A simulation model for epidemics of stem rust in ryegrass seed crops.

A simulation model (STEMRUST_G, named for stem rust of grasses) was created for stem rust (caused by Puccinia graminis subsp. graminicola) in perennial ryegrass grown to maturity as a seed crop. The model has a daily time step and is driven by weather data and an initial input of disease severity from field observation. Key aspects of plant growth are modeled. Disease severity is modeled as rust population growth, where individuals are pathogen colonies (pustules) grouped in cohorts defined by date of initiation and plant part infected. Infections due to either aerial spread or within-plant contact spread are modeled. Pathogen cohorts progress through life stages that are modeled as disease cycle components (colony establishment, latent period, infectious period, and sporulation) affected by daily weather variables, plant growth, and fungicide application. Fungicide effects on disease cycle components are modeled for two commonly used active ingredients, applied preinfection or postinfection. Previously validated submodels for certain disease cycle components formed the framework for integrating additional processes, and the complete model was calibrated with field data from 10 stem rust epidemics. Discrepancies between modeled outcomes and the calibration data (log10[modeled]-log10[observed]) had a mean near zero but considerable variance, with 1 standard deviation=0.5 log10 units (3.2-fold). It appears that a large proportion of the modeling error variance may be due to variability in field observations of disease severity. An action threshold for fungicide application was derived empirically, using a constructed weather input file favorable for disease development. The action threshold is a negative threshold, representing a level of disease (latent plus visible) below which damaging levels of disease are unable to develop before the yield-critical crop stage. The model is in the public domain and available on the Internet.

Evaluation of the Ryegrass Stem Rust Model STEMRUST_G and Its Implementation as a Decision Aid.

STEMRUST_G, a simulation model for epidemics of stem rust in perennial ryegrass grown to maturity as a seed crop, was validated for use as a heuristic tool and as a decision aid for disease management with fungicides. Multistage validation had been used in model creation by incorporating previously validated submodels for infection, latent period duration, sporulation, fungicide effects, and plant growth. Validation of the complete model was by comparison of model output with observed disease severities in 35 epidemics at nine location-years in the Pacific Northwest of the United States. We judge the model acceptable for its purposes, based on several tests. Graphs of modeled disease progress were generally congruent with plotted disease severity observations. There was negligible average bias in the 570 modeled-versus-observed comparisons across all data, although there was large variance in size of the deviances. Modeled severities were accurate in >80% of the comparisons, where accuracy is defined as the modeled value being within twice the 95% confidence interval of the observed value, within ±1 day of the observation date. An interactive website was created to produce disease estimates by running STEMRUST_G with user-supplied disease scouting information and automated daily weather data inputs from field sites. The model and decision aid supplement disease managers' information by estimating the level of latent (invisible) and expressed disease since the last scouting observation, given season-long weather conditions up to the present, and it estimates effects of fungicides on epidemic development. In additional large-plot experiments conducted in grower fields, the decision aid produced disease management outcomes (management cost and seed yield) as good as or better than the growers' standard practice. In future, STEMRUST_G could be modified to create similar models and decision aids for stem rust of wheat and barley, after additional experiments to determine appropriate parameters for the disease in these small-grain hosts.

Pathotype-specific QTL for stem rust resistance in Lolium perenne.

A genetic map populated with RAD and SSR markers was created from F1 progeny of a stem rust-susceptible and stem rust-resistant parent of perennial ryegrass (Lolium perenne). The map supplements a previous map of this population by having markers in common with several other Lolium spp. maps including EST-SSR anchor markers from a consensus map published by other researchers. A QTL analysis was conducted with disease severity and infection type data obtained by controlled inoculation of the population with each of two previously characterized pathotypes of Puccinia graminis subsp. graminicola that differ in virulence to different host plant genotypes in the F1 population. Each pathotype activated a specific QTL on one linkage group (LG): qLpPg1 on LG7 for pathotype 101, or qLpPg2 on LG1 for pathotype 106. Both pathotypes also activated a third QTL in common, qLpPg3 on LG6. Anchor markers, present on a consensus map, were located in proximity to each of the three QTL. These QTL had been detected also in previous experiments in which a genetically heterogeneous inoculum of the stem rust pathogen activated all three QTL together. The results of this and a previous study are consistent with the involvement of the pathotype-specific QTL in pathogen recognition and the pathotype-nonspecific QTL in a generalized resistance response. By aligning the markers common to other published reports, it appears that two and possibly all three of the stem rust QTL reported here are in the same general genomic regions containing some of the L. perenne QTL reported to be activated in response to the crown rust pathogen (P. coronata).

Effects of the Mycoparasite Sphaerellopsis filum on Overwintering Survival of Stem Rust in Perennial Ryegrass.

Sphaerellopsis filum is a mycoparasite of Puccinia graminis subsp. graminicola, a rust fungus that causes widespread crop damage on perennial ryegrass grown for seed. In observations taken over the winter months, S. filum was found naturally colonizing 2% of P. graminis subsp. graminicola uredinia on first-year plantings and 25% of the uredinia on second-year plantings of perennial ryegrass. In controlled experiments conducted in glasshouses and growth chambers, S. filum applied to rust-inoculated plants reduced the lifetime spore production of P. graminis subsp. graminicola pustules by half, from 39,000 to 18,000 spores/pustule. Mist duration, temperature, and P. graminis subsp. graminicola pustule age at the time of S. filum inoculation had significant effects on the proportion of P. graminis subsp. graminicola pustules infected by S. filum. Fifty percent of all P. graminis subsp. graminicola pustules were infected when S. filum was inoculated onto erumpent pustules and incubated above 5°C for 48 h while exposed to mist. Plants inoculated with both fungi under controlled conditions and then planted into the field had a significantly greater proportion of P. graminis subsp. graminicola pustules infected with S. filum, and a significantly reduced P. graminis subsp. graminicola overwintering population, compared with plants inoculated with P. graminis subsp. graminicola only. First-year stands of perennial ryegrass treated in the field with monthly applications of S. filum had more than 10 times the proportion of pustules infected with S. filum and 50% less P. graminis subsp. graminicola disease compared with the nontreated controls. In comparison, plants with one application of fungicide during the winter had 98% lower P. graminis subsp. graminicola severity than the P. graminis subsp. graminicola-only control There were no effects of S. filum or fungicide application on rust severity in 2- or 3-year-old perennial ryegrass stands.

Sensitivity of Disease Management Decision Aids to Temperature Input Errors Associated with Sampling Interval and Out-of-Canopy Sensor Placement.

Many plant disease epidemic models, and the disease management decision aids developed from them, are created based on temperature or other weather conditions measured in or above the crop canopy at intervals of 15 or 30 min. Disease management decision aids, however, commonly are implemented based on hourly weather measurements made from sensors sited at a standard placement of 1.5 m above the ground or are estimated from off-site weather measurements. We investigated temperature measurement errors introduced when sampling interval was increased from 15 to 60 min, and when actual in-canopy conditions were represented by temperature measurements collected by standard-placement sensors (1.5 m above the ground, outside the canopy) in each of three crops (grass seed, grape, and hops) and assessed the impact of these errors on outcomes of decision aids for grass stem rust as well as grape and hops powdery mildews. Decreasing time resolution from 15 to 60 min resulted in statistically significant underestimates of daily maximum temperatures and overestimates of daily minimum temperatures that averaged 0.2 to 0.4°C. Sensor location (in-canopy versus standard-placement) also had a statistically significant effect on measured temperature, and this effect was significantly less in grape or hops than in the grass seed crop. Effects of these temperature errors on performance of disease management decision aids were affected by magnitude of the errors as well as the type of decision aid. The grape and hops powdery mildew decision aids used rule-based indices, and the relatively small (±0.8°C) differences in temperature observed between in-canopy and standard placement sensors in these crops resulted in differences in rule outcomes when actual in-canopy temperatures were near a threshold for declaring that a rule had been met. However, there were only minor differences in the management decision (i.e., fungicide application interval). The decision aid for grass stem rust was a simulation model, for which temperature recording errors associated with location of the weather station resulted in incremental (not threshold) effects on the model of pathogen growth and plant infection probability. Simple algorithms were devised to correct the recorded temperatures or the computed infection probability to produce outcomes similar to those resulting from in-canopy temperature measurements. This study illustrates an example of evaluating (and, if necessary, correcting) temperature measurement errors from weather station sensors not located within the crop canopy, and provides an estimate of uncertainty in temperature measurements associated with location and sampling interval of weather station sensors.

Mapping with RAD (restriction-site associated DNA) markers to rapidly identify QTL for stem rust resistance in Lolium perenne.

A mapping population was created to detect quantitative trait loci (QTL) for resistance to stem rust caused by Puccinia graminis subsp. graminicola in Lolium perenne. A susceptible and a resistant plant were crossed to produce a pseudo-testcross population of 193 F(1) individuals. Markers were produced by the restriction-site associated DNA (RAD) process, which uses massively parallel and multiplexed sequencing of reduced-representation libraries. Additional simple sequence repeat (SSR) and sequence-tagged site (STS) markers were combined with the RAD markers to produce maps for the female (738 cM) and male (721 cM) parents. Stem rust phenotypes (number of pustules per plant) were determined in replicated greenhouse trials by inoculation with a field-collected, genetically heterogeneous population of urediniospores. The F(1) progeny displayed continuous distribution of phenotypes and transgressive segregation. We detected three resistance QTL. The most prominent QTL (qLpPg1) is located near 41 cM on linkage group (LG) 7 with a 2-LOD interval of 8 cM, and accounts for 30-38% of the stem rust phenotypic variance. QTL were detected also on LG1 (qLpPg2) and LG6 (qLpPg3), each accounting for approximately 10% of phenotypic variance. Alleles of loci closely linked to these QTL originated from the resistant parent for qLpPg1 and from both parents for qLpPg2 and qLpPg3. Observed quantitative nature of the resistance may be due to partial-resistance effects against all pathogen genotypes, or qualitative effects completely preventing infection by only some genotypes in the genetically mixed inoculum. RAD markers facilitated rapid construction of new genetic maps in this outcrossing species and will enable development of sequence-based markers linked to stem rust resistance in L. perenne.

Decision aids for multiple-decision disease management as affected by weather input errors.

Many disease management decision support systems (DSSs) rely, exclusively or in part, on weather inputs to calculate an indicator for disease hazard. Error in the weather inputs, typically due to forecasting, interpolation, or estimation from off-site sources, may affect model calculations and management decision recommendations. The extent to which errors in weather inputs affect the quality of the final management outcome depends on a number of aspects of the disease management context, including whether management consists of a single dichotomous decision, or of a multi-decision process extending over the cropping season(s). Decision aids for multi-decision disease management typically are based on simple or complex algorithms of weather data which may be accumulated over several days or weeks. It is difficult to quantify accuracy of multi-decision DSSs due to temporally overlapping disease events, existence of more than one solution to optimizing the outcome, opportunities to take later recourse to modify earlier decisions, and the ongoing, complex decision process in which the DSS is only one component. One approach to assessing importance of weather input errors is to conduct an error analysis in which the DSS outcome from high-quality weather data is compared with that from weather data with various levels of bias and/or variance from the original data. We illustrate this analytical approach for two types of DSS, an infection risk index for hop powdery mildew and a simulation model for grass stem rust. Further exploration of analysis methods is needed to address problems associated with assessing uncertainty in multi-decision DSSs.

Field assessment of a model for fungicide effects on intraplant spread of stem rust in perennial ryegrass seed crops.

Intraplant spread of stem rust (Puccinia graminis subsp. graminicola) in perennial ryegrass during tiller extension is a major determinant of epidemic severity and is dominated by stem extension dynamics. Simple equations for extension of inflorescence and internodes are presented and parameterized. These equations are combined with previously published equations for pathogen latent period and for postinfection efficacy of fungicides to produce a model for effects of fungicide type and timing on intraplant spread. The model is driven by thermal units, calculated from air temperature measurements. Three field experiments, conducted independently from the field experiments that provided data for plant growth model parameterization, were conducted to assess performance of the disease spread model. Either propiconazole or azoxystrobin, the two most commonly used fungicides for stem rust control, was applied to tillers that had stem rust pustules on the flag sheath and in which the inflorescence was partially extended. Intraplant spread of disease to the extending inflorescence (stem and flowerhead) was observed at several dates following treatment and compared with modeled severities. The model estimated accurately the severities of inflorescence infection for most treatments and observation times, with a correlation coefficient of 0.93 for modeled versus observed disease severities across the three experiments. The model correctly estimated the rank order of final severities among the treatments (fungicide type and timing). The model can be extended to intraplant spread of stem rust at all internodes and incorporated into decision support tools for fungicide type and timing in management of this disease.

Interaction of Fungicide Physical Modes of Action and Plant Phenology in Control of Stem Rust of Perennial Ryegrass Grown for Seed.

Azoxystrobin provided protective and curative effects against stem rust (caused by Puccinia graminis subsp. graminicola) of inoculated perennial ryegrass under field conditions, significantly reducing disease severity compared with the nontreated check when applied as much as 15 days before infection or 14 days after infection. Propiconazole had a significant effect when applied 13, but not 15, days before infection or 7, but not 9, days after infection. Either fungicide was very effective when applied near the time of infection, and the effectiveness of each fungicide was well described by a second- or third-order polynomial with time (days or stem rust latent periods) as the independent variable. When symptomatic plants were sprayed with a fungicide, subsequent urediniospore production per pustule was reduced by 73% in propiconazole-treated plants and by 95% in azoxystrobin-treated plants. Azoxystrobin modestly but significantly reduced germinability of urediniospores from sprayed pustules, unlike propiconazole. These differences between the two fungicides in physical modes of action resulted in a marked difference in their effects on secondary, within-plant spread of the disease. In normal disease development, sporulation from the inner (adaxial) face of pustules on the flag-leaf sheath produces inoculum that leads to numerous contiguous secondary infections along the length of the emerging inflorescence, but only 7% of azoxystrobin-treated sheath pustules had sporulation from the adaxial surface compared with 72 and 90% of propiconazole-treated and nontreated pustules, respectively. Tillers treated with propiconazole early in the process of within-plant spread had significantly greater final stem rust severity than those treated with azoxystrobin at this time. Results of these experiments will allow effects of fungicide application to be incorporated into quantitative epidemic models that describe disease development as a function of environment and plant phenology.

Regional Development of Orchardgrass Choke and Estimation of Seed Yield Loss.

A survey for choke, caused by Epichloë typhina, in orchardgrass seed-production fields in Oregon was conducted annually from 1998 to 2003. In all, 99 fields were inspected, 57 in more than 1 year, to produce a set of 217 observations on disease incidence. There was a significant increase in disease incidence in 38% of the revisited fields, and a significant reduction of incidence in 3%. Yearly increases in disease incidence were as high as 29% in individual fields, but the average yearly increase from 1999 to 2003 was 5 to 8%. In 1998, 60% of all surveyed fields were infested with choke and, by 2003, 90% were infested. Average disease incidence in fields in their first year of production was <2%, and average disease incidence in older fields was approximately 10%, in 2003. Seed yield loss was equal to disease incidence (percentage of tillers diseased), and we found no significant yield compensation in diseased stands. An observed correlation of disease incidence with disease prevalence (proportion of sampled sites infested w ithin a field) may permit simple estimation of incidence and, thus, of potential economic loss in an affected field. We estimate regional loss to the 2004 orchardgrass seed crop due to choke o be approximately $0.8 million.

Effect of Autumn Planting Date and Stand Age on Severity of Stem Rust in Seed Crops of Perennial Ryegrass.

Perennial ryegrass (Lolium perenne) grown for seed is planted in autumn for July harvest (first-year seed crop), then kept in production for subsequent yearly harvests. Plots of first-year perennial ryegrass planted in early November had only 3% as much stem rust in June as plots planted in mid-September. In other plots where fungicides were used to prevent rust development, seed yield from the November-planted plots was reduced by 23% compared with September-planted plots. In the second-year seed crop, stem rust severity in June was intermediate between severities in early- and late-planted first-year plots. The association of reduced stem rust severity with late planting for first-year crops was observed for five of six perennial ryegrass cultivars tested. Degree of reduction in disease severity due to planting date was greatest for the cultivars that had the highest disease severity in early-planted stands. Delay of autumn planting date may provide a useful cultural control method for first-year stands of perennial ryegrass seed crops.

New applications of statistical tools in plant pathology.

ABSTRACT The series of papers introduced by this one address a range of statistical applications in plant pathology, including survival analysis, nonparametric analysis of disease associations, multivariate analyses, neural networks, meta-analysis, and Bayesian statistics. Here we present an overview of additional applications of statistics in plant pathology. An analysis of variance based on the assumption of normally distributed responses with equal variances has been a standard approach in biology for decades. Advances in statistical theory and computation now make it convenient to appropriately deal with discrete responses using generalized linear models, with adjustments for overdispersion as needed. New nonparametric approaches are available for analysis of ordinal data such as disease ratings. Many experiments require the use of models with fixed and random effects for data analysis. New or expanded computing packages, such as SAS PROC MIXED, coupled with extensive advances in statistical theory, allow for appropriate analyses of normally distributed data using linear mixed models, and discrete data with generalized linear mixed models. Decision theory offers a framework in plant pathology for contexts such as the decision about whether to apply or withhold a treatment. Model selection can be performed using Akaike's information criterion. Plant pathologists studying pathogens at the population level have traditionally been the main consumers of statistical approaches in plant pathology, but new technologies such as microarrays supply estimates of gene expression for thousands of genes simultaneously and present challenges for statistical analysis. Applications to the study of the landscape of the field and of the genome share the risk of pseudoreplication, the problem of determining the appropriate scale of the experimental unit and of obtaining sufficient replication at that scale.

Role of phenology in host susceptibility and within-plant spread of stem rust during reproductive development of perennial ryegrass.

ABSTRACT Perennial ryegrass (Lolium perenne) was inoculated with urediniospores of Puccinia graminis subsp. graminicola at four stages of reproductive tiller development. All developmental stages, from expansion of the penultimate leaf through anthesis, were equally susceptible to infection measured as number of pustules per total inoculated plant area. However, within each stage from boot through anthesis, there was a negative correlation of susceptibility with age of host tissue, the highest disease severities occurring on expanding or newly expanded leaves or inflorescences. Within-plant spread of the disease from primary lesions on the flag leaf sheath to the elongating stem of the inflorescence was observed and verified. A single primary infection on the leaf sheath can produce an elongated strip of contiguous secondary infections on the inflorescence head and stem, which is a characteristic sign of stem rust. Position and timing of eruption of the secondary pustules on the stem can be explained by assuming that infections occur on the stem slightly distal to the overlying sheath lesion and erupt one latent period later, by which time tiller elongation has moved the stem infection site upward. Repeated infections occur as the stem elongates, producing contiguous lesions. Expansion rate and final lesion area on the stem were correlated with elongation rate of the inflorescence. Microscopic observation of plant tissue sections revealed that the fungus sporulates on the inner surface of the sheath, and germinating urediniospores are attached to the stem surface under the covering sheath starting 1 to 2 cm distal to the location of the sheath infection site.

Prediction of stem rust infection favorability, by means of degree-hour wetness duration, for perennial ryegrass seed crops.

ABSTRACT A weather-based infection model for stem rust of perennial ryegrass seed crops was developed and tested using data from inoculated bioassay plants in a field environment with monitored weather. The model describes favorability of daily weather as a proportion (0.0 to 1.0) of the maximum possible infection level set by host and inoculum. Moisture duration and temperature are combined in one factor as wet degree-hours (DH(w)) (i.e., degree-hours > 2.0 degrees C summed only over time intervals when) moisture is present). Degree-hours are weighted as a function of temperature, based on observed rates of urediniospore germination. The pathogen Puccinia graminis subsp. graminicola requires favorable conditions of temperature and moisture during the night (dark period) and also at the beginning of the morning (light period), and both periods are included in the model. There is a correction factor for reduced favorability if the dark wet period is interrupted. The model is: proportion of maximum infection = 1 - e((-0.0031) (DHw Index)), where DH(w) Index is the product of interruption-adjusted overnight weighted DH(w) multiplied by morning (first 2 h after sunrise) weighted DH(w). The model can be run easily with measurements from automated dataloggers that record temperature and wetness readings at frequent time intervals. In tests with three independent data sets, the model accounted for 80% of the variance in log(observed infection level) across three orders of magnitude, and the regression lines for predicted and observed values were not significantly different from log(observed) = log(predicted). A simpler version of the model using nonweighted degree hours (>2.0 degrees C) was developed and tested. It performed nearly as well as the weighted-degree-hour model under conditions when temperatures from sunset to 2 h past sunrise were mostly between 4 and 20 degrees C, as is the case during the growing season in the major U.S. production region for cool-season grass seed. The infection model is intended for use in combination with measured or modeled estimates of inoculum level, to derive estimates of daily infection.

Evaluation of Postharvest Burning and Fungicides to Reduce the Polyetic Rate of Increase of Choke Disease in Orchardgrass Seed Production.

Epichloë typhina, causal agent of choke disease, is detrimental to orchardgrass seed production. The fungus grows systemically, persists indefinitely as an endophyte within the perennial host, and produces a stroma bearing conidia and ascospores at the time of host flowering. The ascospores or conidia are thought to infect plants through the cut ends of tillers after swathing at harvest. The objective of this study was to evaluate the potential of systemic fungicides and postharvest treatments (burning and reclipping) to reduce the rate of increase of choke disease among plants. The fungicides propiconazole and azoxystrobin reduced germination of conidia of E. typhina in vitro, but had no effect on development of stroma or viability of conidia produced on infected plants. In field tests, fungicides applied to the cut ends of tillers after harvest were ineffective at reducing the rate of increase in disease. Likewise, reclipping of orchardgrass stubble after harvest, in an attempt to remove incipient infections in the tillers, did not reduce the rate of disease increase in the stand. However, propane-assisted burning of postharvest stubble did reduce the polyetic epidemic rate to 2.7% per year, compared with approximately 9.2% per year in plots receiving the fungicide, reclipping, or control treatments. The results suggest that postharvest burning may be useful in controlling choke disease and raise the possibility that there are infection courts other than the pith of cut reproductive tillers.

A temperature-based model for latent-period duration in stem rust of perennial ryegrass and tall fescue.

ABSTRACT A temperature-response curve for latent-period duration in stem rust (caused by Puccinia graminis subsp. graminicola) on perennial ryegrass and tall fescue was developed from constant-temperature experiments with inoculated plants and evaluated in field experiments. Under constant-temperature conditions, time from infection to 50% of pustules erupted for perennial ryegrass ranged from 54 days at 3.5 degrees C to 5.9 days at 26.5 degrees C. The latent period (LP(50)) duration of tall fescue was 69 and 8.5 days at these respective temperatures. The dependence of latent-period completion rate on temperature was best described as a linear increase in rate with temperature up to approximately 26 degrees C, then an exponential decline with temperature up to the maximum (lethal) temperature of approximately 35.5 degrees C. LP(20), the time required for 20% of open pustules to appear, was used as an estimator of latent-period duration for field observations. Percentage of one latent period completed per half hour (half-hourly rate), for perennial ryegrass was modeled as (0.0156T - 0.0206) {1 - exp[0.497(T - 35.5)]}, where T = average temperature ( degrees C) during the half-hour period. For tall fescue the modeled rate was (0.0109T - 0.00214) {1 - exp[0.417(T - 35.5)]}. Latent periods modeled by these equations were compared with observed latent periods in field experiments with potted plants, where half-hourly temperatures were measured. Linear regressions of modeled versus observed latent periods had adjusted r(2) values of 0.96 for perennial ryegrass and 0.93 for tall fescue. The latent-period equations could be used as components of a weather-based disease advisory model to optimize fungicide use in stem-rust management on these crops.

Host Range Differences Between Populations of Puccinia graminis subsp. graminicola Obtained from Perennial Ryegrass and Tall Fescue.

In the Pacific Northwest region of the United States, cool-season grasses grown for seed can be severely damaged by Puccinia graminis subsp. graminicola, causal agent of stem rust. Urediniospores of the pathogen, collected either from perennial ryegrass (Lolium perenne) or tall fescue (Festuca arundinacea), were tested for host range among selected grasses and cereals. Under greenhouse conditions, the inoculum from L. perenne could produce pustules on this host, as well as on Dactylis glomerata, Lolium multiflorum, Poa pratensis, and F. rubra subsp. rubra and subsp. commutata; it caused only limited pustule development (low incidence or pustule type) on F. arundinacea, F. ovina subsp. hirtula, P. annua, Hordeum vulgare, and Secale cereale. No symptoms were produced on Triticum aestivum or Avena sativa. The inoculum from F. arundinacea had a host range that included itself, D. glomerata, L. perenne, L. multiflorum, and F. rubra subsp. rubra and subsp. commutata; there was no sign of pustule development on Poa spp. or the cereal grains tested (T. aestivum, A. sativa, S. cereale, and H. vulgare). The two urediniospore populations differed also in rate of symptom development on most of their common hosts. There was a small, but statistically significant, difference in spore size among the populations from different hosts. No recommendation is made for separate taxonomic status of populations from F. arundinacea and L. perenne, but the adaptation of each to its own host should be considered when devising disease management strategies and studying host genetic resistance.

Freezing Temperature Effect on Survival of Puccinia graminis subsp. graminicola in Festuca arundinacea and Lolium perenne.

The Willamette Valley in Oregon is a major seed production area for cool-season grasses. Puccinia graminis subsp. graminicola survives over winter on its hosts as uredinial infections and causes epidemics of stem rust, the area's major disease on perennial ryegrass and tall fescue. To determine the possible importance of freezing temperature on rust survival, infected plants taken from the field were subjected to controlled freezing across a range of temperatures representative of those that can occur in the region. After treatment, plants were placed in a warm greenhouse, and the number of actively sporulating pustules was recorded at 3-day intervals for 21 days. The pathogen responded similarly to freezing treatments whether in perennial ryegrass or tall fescue. Compared with the nontreated standard, there was no significant reduction in pustule number after exposure to -3 or -6°C. Exposure of infected plants to -10°C caused a 75 to 90% reduction in rust survival, and exposure to -13°C killed all rust infections in tall fescue and over 99% in perennial ryegrass. The decline in rust survival with temperature was slightly steeper for perennial ryegrass than for tall fescue. A higher absolute number of infections in perennial ryegrass than in tall fescue resulted in higher numbers of surviving infections on perennial ryegrass. Survival of rust infections appeared to be primarily a function of host tissue survival. Between 1960 and 1997, years with winter temperatures as low as -10 or -13°C have occurred in the Willamette Valley with frequencies of approximately 39 and 8%, respectively. We conclude that year-to-year variation in winter temperature could have a significant effect on the survival of the grass stem rust pathogen.

Geographical Distribution and Incidence of Orchardgrass Choke, Caused by Epichloë typhina, in Oregon.

A 1998 survey was conducted in the Willamette Valley of Oregon, the major U.S. production area for orchardgrass seed, to determine the extent and severity of choke disease in Dactylis glomerata. This disease is a severe constraint to orchardgrass seed production in other parts of the world, but was unknown in Oregon prior to 1997. Thirty-seven fields, representing 27 cultivars and the geographical extent of production in the Willamette Valley, were selected from a list of fields registered for certification. Choke was found in 26 (70%) of the fields, and disease incidence ranged from <0.05 to 28% tillers affected. Five of the 37 fields had only trace levels of the disease, but four of the fields, representing three counties, had incidences >10%. In a survey of 16 fields located within 3.5 km of the 1997 discovery, choke was found in 14 fields, of which three had incidences >20%. Increase in disease incidence between 1997 and 1998 ranged from 2.1- to 3.3-fold in the three fields where disease increase was measured. One year after its presence was confirmed in Oregon, choke disease of orchardgrass is well-established throughout the orchardgrass seed producing region at damaging levels and is apparently able to increase and spread under the prevailing climatic and cultural conditions.