eldridge et al. (1973) report that poplar cultivars in Australia differed in resistance with the greatest variation in disease severity between provenances and much less within. Again there is a general constant ranking of provenances. Through testing of four M;_ larici-populina isolates against four Poplar cultivars, Heather, Sharma & Miller (1980) suggest that the resistance is polygenically based and caused by minor genes which interact with genes in the pathogen, although no genetic studies were done in this experiment. 11 METHODS There have been no previous studies with this particular pathosystem to search for physiological specialization. Several such studies, however, have been done in other Melampsora-Populus pathosystems, but the differences between this pathosystem and other Melampsora-Populus pathosystems warranted several preliminary studies in methodology. These methodological studies: leaf age effects on susceptibilty, kinetin concentration effects on leaf disk senescence, and hyperparasite effects on uredospores, are presented in this methods section in the pertinent area. Sample Size Unlike many other Host-Pathogen studies, my specimens were to be drawn from wild populations and their disease interaction characteristics were not known. Thus a proper sample size was needed to ensure sufficient sampling: n = (t2) (S2) / E2 S2 = 1.2 (in a disease rating system of 0 to 5, Farmer & Wilcox 1968. Similar value given in Jokela, 1966) E = 1 (set to be significant difference in disease severity which I expect to be able to detect) t = 2.447 (95% level of significance, 6 degrees of freedom) n = (2.447)2 (1.2) / (1) = 7 This calculation was valid for quantitative systems, but as can be seen by the following calculations, the sample size held, 1 2 even conservatively, for qualitative systems. For qualitative systems, the sample size is chosen such that there is a 95% chance of selecting at least two distinct types (clones or isolates). If qualitative effects are important, the most common type should not comprise more than, for example, 50% of the population. The probability of selecting only one type in 'n' trees is .5 to the nth power. Hence, what is 'n' such that .5 to the nth is equal to 0.05 (i.e. 95% confidence)? The sample size 'n' is then found to be 5. Therefore, at least seven trees and seven pathogen races were needed in sampling. As this was the minimum, it was decided that more than seven trees along with their Melampsora rust isolates were to be sampled. Host selection and Telia collection Eleven clones of black cottonwood were selected from around the Lower Fraser Valley in the autumn of 1982 and the spring of 1983. Telia of the autumn collected clones were also gathered at that time and stored outside in open-meshed bags. During the early part of 1983, two clones and their isolates were collected from the B.C. Interior and one from Calgary, Alberta. Listings of all clones and isolates as well as the collection and spore formation dates for all isolates are given in Appendix B. The black cottonwood trees were chosen on the basis of being relatively isolated from other cottonwoods, and having attained reproductive age. Fallen twigs from the tree were usually collected, as reproductive buds and most of the branches 1 3 were in the higher portions of the tree. These abscised twigs due to cladoptosis are known to root readily in greenhouse conditions (Galloway & Worrall 1979). Leaves with telia were also collected from around the base of the tree to be used in virulence trials. Both collections required that the tree be isolated so that the identity of the twigs and leaves could be certain. Isolation was also required so that uredospore cycling should be on the same tree, presumably allowing build-up of a single Melampsora isolate. As isolation was the major criterion for sampling, trees with leaves that had no telia were also selected. The requirement that the trees had attained reproductive age was added so that sexual crosses could be made if the parents were to exhibit qualitative resistance. Since black cottonwoods are mostly dioecious (Schreiner 1974), all crosses were not possible in any case. Following the method of Muhle-Larsen (1970) with twigs possessing reproductive buds set in flasks of distilled water, and with a paint brush to repeatedly dust pollen onto newly emerging female flowers, I did achieve one successsful cross which yielded 13 progeny. However, the parents did not differ significantly in overall resistance, so these progeny were not of interest. Several trees from outside the Lower Fraser Valley were sampled to include geographic variation. Statistical analyses were later done with and without these clones and isolates. The first analysis was to show what type of variation in virulence and resistance may exist in the trichocarpa - M. occidentalis pathosystem at a species level. The second 14 analysis was to show what type of variation may occur in the system at a population level. All collected leaves with telia were kept outside in open-meshed bags until April. Presumably, most of the telia on these leaves belonged to the pathogen race which was most successful on the tree during the past season. These telia were probably the result of auto-infection during the infection cycles within the tree. These telia could not be induced to germinate in the fall, probably because they probably require external overwintering stimuli to germinate in the spring (Longo et al. 1980, Von Weissenberg 1980). The twigs were potted in sterilized soil and grown on long days in the greenhouse in order to produce enough leaves for the resistance trials. The daylength and average temperature in the greenhouse were 16 hrs and 20 degrees C. Greenhouse grown leaves have a low chance of wild infection, and none of my clones became infected in the greenhouse. Uredospore Inoculum Preparation In early April, leaves with telia were brought into the laboratory, washed in cold, running, tap water for one hour, and set on top of moist filter paper in petri dishes to incubate. Upon visible germination of the black telia giving a golden-yellow layer of basidia above the black telia, the leaves were suspended over pots of Douglas-fir seedlings planted in March and April. This procedure follows that of Ziller (1965). The assumption was made that all telia on the leaves of a single 15 cottonwood tree were closely related. The inoculation of Douglas-fir seedlings by these basidiospores should yield pycnia on the upper surface of the Douglas-fir needles, but, unlike Ziller (1965), I did not find inoculation of Douglas-fir seedlings by basidiospores of M.  occidentalis easily achieved. My inoculations yielded pycnia for two isolates, Imp and Pic (Appendix B). Possibly my telial isolates were adapted to an aecial host other than Douglas-fir and thus were not able to cause infection. Inoculation was attempted with freshly washed telia that had not been set to germinate in petri dishes. These inoculations did not yield any pycnia. Attempts were also made to monitor sporefall by sporecounts of vaseline-coated coverslips set in pots of Douglas-fir for the duration of the 3-day inoculation; but contamination and the unevenness of the trapping media prevented accurate counts. The pots of Douglas-fir were kept in insect free cages to avoid spurious fertilization of pycnia. Ziller (1965) has shown that M^ occidentalis is heterothallic and thus it is possible to make controlled sexual crosses by pycnial transfers (Ziller 1959). Single transfers should lead to the formation of aeciospores, half of the time if there are only two mating types. The plan initially was to make only selfing crosses, and then, if the parental isolate lines proved to be of interest during later experimentation, non-selfing crosses would be made. The transfer of droplets of pycnial fluid between pycnia derived from the same telial source yielded selfed progeny in 1 6 the form of aeciospores. This was successful for both isolates with pycnia, Imp and Pic. From selfing crosses between pycnia of the same telial source, some spore lines were isolated. The aeciospores were inoculated onto their respective original host clone. A series of uredospore transfers followed to yield pure spore lines. For all other isolates which could not be selfed due to lack of pycnia, uredia was collected in late summer from the original trees. (The collection dates are presented in Appendix B). The uredospores of several uredia of each leaf were spread onto green-house grown leaves of their respective original host clone. Several uredospore transfers followed before virulence trials. I thought that these uredospores of late summer were likely those of the race which were well adapted towards their particular black cottonwood clones. If physiological specialization were occurring in this natural system, well adapted spores of late summer should be expected to exhibit it. Virulence and Resistance Trials The most recent fully expanded leaf of each clone was collected in the greenhouse. After surface sterilization with .35% sodium hypochlorite for 1 minute, each leaf was cut into ten 17 mm disks with a number 6 cork borer. Disks of one clone were placed into ten different petri dishes on top of number 3 Whatman filter paper saturated with a 5 ppm solution of kinetin. Each petri dish ended up with 14 disks representing the fourteen 17 clones (figure 2); In one experimental block, there were ten petri dishes, one for each isolate. The block was replicated at ten different times leading to ten randomized complete blocks as the experimental design. One block was discarded the eighth day after inoculation due to the unacceptably low infection. The inoculum quality for this block was suspect. Figure 2: A single Melampsora rust isolate was applied to each of the fourteen clones of black cottonwood represented by 17 mm leaf disks in each petri dish. To standardize the effects of leaf age and position on susceptibility to rust, the most recent fully expanded leaf was used. It has been suggested that a recently fully expanded leaf exhibits maximum susceptibility, and that subsequently susceptibility decreases with age (Sharma & Heather & Winer 1980, Lin & Edwards 1974). As well, the most recently expanded 18 leaf may be expected to senesce less quickly than older leaves. A minor experiment was conducted to measure these age effects. Several cottonwood stems were stripped entirely of their leaves for use in this leaf age susceptibility study. Leaf halves were inoculated in petri dishes and incubated for 25 days at which time the disease severity was assessed by number of uredia and telia. The youngest leaves were the most resistant, and did not become diseased. The most recently fully expanded leaf was found to be the youngest or the second youngest leaf susceptible to Melampsora rust. (Appendix C). The host clones were represented by leaf disks cut from leaves surface sterilized with .35% sodium hypochlorite for 1 min (Waller 1981). These disks were floated on filter paper in a 5 ppm kinetin solution in a petri dish, as has been done in other studies of the Melampsora-Populus complex (Shain & Cornelius 1979, Singh & Heather 1981) In a small study, the effect of varying the concentration of kinetin was tested on halves of recently matured leaves. One half of these leaf halves were surface sterilized. The results showed that while there were no significant differences between sterilized and non-sterilized leaves, with the higher concentrations of kinetin (10, 50, or 100 ppm), heavy senescence occurred up to four times faster than with water. There were no great differences between water and up to 5 ppm kinetin. Chandrashekar (1982) found similar results. Inoculations followed the method of Shain & Cornelius (1979). Uredospores were suspended in a .1% agar solution. Three droplets of suspension, each approximately .02 ml, were 19 inoculated with a micropipette onto each disk of one petri dish. Shain and Cornelius (1979) state that the optimum inoculation concentration is between 1250 to 5000 spores/disk (1.7 cm diameter). Indeed, I obtained heavy infection at 1300 spores/disk (1.7 cm diameter) in earlier inoculation tests. Attempts were made to standardize inoculum concentrations to around 3000 spores/disk (three .02 ml drops of 50,000 spores/ml suspension) through a constant spore suspension absorbance reading ( .05 absorbance units or 90% transmittance) on a spectrophotometer prior to inoculation; but the achieved range was 1000 to 20000 with an average at 6000 spores/disk. These spore counts were made on a haemocytometer after the inoculations rather than before, which may have contributed to to this great range in inoculum concentration. The uredospore inoculum load was purposely set higher than the concentration at which heavy infection could occur on some clones. In inoculation curves (infection level vs. spore concentration), it is thought that slope decreases exponentially (concavely curvilinear) and reaches a concentration saturation level (slope goes to 0) where a greater concentration of inoculum does not cause greater infection. My objective was to ensure that inoculum concentrations onto all the clones exceeded their saturation point so infection level should be due purely to resistance and virulence rather than escape effects. As a final check on proper inoculum load, a correlation was later made between inoculum concentration and single petri dish infection averages. (A single spore preparation was applied to a single petri dish). The results (r= -.14) showed no 20 significant correlation between spore concentration and infection levels, which meant that my inoculations were, as proper, in the range of saturation concentrations. As well, a graph of the two showed no apparent trends even among the lowest concentrations. (Appendix D). After each petri dish inoculation, spores were plated onto 5% water agar to determine the percent germination. There was a imperfect fungus commonly associated with the uredospores which may have been inhibiting uredospore germination in the water agar plates. Readings of 0% germination were obtained, although infrequently, yet infection of leaf disks did occur from these same spore suspensions. It was suspected that a hyperparasite which was especially promoted in the water agar petri plate conditions was responsible. An unpublished experiment was done in the same lab to test the effects of this presumed hyperparasite. This experiment was conducted by Elena Klein as a directed studies in the fall of 1983. She showed that although this presumed hyperparasite had a great effect in reducing infection when inoculated onto the leaf a week prior to uredospore inoculation, it had no effects in a 24 hr prior inoculation. However, there was considerable background variation in this experiment. Hyperparasites and antagonists of Melampsora uredospores have been commonly reported (Bier 1965, McBride 1969, Omar & Heather 1979, Sharma & Heather 1981 & 1982), however their role in natural pathosystems is not known. These published studies were all conducted in laboratory conditions, and while 21 antagonists can be promoted in laboratory conditions, antagonistic concentrations may be uncommon in Nature. Finally, the petri dishes of the main experiment were all placed into an incubation chamber at a temperature of 18 degrees C with constant cool white fluorescent light and watered every 3 days with distilled water. Krzan (1980) shows that optimal air temperature for germination of larici-populina uredospores is 16 degrees C, with higher temperatures more inhibiting than lower temperatures. Toole (1967) holds that the temperature optima for the germination of uredospores of medusae and M. larici-populina is 18 degrees C. Shain & Cornelius (1979) demonstrate that leaf disks of deltoides inoculated with M. medusae will have more severe infections at 18 degrees C than at 23 degrees C. Rating System Disease severity may be assessed by several parameters. In classical studies of host:pathogen relationships, infection type with defined categories of infection has been the main screening criterion, but many workers emphasize the importance of epidemiological characteristics such as number of pustules, incubation period, number of spores per pustule, and longevity of sporulation. (Zadoks & Schein 1978 p.7, Sharma & Heather 1979a). Number of spores would be an absolute measure of the virulence/resistance in the aegricorpus (hostrpathogen association, Loegering 1978 p. 311); but, depending on spore production rates, the latent period (time from inoculation to 22 symptoms) could also have important epidemiological consequences. Thus a combination of the two, such as #spores produced / latent period, which is spore production rate, may give a better assessment. Other studies in the Melampsora-Populus complex have used diverse measures for disease severity. In the multitude of publications by Heather and associates at the Australian National University, the measures used are: IPF(incubation period), ULD(uredia per disk), and USM(uredospores per mm2 ). Measurements are made at 15 days after inoculation. Shain and Cornelius (1979) use total uredia and telia counts for disease severity, and measurements are made at 9 to 11 days after inoculation. Spiers (1978) similarly made uredial counts 10 days after inoculation. In all of these studies, the particular reason for the date of measurement is not given. For each of my disks, which were observed daily, measurements were made at a period of twice the latent period so that all disease interactions could be measured at relatively the same stage of development. All uninfected disks were observed up to a minimum of 22 days. The number of spores may be measured indirectly by measuring absorbance of a spore suspension with a spectrophotometer. (Both transmittance and absorbance scales are given on a Spectronic 20 spectrophotometer, but it is easier to read the linear transmittance scale because the absorbance scale is non-linear. After obtaining transmittance readings, one can easily convert them into absorbance units, since absorbance is the log of the inverse of transmittance). Sharma 23 & Heather (1979b) have shown that Melampsora uredospore concentration as measured by a haemocytometer has a high correlation (r=.998) with light absorbance at 640 nm of the same spore suspension. Spore suspensions for readings on a Spectronic 20 were made with modifications of Sharma & Heather's (1979b) procedure. As latent period has a strong relationship with spore production, all disks were assessed upon reaching twice their latent period. Disease assessment took two forms: the first involved microscopic examination of the disks to count the number of uredia and telia. Uredia were rated into one of three diameter classes: <.2mm, .2mm to .5 mm, and >.5mm. The second method involved spore counts for disease severity. The disks were placed singly into test tubes with 1 ml of 5% Tween-20, and the tubes were vigorously agitated for 1 hour one hour on a Burrell wrist-action shaker. One and a half mis of water were then added to each tube to make the 2.5 mis needed for absorbance measurements on a spectrophotometer. The final 2% Tween-20 concentration did not detectably decrease transmittance in a Tween-20 absorbance test. Random checks were carried out on one out of ten tubes with a haemocytometer to derive a correlation for absorbance (colorimeter) and spore count (haemocytometer). Spore suspensions for 154 disks were measured for absorbance and spore count. The derived equation was: SPORE COUNT = 1228300 * ABSORBANCE - 7294, which had a significant r=.9l (Appendix E). This negative y-intercept value indicates that there may have been discoloration 24 of the spore suspensions by the disks. Indeed, over 20 of these suspensions were made from disks with no infection, and these suspensions all showed a small degree of absorbance. Since spore suspensions cannot take telia into account, a value of 500 spores was assigned to each telia to give the final equation: SPORE COUNT = 1228300 * ABSORBANCE - 7294 + 500 * TELIA. This value of 500 spores for each telium was decided upon after calculating that a medium size uredium released this number of spores in the disk shaking process. Sharma & Heather (1979b) derived a similar equation of: UREDOSPORES = 1170000 * ABSORBANCE + 2800 for larici- populina on hybrid poplars. Analysis of variance was then performed on spore production rates as calculated from average spore production during twice the latent period. Analysis of variance was also carried out on total pustule count over twice latent period, on total spore count, and on latent period in days. Components of variance were then calculated for the variance sources used in the analysis of variance on average daily spore production to determine the variance contribution by each source. 25 RESULTS AND DISCUSSION The spore production rates are presented in Table II. Latent period varied from 6 to 12 days with the median at 8 days. Out of 1260 disks, 235 did not become infected, although no single clone-isolate combination was without infection throughout the nine blocks. Analysis of Variance The analyses of variance for all specimens and for only Lower Fraser Valley specimens are presented in Table I. The first analysis was done to show what type of variation in virulence and resistance may exisc in the P_;_ tr ichocarpa - M.  occidentalis pathosystem at a species level. The second analysis was to show what type of variation may occur in the system at a population level. Both analyses showed that there were highly significant differences between clones of black cottonwood and between isolates of Melampsora rust, which meant that there were great differences in clonal resistance and in isolate virulence at both the population and species levels. The clonal means in spore production rate varied from 311 to 1008 spores/disk/day, and the isolate means varied from 283 to 1074 spores/disk/day (Table II). These analyses also showed that there were no significant interactions between clones and isolates. Thus there was no indication of differential interaction. The significant F-value for blocks meant that significant variation was removed by blocking. This variation consisted of random effects, time effects (as blocks were replicated at 26 TABLE I: Analysis of variance for average daily spore production during twice the latent period, and isolate and clonal rankings (***). a. For all specimens SOURCE Block Treatment Isolates Clones Interaction Error Total D. . F. 8 1 39 F-VALUE 6.942 2.357 5.318 14.442 .929 F-PROB ** 9 1 3 1 1 7 .0000 .0000 .0000 .0000 .6884 1112 1259 Clonal ranking: * * * Ken<Pond<Thun<Sal<Poco<Mar<Pic<Hebb<Rup<Imp<Gran<Cord<TrQ<Cal Isolate ranking: * * * IMP <GRAN <KEN <THUN <PIC <PRG <HEBB <ALK <RUP <POCO b. For Lower Fraser Valley Specimens SOURCE D. F. . F-VALUE F-PROB Block 8 5.003 .0000 Treatment 76 1 .885 .0001 Isolates 6 8.600 .0000 Clones 1 0 5.319 .0000 Interaction 60 .870 .7457 Error 608 Total 692 Clonal ranking: Ken<Thun<Sal<Pond<Mar<Poco<Hebb<Rup<Imp<Cord<Gran Isolate ranking: IMP <KEN <GRAN <THUN <HEBB <POCO <RUP * Specimens from outside the Lower Fraser Valley ** F-prob is the probability that the null hypothesis, which states that there are no differences, is true. *** The rankings result from Duncans multiple range tests (95% confidence), and all specimens underscored by the same line are not significantly different. 27 different times), variation between leaves of a single clone, and variation from separate preparations of the same isolate. The lack of significant clone-isolate interaction coupled with significant differences beteen clones and between isolates means that there is constant ranking. From the isolate and clone rankings (Table I), it can be seen that although the absolute rankings differ slightly between the Lower Fraser Valley rankings and the all-specimens rankings, there are no statistically significant changes in ranking. Within one underlined group (Table I), specimens may change order between all-specimens and Fraser Valley specimens, but rankings do not change between groups. Analyses of variance on other parameters of disease severity (average pustule production, total spore count, and latent period) showed the same results (Appendix F). Clones as well as isolates differed significantly, while there was no sign of differential interaction. These analyses were done because, depending on the resistance mechanism involved, it would have been possible for one parameter to demonstrate differential interaction while another didn't. For example, if the resistance mechanism were involved in impeding the penetration process, this could show up most clearly in latent period differences. Whereas if the resistance mechanism were involved in impeding parasitic growth and sporulation after penetration, this would probably show up better in total spore count than in latent period differences. 28 Physiological Specialization Another presentation of the lack of detectable specialization or adaptation by the isolates toward their original host clones is given in Table II. It shows that in no case did an isolate produce more spores on its original host than on all other hosts. A t-test done on the difference in spore production rates between means of all original host-clone combinations and of all experimental combinations showed no significant difference between the two. Thus there was no indication of physiological specialization by pathogen isolates toward their original host clones. TABLE II: Average daily spore production during twice the latent period for all isolates on all clones (units are spores/disk/day). (isolates) Alk Gran Hebb Imp Ken Pic Poco PrG Rup Thun MEAN (clones) CAL 2152 639 2964 1 020 1 346 1 740 3443 1 270 2700 3245 2052 CORD 1 333 813 1466 1 56 367 783 1 362 903 1110 1 669 997 GRAN 377 723* 348 392 1 125 1034 1 937 733 1 767 1089 953 HEBB 1 263 622 358* 316 631 623 672 885 817 470 666 IMP 955 1596 324 1 1 3* 547 560 583 671 1115 723 719 KEN 1 54 353 1 2 1 58 241 * 361 591 490 405 341 31 1 MAR 890 376 953 221 237 433 870 1093 1 1 43 31 1 653 PIC 645 261 804 509 473 1 1 06* 813 773 602 481 655 POCO 678 406 868 89 688 847 644* 779 914 264 618 POND 339 178 181 196 364 456 642 313 456 203 333 RUP 770 316 490 398 328 868 1037 885 1 091 * 657 684 SAL 714 337 524 80 835 292 556 403 528 164 443 THUN 338 37 266 76 425 233 565 747 806 289* 371 TRQ 986 1322 1068 313 1516 987 1323 798 1 159 606 1008 MEAN 828 570 765 283 652 737 1074 767 1 044 751 I 747 *Original clone-isolate combinations 29 In my sampling of isolated trees, there was inherently a bias toward sampling those cottonwoods which had survived singly in their immediate area. They may have had some special characteristics such as resistance genes which allowed them to survive. In addition, isolates were taken from these same sampled trees, so that if physiological specialization were occuring, these isolates should surely demonstrate it. This can be contrasted to total random selection of both clones and isolates where the theoretical corresponding genes may not have been sampled. The sampling procedure thus further strengthens the result, that there are no signs of physiological specialization in this natural pathosystem. Because I did not find differential interactions in the samples need not mean that there are none. Quite possibly differential interactions do occur to a minor extent in this pathosystem. Manipulation of the data by eliminating several clones and isolates lead to a significant F-value for isolate-clone interaction (Appendix G), but this data manipulation may be statistically unsound, and also this interaction may have no biological importance. If differential interactions originating from physiological specialization do occur in this pathosystem, they are very rare. Mathematically, the sample size of 14 randomly selected clones allows me to say with 95% confidence that trees with qualitative resistance will play a part in 20% or less of all disease interactions in this pathosystem (calculations: what is x such that (1-x)1* = .05? x is found to be .20). 30 This is further demonstrated by an analysis of the components of variance: although this experiment contains considerable error, it was still possible to see the strongly significant differences between clones and between isolates (Table III). TABLE III: Components of variance for the sources: isolates, clones, blocks, interaction, and error. isolates clones blocks isolate-clone interaction error percent variance accounted for: all specimens Vancouver only 2.6% 5.7% 11.5% 4.63.5% 5.0% 0.0% 0.082.4% 84.7% If there had been no detectable differences between isolates or between clones, then one could claim that this experiment was not sensitive enough to detect differential clone-isolate interaction either. The relative variance contributions change somewhat for all sources except isolate-clone interaction, which remained at 0% (Table III). Calculations for components of variance are given in Appendix H. Effects of Disease Aside from the hypothesized genetics of disease interaction, this disease may have a profound effect on its host. In an unpublished experiment conducted in this lab by Susan Hruszowy on the effects of M_;_ occidentalis on photosynthesis in black cottonwood, she found that 31 photosynthesis of diseased leaves was reduced significantly by 31% as compared to disease-free leaves on the same plant. This experiment was done in the greenhouse on leaves still attached to the plant. Similar results have been reported in published studies. Uredopustules of some rusts act as sinks to accumulate metabolic products in competition with other plant parts (Wang 1961). Heavy infection by Melampsora rust causes premature senescence of leaves and defoliation (Schipper & Dawson 1974). Moderate rust infection can cause a 46% growth reduction (Widin & Schipper 1976). Defoliation has a great effect on stem height and basal stem diameter and consequently on the volume of wood produced by infected trees (Joly 1959, Widin & Schipper 1976). Donnelly (1974) reports that basal leaves of poplars export photosynthate mainly to the basal stem while higher leaves contribute more toward stem height; thus as basal leaves are most often first infected and abscised, basal stem diameter is greatly affected. However, these figures for growth reduction are extreme and it may have been environmental interactions which allowed for such great severity. Palmberg (1977) found that poplar disease resistance to Melampsora rust was not significantly different between different clones on the same site, but was significantly different between sites. He also found very high genotype X environment interactions. Heather & Chandrashekar (1982) claim that environmental factors are most important in disease stability, and that light and temperature contribute more to variation than do cultivars or races. 32 Natural Pathosystems There have been very few studies done on natural pathosystems. Robinson (1980) states that: "virtually no factual studies of a wild plant pathosystem have ever been undertaken" (p. 204), but does mention that two researchers have been working on a wild plant pathosystem in the Netherlands. Other studies of wild plant pathosystems include Puceinia spp. on Avena spp. in Israel (Wahl et al. 1978, Dinoor 1977), two foliar diseases in a Trifolium repens population in Wales (Burdon 1980), Erysiphe graminis hordei on Hordeum sp. (Wahl et al. 1978), a leaf rust disease on Glycine sp. in Australia (Burdon & Marshall 1981), Puce inia spp. on Avena spp. in Australia (Burdon, Oates & Marshall 1983), and a rust disease of wild sunflower species (Zimmer & Rehder 1976). The problem with several of these studies is that the wild plants grew near their cultivated relatives, so that the selection pressure posed by the nearby uniform cultivars could have altered the pathogen population such that these wild pathosystems may not have been natural. I could find no studies in the literature on natural forest tree pathosystems in this area of pathosystem theory. Even my study is not wholly natural since most of the specimens were selected from near urban settings. However, neither the host black cottonwood nor the pathogen Melampsora rust have been manipulated to any extent in the area, so that the unnatural and powerful selection forces arising from homogenous plantations have not had an effect. 33 Robinson (1979 p.21) speculates that qualitative resistance can evolve in natural discontinuous pathosystems but it need not do so. He also suggests that discontinuity (which refers to the continuity of the hostrparasite interaction) would be the force which creates and maintains qualitative resistance, since continuous systems would have no use for qualitative resistance. However, unless the pathogen poses a great selection pressure on its host, qualitative resistance would not be necessary in the host in whichever system. The M^ occ idental i s - P_j_ tr ichocarpa pathosystem can be considered discontinuous, since the rust must cycle to the alternate host every year and there is normally no interaction between the rust and live cottonwood leaves during the winter. Foliage emerges each spring free of Melampsora rust. However, Melampsora rust does not appear to pose a continuous selection pressure on its host. Some years, nearly all trees are heavily infected; during other years, many of the same trees remain free from infection. Furthermore, even during years of heavy infection, major defoliation is not often observed and at any rate does not occur until late in the growing season. Major resistance genes and virulence genes have been isolated from natural populations, and from this it is often inferred that major genes do have a role to play in disease systems that are stable. However another interpretation could be that these genes have a function other than in disease resi stance. It is possible to imagine that major virulence genes can be maintained in the pathogen population if there are powerful 34 selection pressures which act to favor distinct polymorphisms. One such pressure could be a physiological barrier such as lack of sexual reproduction. For example, in a microcylic rust without spermogonia, a mutation could give rise to a powerful virulence gene which allows its possessors to spread through the population. Lack of meiotic crossing-over would prevent gene combinations or a new genetic background for this major virulence gene where its effects would be modified, or where it would be incompatible. Another selection presssure could be the powerful artificial selection posed by monocultures of a cultivar with a single set of strong resistance genes. Powerful virulence genes would then be required in the pathogen population for survival. Sidhu (1980 p. 396) lists 27 pathosystems in which gene-for-gene relationships have been implied, suggested, or demonstrated. None of these systems are naturally wild. However the occidentalis - P_;_ trichocarpa pathosystem which I studied is wild, and not under strong artificial selection pressures to produce major resistance or virulence genes. As well both partners are quite capable of outcrossing with other members of their species so no physiological pressure is present to maintain distinct disease interaction polymorphisms. There is a valuable practical implication from these results. The lack of qualitative resistance and virulence in this system holds the promise that cottonwood resistance will not be devastatingly overcome when used in plantations. Cottonwoods reproduce very easily asexually by cladoptosis (twig 35 drop), and so it would not be uncommon to find neighboring individuals of the same genotype. Thus one could infer that even limited monoculture could be possible for plantations of black cottonwood. However, uniform plantations might strongly favor other diseases and disorders of black cottonwood. There is also a theoretical implication of these results. Very little work has been done on natural pathosystems in the area of host:parasite genetics. The lack of qualitative resistance and virulence in the M. occidentalis - P. trichocarpa pathosystem indicates that qualitative interactions do not play a major role in disease in this system. However, many more studies of other natural pathosystems are required before making the general conclusion that qualitative interactions arising from physiological specialization due to gene-for-gene effects do not occur in natural pathosystems. 36 CONCLUSION An analysis of variance of average daily spore production by 10 isolates of Melampsora occidentalis on 14 clones of Populus trichocarpa showed no indications of physiological specialization. Clonal resistance as well as isolate virulence differed significantly. Analysis of variance of total spores, average daily pustule production, and latent period all gave this same result. The overall average spore production during the time from inoculation to twice the latent period was 650 spores/disk/day. Latent period ranged from 6 to 12 days with a median at 8 days. The sampling technique was biased toward selecting specimens which had the opportunity to be physiologically specialized, since isolates were collected along with their hosts in late summer; yet this specialization was not found. The lack of qualitative resistance and virulence indicates that qualitative interactions do not play a major role in disease in this system. This finding holds the promise that cottonwood resistance will be of a durable nature and not devastatingly overcome when used in plantations. » 37 LITERATURE CITED Bier, J. 1965. Some effects of foliage saprophytes in the control of Melampsora leaf rust on Black Cottonwood. Forestry Chronicle 41:306-315. Browning, A., M. Simons & E. Torres. 1977. Managing host genes: epidemiologic and genetic concepts. in: Plant Disease, Vol. I. eds: J. Horsfall & E. Cowling. Academic Press, New York. pp. 191-212. Burdon, J. 1980. Variation in disease-resistance within a population of Trifolium repens . Journal of Ecology 68:737-744. Burdon, J. & D. Marshall. 198.1. Inter- and Intra-specific diversity in the disease-response of Glycine species to the leaf-rust fungus Phakopsora pachyrhizi . Journal of Ecology 69:381-390 Burdon, J., J. Oates & D. Marshall. 1983. Interactions between Avena and Puccinia species: I. Wild hosts: Avena  barabata r A_^_ fatua , A_j_ ludoviciana . Journal of Applied Ecology 20:571-584. Chandrashekar, M. 1982. Effects of some chemicals employed in the detached leaf culture of Populus on the infection of Melampsora larici-populina . Eur. J. For. Path. 12(6/7):30l-307. & W. Heather. 1980. Reactions of poplar clones to physiologic races of Melampsora larici-populina . Phytopathology 72:327-330. Dinoor. A. 1977. Genes for disease resistance from native populations, pp. 15-25 in: Induced Mutations against Plant Diseases. International Atomic Energy Agency, Vienna, Austria. Donnelly, J. 1974. Seasonal changes in photosynthate transport within elongating shoots of Populus grandidentata . Can. J. Bot. 52:2547-2559. Eldridge, K. & A. Matheson & W. Stahl. 1973. Genetic Variation in resistance to poplar leaf rust. Austral. For. Res. 6:53-59. Ellingboe, A. 1975. Horizontal resistance: An artifact of experimental procedure? Australian Plant Pathology Society Newsletter 3:44-46. Ellingboe, A. 1981. Changing concepts in host:pathogen genetics. Annual Review of Phytopathology 19:125-143. FAO/IPC. 1980. Poplars and Willows in wood production and land use. FAO forestry series #10, Rome. 328 pp. 38 Farmer, R. 1970. Genetic variation among open-pollinated progeny of eastern cottonwood. Ann. Appl. Biol. 31:149— 151. Farmer, R. & J. Wilcox. 1968. Preliminary testing of eastern cottonwood clones. Theoretical and Applied Genetics 38:197-201 . Fleming, R. & C. Person. 1982. Consequences of polygenic determination of resistance and aggressiveness in nonspecific host-parasite relationships. Canadian Journal of Plant Pathology 45:680-685. Flor, H. 1955. Host-parasite interaction in flax rust — its genetics and other implications. Phytopathology 45:680-685. Fowells, H. 1965. Silvics of Forest Trees of the United States. Black Cottonwood. pp. 508-518. USDA Forest Service Agriculture Handbook 271. 762 pp. Fry, W. 1982. Principles of Plant Disease. Academic Press, Toronto. pp. Galloway, G. & J. Worrall. 1979. Cladoptosis: a reproductive strategy in black cottonwood? Can. J. For. Res. 9:122-125. Gremmen, J. 1980. Problems and prospects in breeding Melampsora rust resistant poplars. Folia Forestalia 422:5-9. Hansen, E., L. Moore, D. Netzer, M. Ostry & H. Phipps & J. Zavitkovski. 1983. Establishing intensively cultured hybrid poplar plantations for fuel and fiber. USDA For. Serv. Gen. Tech. Rep. NC-78. 24 pp. Heather, W. & M. Chandrashekar. 1982. Stability of disease resistance in natural and agro-ecosystems. Trans. Brit Mycol. Soc. 78(2) : 381-383 -& J. Sharma. 1977. Some aspects of poplar rust research in Australia. Australian Forestry 40(l):28-43. , & A. Miller. 1980. Physiologic specialization in Melampsora larici-populina on clones of poplar demonstrating partial resistance to leaf rust. Australian Forestry Research 10:125-131. Heybroek, H., B. Stephan & K. von Weissenberg. 1982. Resistance to Diseases and Pests in Forest Trees. Centre for Agricultural Publishing and Documentation, Wageningen, The Netherlands. 502 pp. Johnson, R. 1979. The concept of durable resistance. Phytopathology 69:198-199. 39 Jokela, J. 1966. Incidence and heritability of Melampsora rust in Populus deltoides . in: Breeding pest-resistant trees, eds: H. Gerhold et al. Pergammon Press, Oxford. pp. 111-117. Joly, R. 1959. Influence des defoliateurs forestiers sur 1'accroissement. (The influence of forest defoliation on increment, english summary in Forestry Abstracts 21:2046). Revue Forestiere Francaise 11:775-784. Kiyosawa, S. 1982. Genetics and epidemiological modelling of plant disease resistance. Ann. Rev. Phytopath. 20:93-117. Krzan, Z. 1980. Effect of climatic factors on development of the disease caused by Melampsora larici-populina . Folia Forestalia 422:10-13. Lin, M. & H. Edwards. 1974. Primary penetration process in powdery mildewed barley related to host cell age, cell type, and occurrence of basic staining material. New Phytologist 73:131-137. Loegering, W. 1978. Current concepts of interorganismal genetics. Annual Review of Phytopathology 16:309-320. Longo, N., F. Moriondo S< B. Longo. 1980. Overwintering and germinability of teliospores of Melampsora pinitorqua . Folia Forestalia 422:19-24. McBride, R. 1969. A microbiological control of Melampsora  medusae . Can. J. Bot. 47:711-715. McNabb, H., R. Hall & M. Ostry. 1982. Biological and physical modifications of the environment and the resulting effect upon the host-parasite interactions in short-rotation tree crops. in: Resistance to Diseases and Pests in Forest Trees. eds: H. Heybroek, B. Stephans & K. von Weissenberg. Center for Agricultural Publishing and Documentation, Wageningen. pp. 60-71. Muhle-Larsen, C. 1963. Considerations sur 1'amelioration du genre Populus et specialement sur la section Aigeiros. FAO/FORGEN 63~2b/9. 8 pp. 1970. Recent advances in poplar breeding. International Review of Forestry Research 3:1-67. Nelson, R. 1978. Genetics of Horizontal Resistance to Plant Diseases. Ann. Rev. Phytopath. 16:359-378. Omar, M. & W. Heather. 1979. Effect of saprophytic phylloplane fungi on germination and development of Melampsora larici- populina . Trans. Brit. Mycol. Soc. 72:225-231. 40 Palmberg, C. 1977. Selecting for rust resistance in poplars in Australia. Third World Consultation on Forest Tree Breeding. Canberrra, 1977. pp. 220-230. Robinson, R. 1979. Permanent and impermanent resistance to crop parasites; A re-examination of the Pathosystem concept with special reference to Rice Blast. A. Pflanzenzuchtung 83:1-39. Robinson, R. 1980. New concepts in breeding for disease resistance. Annual Review of Phytopathology 18:189-210. Schipper, A.L. & D.H. Dawson. 1974. Poplar leaf rust -- a problem in maximum wood fiber production. Plant Disease Reporter 58:721-723. Schreiner, J. 1974. Populus L . Poplar. in: Seeds of Woody Plants in the United States. USDA Forest Service Agriculture Handbook #450. pp. 645-655. Scott, P. & R. Johnson & M. Wolfe, & H. Lowe & F. Bennett. 1978. Host specificity in cereal parasites in relation to their control. Applied Biology 5:349-393. Shain, L. 1976. Etiology, epidemiology and control of Melampsora rust of Cottonwood. in: Symposium on eastern cottonwood and related species. eds: B. Thielges & S. Land. Louisiana State University, pp. 189-198. Shain, L. & P. Cornelius. 1979. Quantitative inoculation of eastern cottonwood leaf tissue with Melampsora medusae under controlled condtions. Phytopathology 69:301-304. Sharma, J. & W. Heather. 1979a. Comparison of disease parameters for quantitative assessments of Melampsora leaf rust in clones of Populus species. Trans. Brit. My col.. Soc. 72(3) :483-488. & 1979b. A method for determining density of suspensions of urediniospores of Melampsora larici-populina causing leaf rust of. Poplar. Trans. Brit. Mycol. Soc. 72(3):479-482. & 1981. Antagonism by three species of Cladosporium to three races of Melampsora larici-populina Kleb. Austral. For. Res. 11:283-293. & 1982. Temperature sensitivity of the antagonism of Cladosporium species to races of Melampsora larici- populina on cultivars of ) U Populus X euramericana (Dode) Gainier. Phytopath. Zeitshcrift 105:61-70. , & P. Winer. 1980. Effect of leaf maturity and shoot age of clones of Populus species on susceptibility to Melampsora larici-populina . Phytopathology 70:548-554. 41 Sidhu, G. 1980. Genetic analysis of plant parasitic systems. Proceedings of the XIV International Congress of Genetics, Vol. 1, book two. pp. 391-408. Singh, S. & W. Heather. 1981. An improved method for detached leaf culture of Melampsora leaf rust of Populus species. Trans. Brit. Mycol. Soc. 77:436-437. Spiers, A.G. 1978. Effect of light, temperature, and relative humidity on germination of urediniospores of, and infection of poplars by Melampsora larici-populina and M_j_ medusae . New Zealand Journal of Science 21:393-400. Thielges, B. & J. Adams. 1975. Genetic variation and heritability of Melampsora leaf rust resistance in Eastern cottonwood. Forest Science 21:278-282 Thielges, B. & S. Land. eds. 1976. Symposium on eastern cottonwood and related species. Louisiana State University, Baton Rouge, Louisiana. Toole, R. 1967. Melampsora medusae causes cottonwood rust in the lower Mississippi valley. Phytopathology 57:1361-1363. Vanderplank, J. 1963. Plant diseases, epidemics and control. Academic Press, New York. 349 pp. 1982. Host-pathogen Interactions in Plant Disease. Academic Press, Toronto. 207 pp. Vloten, H. van. 1949. Kruisingsproeven met rassen van Melampsora larici-populina (klebahn) . (Crossing experiments with strains of Melampsora larici-populina , summary in english) Tijdschrift over Plantenziekten 55:196-209. Wahl, I., N. Eshed, A. Segal & Z. Sobel. 1978. Significance of wild relatives of small grains and other wild grasses in cereal powdery mildews. in: The Powdery Mildews. ed: D. Spencer. Academic Press, New York. pp. 83- 100 Waller, J. 1981. Fungal pathogens in the air and in plant shoots. in: Methods in plant pathology, ed: R. Burchill. Phytopathological Papers 26:9-16. Wang, D. 1961. The nature of starch accumulation at the rust infection sites in leaves of pinto bean plants. Can. J. Bot 39:1595-1604. Weissenberg, K. von. 1980. Effect of temperature on dormancy, activation, and long-term storage of teliospores of Melampsora pinitorqua . Folia Forestalia 422:32-36. Widin, K. & A. Schipper. 1976. Epidemiology and impact of Melampsora medusae leaf rust on hybrid poplars. in: Intensive plantation culture: five years research. USDA 42 For. Serv. Gen. Tech. Rep. NC-21, p. 63-74. Wilcox, J. & R. Farmer. 1967. Variation and inheritance of juvenile characteristics of eastern cottonwood. Silvae Genetica 16:162-165. Zadoks, J. 1961. Yellow rust on wheat, studies in epidemiology and physiologic specialization. T. Plziekten 67:69-256. (as quoted in Zadoks, 1972. pp 43-63 in Biology of Rust Resistance in Forest Trees). Zadoks, J. & R. Schein. 1978. Epidemiology and plant disease management, the known and the needed. in: Comparative Epidemiology. eds: J. Paltri & J. Kranz. Third International Congress of Plant Pathology, pp. 1-17. Ziller, W. 1959. Studies of western tree rusts IV: Uredinopsis  hashiokai and tL pteridis causing perennial needle rust of fir. Can. J. Bot. 43:217-230. Ziller, W. 1965. Studies of western tree rusts VI: The aecial host ranges of Melampsora albertensis , M^_ medusae , and M._ occidentalis . Canadian Journal of Botany 43:217-230. 1974. The tree rusts of Western Canada. Canadian Forest Service Dept. of Environment Publication #1329. 272 pp. Zimmer, D. & D. Rehder. 1976. Rust resistance of wild Helianthus species of the North Central United States. Phytopathology 66:208-211. 43 APPENDIX A: Various labels have been applied to the resistance, virulence and disease interaction. The following list is adapted from Browning, Simons & Torres (1977): VERTICAL PERPENDICULAR SPECIFIC MAJOR GENE MONOGENIC OLIGOGENIC MULTIPLE ALLELE QUALITATIVE HIGH SEEDLING HYPERSENSITIVE PROTOPLASMIC DISCRIMINATORY DIFFERENTIAL HORIZONTAL LATERAL NON-SPECIFIC MINOR GENE POLYGENIC MULTIGENIC MULTIPLE GENE QUANTITATIVE LOW,MODERATE ADULT NON-HYPERSENSITIVE NON-PROTOPLASMIC DILATORY UNIFORM FIELD DURABLE TOLERANCE Vanderplank 1963 Vanderplank 1963 Zadoks 1961 Browning et al. 1977 Vanderplank 1963 Johnson 1979 These are but some of the terms which have been used to describe resistance, virulence, and disease interaction. Within each column the terms given are at least partially synonymous. Publications in the field of host-parasite genetics most commonly use the terms vertical and horizontal resistance. However, these terms have become burdened with both genetic and epidemiological definitions which, according to many people, are not necessarily equivalent (Johnson 1979, Ellingboe 1981). Furthermore, the two columns represent two distinct categories to some authors (as above) while others do not see a distinction and view the corresponding terms as representing two extremes of a continuum (Nelson, 1978; Ellingboe, 1975). The terms used in this paper, qualitative and quantitative, are not necessarily synonymous with other terms in the same column, and should be read solely with the definitions given them in this paper. APPENDIX B: 1. LISTING OF ALL CLONES AND ISOLATES Name Isolate Clone Location ALK X Allison Lake, B.C. CAL X Calgary, Alberta CORD X Corduroy Trail, U.B.C Endowment Lands GRAN X X Grandview & Nanaimo Str, Vancouver HEBB X X Hebb Str. (Renfrew & 10th), Vancouver IMP X X Imperial Drive & 16th, Vancouver KEN X X Boundary & Nelson Str, Vancouver MAR X Fraser Viewpoint, U.B.C. PIC X X Picnic Point, Kamloops, B.C. POCO X X Port Coquitlam, B.C. POND X Ponderosa Cafeteria, U.B.C. PRG X Prince George, B.C. RUP X X Rupert & Grandview, Vancouver SAL X Salish Trail near B.C. Res., U.B.C. THUN X X Thunderbird Stadium, U.B.C. TRQ X Tranquille Farm, Kamloops, B.C. These clones did not survive in the greenhouse: ALK and PRG. These isolates were not present: Cal,Cord,Mar,Pond,Sal,TrQ. 2. ISOLATE COLLECTION AND FORMATION DATES COLLECTION FORMATION Isolate Telia Uredia Basidia Pycnia Aecia Uredia Alk 10/83 Gran 10/82 8/83 7/83 Hebb 8/83 Imp 1 0/82 20/5/83 31/5/83 16/6/83 3/7/83 Ken 1 1/82 8/83 6/83 Pic 1 0/82 18/6/83 27/6/83 15/7/83 27/7/83 Poco 8/83 PrG 8/83 Rup 8/83 Thun 10/82 8/83 7/83 APPENDIX C: Leaf age susceptibility study Three stems were stripped entirely of their leaves and the leaves were cut into halves for inoculation. The leaves were inoculated with several droplets of the spore suspensions of an isolate known to be virulent on that clone. The droplets were then spread all over the leaf halves with a paintbrush. leaf height* length width pustules (cm) (cm) (cm) (number) 1-1 2.2 8.0 2.5 0 1-2 4.6 11.6 4.7 1 1-3 7.2 14.9 6.5 g ** 1-4 10.5 15.5 5.7 8 1-5 14.0 14.6 5.5 4 1-6 18.0 13.8 5.9 1 1-7 21.2 10.8 5.4 10 1-8 24.8 11.5 5.2 10 2-1 .2 5.2 1 .8 0 2-2 .5 5.5 1 .9 0 2-3 1 .2 4.9 2.0 1 o ** 2-4 2.4 4.6 2. 1 2 2-5 4.7 4.8 1 .9 5 2-6 6.5 5.4 1 .8 5 2-7 7.8 4.3 1 .7 10 2-8 9.0 5.0 1 .9 1 2-9 10.2 5.0 1 .9 12 2-10 12.0 6.0 2.3 8 3-1 1 .0 7.5 3.9 0 3-2 2.0 7.0 3.4 0 3-3 2.3 8.2 3.5 7 * * 3-4 4.5 9.2 4.3 8 3-5 6.0 6.2 3.5 1 3-6 7.2 8.1 3.8 25 3-7 8.2 7.5 3.5 2 3-8 9.8 8.0 4.0 5 3-9 11.0 9.2 4.0 7 3-10 13.2 7.2 3.4 8 * height refers to distance from the top of the plant, and the smallest height means the youngest leaf, (e.g. leaf 1-1 was the youngest leaf on stem 1). ** This was judged to be the most recently expanded leaf based upon the length and width of the leaf. By pustule number, this leaf was also found to be the youngest or second youngest leaf susceptible to the rust. APPENDIX D: Mean average daily spore production df each petri dish The correlation is r=-.14, which is not significant. vs. inoculum concentration applied to i dish. MEAN SPORE PRODUCTION 23396. + 20276. 17156. 14037. 10917. 7796.8 + * 4676.9 + * * * * 1557.0 + + 1000.0 5111.1 9222.2 3055.6 7166.7 11278. 13333. 17444. 15389. INOCULUM CONCENTRATION 19500. ON ON APPENDIX E: Haemocytometer spore count vs. light absorbance of the same spore suspension. This relationship was used to derive an equation for calculating spores from absorbance. The derived equation: SPORES = 1,228,300 * ABSORBANCE - 7294 has an r=.91 .14000 +6+ HAEMOCYTOMETER SPORE COUNT + * .12000 +6+ + .10000 +6+ 80000. + 60000. 40000. 20000. * * * 2 22 * * *** * *2 + 3 * * 22*2*2* *2 * * 3 5*2*2*3 * * +*789*722*** 2 * + + + + + +-.86950 -3 .23426 -1 .12148 -1 .45982 -1 .68538 -1 .91095 -1 .34704 -1 .57260 -1 .79817 -1 LIGHT ABSORBANCE 10237 -P-48 APPENDIX F: ANALYSIS OF VARIANCE 1. Average spore production during twice the latent period for all specimens SOURCE Block Treatment Isolates Clones Interaction Error Total i. F, 8 1 39 1112 1 259 9 1 3 1 1 7 F-VALUE 6.944 2.357 5.318 14.442 .923 F-PROB .0000 .0000 .0000 .0000 .6884 2. Average spore production for Vancouver Region specimens SOURCE D. F. F-VALUE F-PROB Block 8 5.003 .0000 Treatment 76 1 .885 .0001 Isolates 6 8.600 .0000 Clones 1 0 5.319 .0000 Interaction 60 .870 .7457 Error 608 Total 692 3. Pustule count over twice latent period SOURCE Block Treatment Isolates Clones Interaction Error Total i. F, 8 139 1112 1 259 9 1 3 1 17 F-VALUE 6.693 2.514 8. 160 17.293 .851 F-PROB .0000 .0000 .0000 .0000 .8673 4. Number of spores produced during twice latent period SOURCE Block Treatment Isolates Clones Interaction Error Total D. F, 8 1 39 1112 1 259 1 1 F-VALUE 6.865 341 219 1 1 7 ,831 2, 7, 16, F-PROB .0000 .0000 .0000 .0000 .8989 5. Latent Period (time in days from inoculation to symptoms) SOURCE Block Treatment Isolates Clones Interaction Error Total D. F, 8 1 39 1112 1259 1 1 F-VALUE 3.548 1 .825 7.714 8. 105 .869 F-PROB .0000 .0000 .0000 .0000 .8321 49 APPENDIX G: Analysis of variance for isolate-clone interaction where lowly correlated clones and isolates have been subject to analysis on U.B.C. MTS program *ANOVAR MODEL,SPORIN=A+B+C+AC+E LIMITS,2,9,2 FACTORIAL DESIGN A1=Hebb A2=Gran C1=CAL C2=IMP DATA,-DAT RANDOM,B RENAME,BLOCK=B,ISOLATES=A,CLONES=C,INTERACTION=AC RANGE,(SOURCE=ALL),(SIGREQD=NONE),(CORREL=YES) OPTIONS,PRNTEMS INPUT,A(5,2),B(2,2),C(8,2),SPORIN(55,8) ANALYSIS OF VARIANCE FOR AVERAGE SPORE PRODUCTION RATE SOURCE D.F. F-VALUE F PROB ISOLATES 1 0.9022 0 .3543 CLONES 1 3.1251 0 .0864 INTERACTION 1 10.8899 0 .0031 BLOCK 8 1.1114 0 .3905 ERROR 24 TOTAL 35 I SOL CLONE MEAN STD ERROR RANGE TEST 1. Hebb * CAL 2324.000 168.444 A 2. Hebb * IMP 454.889 40.725 B 3. Gran * CAL 756.556 26.8464. Gran * IMP 1321.667 177.073 B A ** This was the only significant F-value ( < .05 ), so a Duncan's Multiple Range Test was performed on these means of isolate-clone interaction. Means followed by the same letter (A or B) are not significantly di fferent. The conclusion can be made that elimination of closely correlated clones and isolates (with respect to disease severity) can lead to a significant isolate-clone interaction but without the more important significant differences between clones or between isolates. 50 APPENDIX H: CALCULATIONS FOR COMPONENTS OF VARIANCE Begin with an analysis of variance with mean squares: Source degrees of freedom mean square blocks 8 9013334 isolates 9 6415802 clones 13 1742379interaction 117 1206473 error 1112 1298330 Then calculate the expected mean squares (EMS): EMS(blocks) = VAR(error) + N * A * C * VAR(blocks) EMS(isolates) = VAR(error) + N * B * C * VAR(isolates) + N * B * VAR(interaction) EMS(clones) = VAR(error) + N * A * B * VAR(clones) + N * B * VAR(interaction) EMS(interaction) = VAR(error) + N * B * VAR(interaction) EMS(error) = VAR(error) Let the EMS values equal their mean square counterparts, and from the analysis of variance: N is the number of replication's 0 A is the number of isolates 10 B is the number of blocks 9 C is the number of clones 14 Then solve for the variances (VAR): VAR(blocks) = ( MS(blocks) - MS(error) ) / N * B = 55107 VAR(isolates) = ( MS(isolates) - MS(interact) ) / N * B * C =41344 VAR(clones) = ( MS(clones) - MS(interact) ) / N * A * B = 180192 VAR(interact) = ( MS(interact) - MS(error) ) / N * B = 0 VAR(error) = MS(error) =1298330 VAR(total)= 1574973 (the sum of these variances) Now calculating all the variances as a percentage of the total will give what is found in Table III. 


Citation Scheme:


Usage Statistics

Country Views Downloads
United States 7 0
China 4 5
Japan 3 0
City Views Downloads
Ashburn 4 0
Tokyo 3 0
Wilmington 3 0
Beijing 2 0
Shenzhen 2 5

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items