© 2007 Plant Management Network.
Spatiotemporal Spread of Potato virus S and Potato virus X in Seed Potato in Tasmania, Australia
Susan J. Lambert, Frank S. Hay, and Sarah J. Pethybridge, Tasmanian Institute of Agricultural Research (TIAR), University of Tasmania, Cradle Coast Campus, P.O. Box 3523, Burnie, Tasmania 7320, Australia; Calum R. Wilson, TIAR, University of Tasmania, New Town Research Laboratories, 13 St. John’s Avenue, New Town, Hobart, Tasmania 7008, Australia
Lambert, S. J., Hay, F. S., Pethybridge, S. J., Wilson, C. R. 2007. Spatiotemporal spread of Potato virus S and Potato virus X in seed potato in Tasmania, Australia. Online. Plant Health Progress doi:10.1094/PHP-2007-0726-07-RS.
The spatial and temporal distribution of Potato virus S (PVS) and Potato virus X (PVX) was studied in two trials within each of four commercial fields of seed potato var. Russet Burbank in Tasmania, Australia. In the first trial (plots) 20 leaflets were collected from each of 49 plots (each approximately 8 m wide by 10 m long), with plots arranged in a 7-×-7 lattice. In the second trial (transects), leaflets were collected at 1-m intervals along seven adjacent, 50-m long rows. The mean incidence of PVS increased during the season by 5.2% in one of four plot trials and 25.5% in one of four transect trials. The mean incidence of PVX increased during the season by 10.1%, in one of two transect trials. Spatial Analysis by Distance IndicEs and ordinary runs analysis detected aggregation of PVS infected plants early in the season in one and two fields respectively, suggesting transmission during seed-cutting or during planting. An increase in PVS incidence mid- to late season in one field was associated with aggregation of PVS along, but not across rows, which may be related to the closer plant spacing within rows and hence increased potential for mechanical transmission along rows. Results suggested limited spread of PVS and PVX occurred within crops during the season.
Approximately 26% of the Australian potato (Solanum tuberosum L.) crop, equating to 350,000 metric tonnes, is produced in Tasmania annually (1). The Tasmanian seed potato industry has remained relatively virus free due to geographic isolation and a seed potato certification scheme, which was introduced in the 1930s. However two viruses, Potato virus S (Genus Carlavirus; PVS) and Potato virus X (Genus Potexvirus; PVX), were recently found in seed potato crops in Tasmania. Both viruses are prevalent in potato growing regions around the world (25) and have been reported in other states of Australia (30). In other countries, PVS and PVX have been associated with yield losses of up to 20% (25).
Production of seed potato in Tasmania involves planting of minitubers and growing seed crops over four subsequent seasons or "generations" (G1 to G4) to produce sufficient seed tubers for ware production. Each generation is visually inspected during the season and certified if disease incidence is below set thresholds. However under field conditions, PVS and PVX produce only mild symptoms and because no laboratory testing is conducted for these viruses as part of the certification scheme in Tasmania, virus incidence can increase undetected within each generation of seed potato.
Research in other countries has shown PVS and PVX to be spread predominately by mechanical transmission, e.g., during seed cutting operations (12,17) by contact between plants or from machinery movement within the crop (3,6,14,20). In addition, some strains of PVS are transmitted in a non-persistent manner by some aphid species (29). Weed hosts could also serve as a source of inoculum for PVS and PVX. The host range of PVS includes solanaceous and some non-solanaceous species such as Chenopodium quinoa L. and C. amaranticolor L. (6), while PVX has a limited host range constrained to solanaceous species such as Solanum nigrum L., S. tuberosum L., Nicotiana spp., Petunia hybrida L., Datura stramonium L., Cyphomandra betacea (Cav.) Sendt., and Lycopersicon esculentum Mill. (6).
The objectives of this study were to provide information on the epidemiology of PVS and PVX in Tasmania by (i) quantifying spatial patterns of PVS and PVX infected plants during crop development; (ii) characterizing the temporal progression of PVS and PVX epidemics; (iii) quantifying the spatiotemporal relationships between infected plants at successive time periods; (iv) assessing the role of weeds as alternative virus hosts for PVS and PVX; and (v) determining the role of aphids in transmission of PVS. Information from this study will assist the Tasmanian potato industry in the development of cost-effective strategies for control of these viruses.
Field Sites and Data Collection for Spatial Analyses
Two trials ("plots" and "transects") were established in each of four commercial fields of seed potato var. Russet Burbank over two years. Plot trials consisted of 49 plots, with plots arranged in a 7-×-7 lattice. Twenty leaflets were collected at random from each plot at each sampling time and tested for virus (below). Transect trials consisted of seven adjacent rows, each 50 m long. At each sampling time, one leaflet was collected at 1-m intervals along each row and tested for virus (below).
In 2003, two fields (fields 1 and 2) were located at Riana (UTM coordinate 55G 414798 5434841) in Tasmania. Field 1 (G4) was planted on 10 December 2003 and plots were 10 rows wide (8 m) × 10 m long, with approximately 500 plants per plot. Field 2 (G4) was planted on 20 November and plots were nine rows wide (8.3 m) × 10 m long, with approximately 450 plants per plot. Transect trials were established in each field as described above. Leaflets for virus testing were collected from field 1 at 58 and 129 to 130 days after planting (DAP) and from field 2 at 56 to 58 and 132 to 133 DAP. In 2004, one field (field 3) was again located at Riana and the other (field 4) at Scottsdale (UTM coordinate 55G 541955 5444470) in Tasmania. In field 3 the plots (G3) and transects (G4) were planted on 19 November and field 4 (G3) was planted on 17 November. Plots within field 3 consisted of eight rows (8.3 m) wide × 10 m long, with approximately 480 plants per plot. Plots within field 4 consisted of seven rows (7 m) wide × 10 m long, with approximately 420 plants per plot. Transect trials were established in each field as described above. Leaflets for virus testing were collected from field 3 at 31 and 107 DAP and at field 4 at 30, 54 and 105 DAP. Commercial recommendations were followed for planting, fertilizer application and weed control at all fields. Irrigation water was applied by traveling gun-type irrigator at field 1 and 4 and by solid set sprinklers at fields 2 and 3.
For virus testing, approximately 0.1 g of each leaflet was homogenized in a rotary leaf press in 1.0 ml of 0.01M phosphate-buffered saline (pH 7.4) containing polyvinylpyrrolidone (MW 40,000; 20 g/liter), bovine serum albumin (2.0 g/liter) and Tween 20 (20 ml/liter). Sap extracts (100 μl) were tested for the presence of PVS and PVX by double antibody sandwich immunosorbent assay (DAS-ELISA) (7) using polyclonal antisera (Agdia Inc., Elkart, IN) in 96-well polystyrene microtitre plates (Nunc). Absorbance at 405 nm was recorded for each well using a Titertek photometer (Flow Laboratories, Helsinki, Finland). Samples were considered positive if absorbance was greater than the mean absorbance of the negative controls plus three times the standard deviation of the buffer only (26).
The Score interval method (28) was used to calculate lower and upper 95% binomial confidence limits on the mean virus incidence at each assessment time. No change was detected in the incidence of PVS in plots or transects of fields 1 or 2, with the 95% CI’s around the mean incidence overlapping at each sample time (Table 1). An increase in incidence of PVS was detected in plots (but not transects) of field 3 (5.2%) and in transects (25.5%), but not plots, of field 4 (Table 1). PVX was not detected in plots and transects of field 1 early in the season. Later in the season a trace amount (0.1%) of PVX was detected in plots of field 1, but not in transects (Table 1). Similarly in plots and transects of field 4, no PVX was detected throughout the season (Table 1). In field 2, no increase in PVX was detected in plots or transects, while in field 3, there was no increase in plots, but an increase of 10.1% in the transects (Table 1).
Table 1. Mean incidence (and 95% confidence intervals) of Potato virus S (PVS) and Potato virus X (PVX) in var. Russet Burbank seed potato in plotx and transecty trials from four seed potato fields.
x 980 leaflets and y 357 leaflets tested at each time interval.
z 95% confidence intervals around mean virus incidence calculated by the Score method (28).
Spatial patterns were depicted using geostatistical-based techniques. In transects spatial analysis was conducted by ordinary runs analysis (19) and on virus incidence in plots by Spatial Analysis by Distance IndicEs (SADIE Version 1.22) (23). Analysis was conducted only when virus incidence was greater than 5% and less than 95%.
For ordinary runs analysis a "run" was defined as a succession of one or more like events (i.e., infected or uninfected plants). A non random distribution of infected plants was concluded (P = 0.05) if the Z-statistic calculated according to (19), was less than -1.64. Runs were assessed by joining plants both along and across rows, to test for aggregation in both directions. SADIE has been described previously (21,22,23,24,32). Briefly, SADIE uses a transportation algorithm, to calculate the shortest distances required to move spatially referenced data to obtain both ‘regular’ and ‘crowded’ spatial patterns. The overall distances required for these moves are then summed and compared to random simulations based on re-sampling of the diseased measure locations. All simulations used the maximum number of randomizations, 5967. Deviation of the index of aggregation (Ia), the ratio of the expected and observed distances to regularity, from the null hypothesis of no spatial dependence was assessed by a 1-sided test for aggregation. Values of Ia equal to 1 indicate a random spatial pattern, values less than 1 indicate a regular pattern and values greater than 1, an aggregated pattern.
Temporal associations in spatial patterns between two consecutive sampling times were analysed using the Association Extension of SADIE (Version 1.22) (31). Overall association (X) was calculated as the mean of individual local associations between the clustering indices, which estimate the net distance individuals are required to move to achieve regularity. Significance of X was tested by the maximum number of randomizations of the local association values, allowing for small-scale autocorrelation with the Dutilleul adjustment, using a two-tailed test. The null hypothesis of no association was used (31).
Spatial analysis of PVS incidence by SADIE detected a random spatial pattern of infected plants in plots in fields 1 and 3 and an aggregated spatial pattern at all times in field 4 (Table 2). Ordinary runs analysis of transects demonstrated aggregation of PVS infected plants along (but not across) rows at early and late season in field 1, and aggregation across (but not along) rows at early season in field 3 (Table 2). In field 4, ordinary runs analysis detected aggregation along rows at early, mid and late season and across rows at early and mid-season. The incidence of PVX was sufficient for analysis (> 5% incidence) by SADIE only in plots of field 2 and by ordinary runs analysis in transects of fields 2 and 3. A random distribution of PVX infected plants was detected by SADIE in plots and along and across rows in transects by ordinary runs analysis in fields 2 and 3 at all times (Table 2).
Table 2. Spatial analysis of the incidence of Potato virus S (PVS) and Potato virus X (PVX) in plot and transect trials within four fields of seed potato using Spatial Analysis by Distance IndicEs (SADIE) and ordinary runs analyses for plot and transect trials respectively.
w Ia is the index of aggregation.
x Z-statistic less than -1.64 indicates non random distribution of infected plants either along or across rows.
y ns (not significant), * (P ≤ 0.05), ** (P ≤ 0.01)
z Not analyzed due to virus incidence being too low (<5%) for spatial analysis.
Significant spatial association was detected between the distribution of PVS infected plants at 30 and 54 DAP (X = 0.376; P = 0.006) and 54 and 105 DAP (X = 0.384; P = 0.007) in field 4. However, no significant spatial association was detected in spatial distribution of virus infected plants in plots between time periods at other fields (data not shown).
Prevalence and Incidence of PVS and PVX in Weeds
The prevalence and incidence of both viruses in weeds were assessed at each field. Leaf samples from weeds within plots and at edges of each field were tested for PVS and PVX by DAS-ELISA (described above). Leaflets were collected from fields 1, 3, and 4 at 130, 107 and 105 DAP, respectively, and from field 2 at 77 and 133 DAP.
In field 1, PVS was not detected in samples of Chenopodium album L. (total number of samples, n = 32) or Solanum nigrum L. (n = 13). In field 2, PVS was not detected in samples of C. album (n = 86), S. nigrum (n = 37), Polygonum persicaria L. (n = 28), Rumex acetosella L. (n = 9), Sisymbrium officinale (L.) Scop. (n = 31), Fagopyron esculentum Moench. (n = 2), Trifolium repens L. (n=5), Geranium dissectum L. (n = 1), Carduus teniuflorus (Curt.) (n = 1), Coronopus didymus (L.) Sm. (n = 1), Senecio vulgaris L. (n = 2), Sonchus asper (L.) Hill (n = 2), Rumex crispus L. (n = 1), and Capsella bursa-pastoris (L.) Medik. (n = 3). PVS was detected in 1 of 2 C. album and 2 of 58 S. nigrum collected in field 3. In field 4, PVS was detected in 2 of 53 C. album, 1 of 11 Malva sylvestris L. and 1 of 6 Rumex obtusifolius L. In addition, PVS was not detected in samples of S. nigrum (n = 37), Raphanus raphanistrum L. (n = 1), or Amaranthus powellii S. Wats. (n = 1) in field 4. PVX was not detected in weed samples collected from any of the fields.
To monitor flights of alatae (winged) aphids, two yellow sticky aphid traps (9.5 cm wide × 23.0 cm long) were placed 1 m above ground at the field edge facing into the prevailing wind. Traps were placed in fields 1 to 4 at 80, 70, 31, and 30 DAP respectively and replaced weekly (fields 1 and 2) or fortnightly (fields 3 and 4). Trapping occurred between 3 February 2004 and 7 April 2004 (fields 1 and 2), 20 December 2004 and 7 March 2005 (field 3), and 7 December 2004 and 3 March 2005 (field 4). Traps were stored at 10°C until processed for aphid identification. Aphids were located under a dissecting microscope (50×) and removed from traps by soaking in a dipentine-based solvent (DeSolvit, RCR International, Victoria, Australia). Aphids were identified to species using keys (5) and by comparison with type specimens held in the insect collection at the Department of Primary Industries, Water, New Town, Tasmania.
Small numbers of aphids were trapped, mostly late in the development of the potato crops. In field 1, Aphis gossypii Glover (melon/cotton aphid) occurred on traps retrieved on 10 March (n = 1), 1 April (n = 1), and 7 April 2004 (n = 2) and Macrosiphum euphorbiae Thomas (potato aphid) occurred on traps retrieved on 7 April 2004 (n = 1). In field 2, A. gossypii occurred on traps retrieved on 17 March (n = 2) and 7 April 2004 (n = 28), respectively. No aphids were detected on traps in field 3. In field 4, a single M. euphorbiae was trapped on 3 March 2005.
Limited increase in incidence of PVS and PVX was detected in Tasmanian seed potato crops during the growing seasons. An increase in the incidence of PVS was detected in only one of the four plot trials (5.2% at field 3) and one of the four transect trials (25.5% at field 4). Similarly no significant increase in PVX incidence was detected in any of the three plot trials where PVX was detected and only one of the two transect trials (10.1% at field 3). In those trials in which PVX was not detected early in the season, either none or only trace amounts of PVX were detected late in the season, suggesting no transmission from external sources.
Spatial analyses of PVS incidence in plots in fields 1 and 3 by SADIE suggested a random distribution at planting throughout the season. This distribution was indicative of the planting of infected tubers, with no evidence of further spread during the season. However, aggregation along rows was detected by ordinary runs analysis in field 1, indicating some mechanical transmission of PVS between tubers at planting, as no further increase in incidence of PVS was detected during the season. In field 4, SADIE detected aggregation at all times and ordinary runs analysis detected aggregation along and across rows at early and mid season, but only along rows at late season. Aggregation of infected tubers at planting may result from the planting of seed from particular bins that had been harvested from areas of high virus incidence the previous season, or virus transmission between tubers within particular bins by seed cutting prior to planting or by mechanical transmission during handling and planting. Sprouts can contain high virus concentrations (12) and may be damaged during planting operations, increasing the potential for transmission between tubers at this time. The aggregation of PVS infected plants in transects of field 4 along, but not across rows, later in the season coincided with an increase in virus incidence and suggested mechanical transmission predominately along rows. This may reflect the closer spacing of plants along rows compared to across rows, leading to plants contacting each other along rows earlier in the season and therefore providing more opportunity for mechanical transmission in comparison to plants across rows.
The increase of PVS in transects, but not in plots in field 4 may have resulted from the former having a higher virus incidence and therefore increased opportunity for transmission between plants than the latter. In addition, transects were located at the field edge, while the plots were within the potato field. No traffic occurred along the field edge bordering transects to facilitate virus transmission. However, transects at the field edge may have been more exposed to wind damage than plots, with consequently more opportunity for mechanical transmission. Although aphids can preferentially alight at field edges, leading to a higher virus incidence at edges (9), aphid trapping did not suggest the presence of known aphid vectors at any of the fields.
For PVX, random distributions were detected at all times in field 2 by both spatial analyses methods and in field 3 by ordinary runs analysis. As PVX is transmitted only by mechanical means (6), any increase in incidence might have been expected to occur between adjacent plants, forming an aggregated distribution. Alternative means of transmission of PVX have been documented, such as by chewing insects (4), which might account for the random distribution following spread in transects at field 3.
Limited increase in the incidence of PVS and PVX in Tasmanian fields is unlike the findings of many overseas studies (8,10,11,13,15). However, direct comparison with these studies is difficult. Virus testing methods have improved greatly since some studies were conducted, and agronomic methods have changed considerably and differ between countries. In older trials, there were often considerable mechanical operations through the crop for weed control, moulding of rows, application of pesticides by ground equipment and the use of planting equipment with metal spikes to plant seed pieces. These agronomic operations may have led to significant mechanical virus transmission. By contrast, in the four fields studied in Tasmania, minimal traffic occurred through the crops with no moulding of rows, the use of aerial spraying, the use of irrigation via solid set sprinklers or traveling irrigators with wide laneways and the use of cup-type planters. These agronomic operations may have minimized mechanical virus transmission. In some trials, viruliferous aphids have been implicated as a significant cause of reinfection of seed potato with PVS (15), while in others mechanical transmission has been considered more important (10). In our trials, no known aphid vectors of PVS were detected on traps, and the minimal spread of PVS in all but field 4, suggested aphids did not contribute to transmission.
Although limited studies have been conducted, weeds have not been found to be important sources of PVS (16,27). Serological testing in this study indicated PVS to occur infrequently in some weeds in the fields studied. Several weeds have been shown to be sources of PVX in overseas studies (2,18), however PVX was not detected in weeds in our study. These results suggest that weeds were unlikely to be a major source of inoculum in the particular fields studied. Further evidence for limited external sources of PVS inoculum is gained from the significant spatiotemporal association of PVS infected plants between time periods.
We have been able to demonstrate that limited increase in the incidence of PVS and PVX occurs in seed potato fields in Tasmania, indicating that agronomic practices in Tasmania are effective in minimizing transmission of PVS and PVX during the growing season. Furthermore in the four fields studied, weed hosts appeared unimportant in the epidemiology of PVS and PVX, with no suggestion of aphid transmission of PVS.
This research was supported by the Australian Federal Government through the Department of Agriculture, Fisheries and Forestry (Science and Innovations Award for Young People in Agriculture, Fisheries and Forestry 2004). We also thank seed potato growers for access to their crops, staff of Simplot Australia Ltd., Mr. C. Palmer and Ms. H. Jenson for technical assistance, and Ms. C. Young and Ms. S. Young (Department of Primary Industries, Water and Environment, New Town, Tasmania) for aphid identification.
2. Allen, T. C., and Davis, J. R. 1982. Distribution of tobacco rattle virus and Potato virus X in leaves, roots, and fruits and/or seeds of naturally infected weeds. Am. Pot. J. 59:149-153.
3. Beemster, A. B. R., and de Bokx, J. A. 1987. Survey of properties and symptoms. Pages 84-113 in: Viruses of potato and seed potato production. J. A. de Bokx and J. P. H. van der Want, eds. Pudoc, Wageningen, the Netherlands.
4. Bercks, R. 1970. Potato virus X. C.M.I. AA.B. Descr. Plant Vir. 4.
5. Blackman, R. L., and Eastop, V. F. 2000. Aphids on the World’s Crops. John Wiley and Sons Ltd., UK.
6. Brunt, A. A., and Loebenstein, G. 2001. The main viruses infecting potato crops. Pages 65-133 in: Viruses and Virus-Like Diseases of Potato and Production of Seed-Potatoes. G. Loebenstein, P. H. Berger, A. A. Brunt, and R. H. Lawson, eds. Kluwer Academics Publ., Dordrecht, the Netherlands.
7. Clark, M. F., and Adams, A. N. 1977. Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. J. Gen. Virol. 34:475-483.
8. Cockerham, G. 1958. Observations on the spread of virus X. Pages 144-148 in: Proc. of the Third Conf. on Potato Virus Dis., Lisse-Wageningen.
9. DiFonzo, C. D., Ragsdale, D. W., and Radcliffe, E. B. 1996. Crop borders reduce potato virus Y incidence in seed potato. Ann. Appl. Biol. 129:282-302.
10. Fletcher, J. D. 1984. Levels of virus incidence in pathogen tested and group 1 seed potatoes in New Zealand. Aust. Plant Pathol. 13:33-35.
11. Fletcher, J. D., Lethwaite, S. L., Boddington, H. J., Nott, H. M., and Wood, R. J. 1996. Virus disease surveys of ware potato crops, Franklin County, North Island, New Zealand. N. Z. J. Crop Hort. Sci. 24:7-12.
12. Franc, G. D., and Banttari, E. E. 1984. The transmission of Potato virus S by the cutting knife and retention time of infectious PVS on common surfaces. Am. Pot. J. 61:253-260
13. Franc, G. D., and Banttari, E. E. 1996. Translocation and mechanical spread of a Minnesota isolate of Potato virus S in potatoes. Am. Pot. J. 73:123-133.
14. Franc, G. D., and Banttari, E. E. 2001. Mechanically transmissible viruses of potato. Pages 159-175 in: Viruses and Virus-Like Diseases of Potato and Production of Seed-Potatoes. G. Loebenstein, P. H. Berger, A. A. Brunt and R. H. Lawson, eds. Kluwer Academic Publ., the Netherlands.
15. Kamenikova, L. 1978. The spread of Potato virus S. Ochrana Rostlin 14:101-106
16. Khalil, E. M., and Shalla, T. A. 1982. Detection and spread of Potato virus S. Plant Dis. 66:368-371.
17. Larson, R. H. 1950. The spread of ringspot virus X by cutting knife. Am. Pot. J. 27:53-54.
18. Locatelli, E. A., Allen, T. C., Koepsell, P. A., and Appelby, A. P. 1978. Diagnosis of tobacco rattle virus (TRV) and other viruses in weed and rotation crops in potato fields. Am. Pot. J. 55:249-257.
19. Madden, L. V., Louie, R., Abt, J. J., and Knoke, J. K. 1982. Evaluation of tests for randomness of infected plants. Phytopathology 72:195-198.
20. Manzer, F. E., and Merriam, D. C. 1961. Field transmission of the potato spindle tuber virus and virus X by cultivating and hilling equipment. Am. Pot. J. 38:346-348.
21. Perry, J. N. 1995. Spatial analysis by distance indices. J. Anim. Ecol. 64:303-314.
22. Perry, J. N. 1998. Measures of spatial patterns for counts. Ecology 79:1008-1017.
23. Perry, J. N., Bell, E. D., Smith, R. H., and Woiwod, I. P. 1996. SADIE: software to measure and model spatial pattern. Aspects Appl. Biol. 46:95-102.
24. Perry, J. N., Winder, L., Holland, J. M., and Alston, R. D. 1999. Red-blue plots for detection cluster in count data. Ecol. Lett. 2:106-113.
25. Stevenson, W. R., Loria, R., Franc, G. D., and Weingartner, D. P., eds. 2001. Compendium of Potato Diseases. The American Phytopathological Society, St. Paul, MN.
26. Sutula, C. L., Gillet, J. M., Morrissey, S. M., and Ramsdel, D. C. 1986. Interpreting ELISA data and establishing the positive-negative threshold. Plant Dis. 70:722-726.
27. Thomas, P. E., and Richards, K. 2003. Survival and spread of potato viruses in the Columbia Basin 1: Potential for virus survival in overwintering weeds and crops. Pot. Prog. 3:1-4.
28. Vollset, S. E. 1993. Confidence intervals for binomial proportion. Statist. Med. 12:809-824.
29. Wardrop, E. A., Gray, A. B., Singh, R. P., and Peterson, J. F. 1989. Aphid transmission of Potato virus S. Am. Pot. J. 66:449-459.
30. Wilson, C. R., and Jones, R. A. C. 1990. Virus content of seed potato stocks produced in a unique seed potato production scheme. Ann. Appl. Biol. 116:113-109.
31. Winder, L., Alexander, C. J., Holland, J. M., Woolley, C., and Perry, J. N. 2001. Modelling the spatio-temporal response of predators to transient prey patches in the field. Ecol. Lett. 4:568-576.
32. Xu, X. -M., and Madden, L. V. 2004. Use of SADIE statistics to study spatial dynamics of plant disease epidemics. Plant Pathol. 53:38-49.