| Part 1 - The Iowa State RRD Field Tests |
| Part 2 - Rose Culture Questions Ann asked |
| Part 3 - Summary of Purdy Report: |
| Part 4 - Original Source Material |
| Part 1 - The Iowa State RRD Field Tests by Larry Peck updated July 7, 2002 |
| Rose rosette disease (RRD) is contagious, incurable and deadly to ornamental roses. Results from field trials conducted at Iowa State University are being used to justify intentional spread of this deadly pathogen in several states. The “transmissibility trials” of the RRD Field Tests conducted at Iowa State University were part of a multipurpose test. The statement “Risk to ornamental rose seems to be low” (Epstein et al. 1997) has been widely quoted, by proponents of RRD as a weed control, but the qualifier “under conditions of this study” has widely been overlooked. In point of fact, they didn't say under conditions of this study in some of their own publicity. Epstein is quoted, "No biological control is 100 percent, whereas chemicals can kill non-target plants. RRD works selectively. That's one of the reasons it is so safe" in a 1996 ISU News Releases. |
| Based on this misrepresentation, or misinterpretation, scientists, farmers and managers, have employed RRD as a weed control, no doubt hastening its spread. This poses a major risk to rose culture in the United States - a billion dollar per year industry. |
| Any representation that results reported for these tests can be applied over a wide geographic area, to roses grown using normal cultural practices is untrue. The ornamental roses in this study were grown under very peculiar circumstance giving added importance to the term; "under conditions of this study". This is addressed below because test garden layouts did not in any way emulate a real garden, test roses were not grown under conditions or using cultural methods used on most ornamental roses, pertinent information from previous works was ignored or dismissed and pertinent data from the tests themselves may have been disregarded. What were to have been tests of four field plots were reduced by bad luck or poor planning to a single field test. |
| Lack of a representative sample: No explanation has been given of the criteria used to determine which three roses would represent over twenty four thousand cultivars of roses. This is particularly bothersome for a group of plants as diverse as roses. This already pathetic situation was made worse by continued use of 'Bonica', a rose Dr. Epstein himself concluded “seems to be resistant to mite infection, but is susceptible to graft inoculation (The glossy leaf results from unusually thick cuticle, and the mites may not be able to feed through it)” (Epstein & Hill, 1997). With 'Bonica', they were for all practical purposes throwing out one third of an already minuscule representative sample. By using 'Color Magic', a rose notoriously susceptible to winter die back, they were throwing out another third as addressed below. |
| Health of test plants: No attempt was made to ascertain the state of health of roses before the field tests as it pertains to rose mosaic virus. This would and probably did effect test results, as RMV infected plants are more susceptible to freeze damage and in our experience, are smaller and have greatly stunted root development when even slightly stressed. Inquiries were answered as follows: “Regarding your question concerning rose mosaic, certified roses were used. They have been subject to the normal virus certification process appropriate to the industry. Often the certification process includes indexing of mother plants on indicator plants, and various serological tests such as ELISA. On occasion, PCR can be employed.” |
| The folks at Iowa State seem to believe that “certified roses” is some sort of grading or certification system. Nothing could be further from the truth! It's a brand name. Not only is it a brand name, but CR is a brand that does not enjoy a good reputation among knowledgeable rose growers. Robert B. Martin Jr. in his book, "Showing Good Roses" refers to Tyler, Texas (the home of Certified Roses) as "The Rose Virus Capital of the World" (p.127). Ann and I have never gotten a rose from CR that was not infected by Rose Mosaic Virus. It is unfortunate that the use of probable RMV infected roses casts doubt on results of much of the RRD work done at Iowa State. |
| Plant Density: Risk assessment plots were established as 7.3 meter (23.62 ft.) squares, each containing ten plants of 'Chrysler Imperial', 'Color Magic', 'Bonica' and seedlings of 'R. multiflora' (note: Seedlings can't be infected with rose mosaic virus, because it is not seed transmissable. Thus they will be big healthy plants that can snag a mite). That's 558 sq. feet for 40 roses or 14 sq. ft. per rose. Living on a farm, Ann and I have lots of space for roses, but we don't have unlimited money to build beds, so there is some limit on the area devoted to roses. I went out to determine our plant density, so I picked three typical beds; two beds of large Austins (3.5 ft. x 30 ft. with 19 'Graham Thomas' and 4.25 ft x 24 with 16 assorted Austin Roses) and one bed of Ht's ( 6 ft. x 32 ft. with 32 Ht's). Average density in my garden was 6 sq. ft. per rose. On a trip to Ohio, we visited the garden of a major exhibitor. His plants, mostly HT's, were smaller than our southern roses because they lose canes in winter and were spaced about two feet apart. For a 50 ft. x 4 ft. bed with 50 HT's that's 4 sq. ft. per plant. The field tests used 3.5 times as much land per plant as he used. |
| Why is this important? We found a dozen RRD infected potted roses in a local home improvement store. They were stored pot to pot and had been there less than five weeks. Except for frost conditions for three mornings, May and June temps (2002) had been in the high 80's and 90's and mite action apparently had begun early. I guess they hadn't gotten the word that, as ornamental roses, they were suposed to be at low risk for RRD, but once one plant gets it, plant density plays a big part in spread. If rose growers had to use "conditions of this study" they would need 233 to 350 percent of the land area they currently use, based on densities in the preceeding paragraph. Keep in mind, this is actually a far greater difference than the numbers indicate because our well cared for Austins and Ht's are large plants compared to the small possibly virused ornamental test roses used in the study. |
| Water: What about water? If not on an irrigation system, the roses would get only three inches of rain or even less per month, compared to the equivalent of six to twelve or more in some real rose gardens. The resultant decrease in succulent growth would be important, denying infected mites a food source and thus an opportunity to infect. According to rose expert Bob Martin in his Book "Showing Good Roses" water requirements are as follows: "A mature, full-size hybrid tea in typical Southern California soil requires about 6-9 gallons of water a week when the high temperatures are in the 70s. As temperatures rise into the 80s the rose will require about 9 gallons of water per week. In the 90s, the rose will require about 12 gallons per week and even more. These figures are rough and based on the amount of water needed to maintain the highest level of show quality; the rose will stay alive on considerably less. One gallon of water equals 231 cubic inches by volume. A typical full size rose in Southern California has a root zone with a diameter of about 30 inches. Using my calculator to figure out the pi times the radius squared etc., I have learned that 1-inch of water delivers about 3 gallons. So I deliver this amount 2-5 times a week depending on temperature. " - So, he gives equivilant of 6 to 15 inches a week in his rose garden compared to the Iowa State Test roses getting 3 inches per month of natural rainfall. Which would have more succulent growth available for mite infestation? How valid is the ISU test based just on this? It gets worse. |
| Winter Protection: From the brief description of cultural practices (Epstein et al, 1997:174) we have “the plants were monitored at monthly intervals for symptoms of RRD infection. At the beginning of the 1994 and 1995 growing seasons, any symptomatic plants were replaced with identical non symptomatic cultivars.” No account of winter protection was given so if they were grafted plants, “identical” should mean dead! Dead plants don't get RRD. Giving them the benefit of the doubt that the plants somehow survived winter with or without protection, we come to the question of pruning. |
| Assuming they had winter protection, this would be done in mid-November in central Iowa. One assumes they would have followed the advise of Iowa State University (Nov. 11,1994 issue of Hort. & Home Pest Newsletter pp154-155,“Growing Roses in Iowa - Winter Protection for Roses” by Richard Jauron, Dept of Hort, ISU); “Extremely tall canes can be cut back to 2.5 to 3 feet.”. This is also of importance for RRD. Looking at RRD in garden roses (in real gardens) it is a disease of the canes where a cane has the misfortune to snag an infected mite. That's why big plants like 'R. multiflora' catch it more than small plants. Early after infection, the initial point of infection appears as a small witches' broom in the cane. From there, infection moves slowly within the cane eventually reaching the root from which it infects the entire bush. Early infection on other canes is often caused by mite transmission from the infected cane to adjacent canes rather than from infection reaching the root and moving out the other canes, because infected mites are blown to or dropped onto nearby canes. Migration within the plant itself is slow! That is why we can sometimes save big plants if we catch it early enough. The fact that multiflora is not cut back for winter and its canes don't die back, makes the comparison with ornamental roses totally invalid. |
| With maximum mite migration for Iowa in late August and later (Epstein et al, 1997), infection introduced at or near the ends of long canes did not have enough time to move within canes before they were cut back for winter protection. If there was no winter protection, the canes simply froze to death perhaps before infection was transmitted to the roots. Further south, there is more time for virus sensu lato migration especially where the canes are not cut back in Fall. |
| Bad Rose culture during the season: Having cut them back in Fall for Winter protection the bushes the next Spring would not serve as a wind break to trap infection carrying mites as they would in the South. Larger roses, Austins for example, and cold hardy roses not cut back, like 'R. multiflora', would have been much bigger targets. Roses infected with Rose Mosaic Virus experience more winter die back, resulting in more severe Spring pruning and presenting a further reduced target for infection bearing mites. This is also true of healthy roses that are particularly susceptible to winter die back such as 'Color Magic' - one third of their ornamental sample. |
| Weeding: We have found no reference to weeding of the test beds. The likelihood of a mite getting to one of these small cut back cultivated roses was small, because unless the test plots were weeded and mulched (like a real garden), the roses were surrounded by a protective screen of weeds. The smaller the roses the more protection offered by the weeds. The fact that weeds grow much faster than unwatered rose bushes, means a physical barrier was in place quickly. This physical barrier to the mites could further stunt the roses' growth introducing a multiplier effect. If the weeds screened out 99% of the mites, for example and the lack of succulent growth killed 9 out of 10 mites landing on the bush that would have otherwise been able to feed and infected the rose, then one would have under reported infection rates by (99 x 9) 891 times. Introduce the other factors mentioned here, each as a multiplier, and the magnitude of the under reporting of the infection rate becomes very large. One also assumes the roses were not fertilized (since no fertilizer regime is mentioned). This would also keep the roses smaller and further reduce resemblance to normal pampered garden roses. |
| Deadheading: We have also found no reference to "deadheading", the removal of old blooms (before the plant sets hips) to restart the bloom cycle on ornamental roses. Most rose growers do this to get as many bloom cycles out of their roses as possible. This would result in more new succulent growth each cycle, thus more opportunity for infection. This could not have been done if they only "monitored" the plants once a month. Failure to do this would lead the roses to expend their energy "setting hips" and greatly delay the next bloom cycle. The once blooming 'R. multifloras' don't grow in the same way. Comparing them to reblooming ornamentals is not a valid comparison. |
| One other "condition of this study": The plants were monitored at monthly intervals by a qualified expert on Rose Rosette Disease. Most rose gardens don't have a plant pathologist visit once a month to inspect for RRD and most rose growers have never heard of it. This puts their roses at greater risk of spread within the garden once RRD is established there. |
| Ignoring or Dismissing Previous Works: Reports on this subject out of Iowa State have consistently overlooked or dismissed published reports of RRD infections of cultivated roses. In 1992 they reported RRD "was first reported on "wild rose" in Manitoba in 1941. There have been no further reports of its occurrence in that area but the disease has been endemic in the midwest states for a number of years." |
| Not true! There is no indication that the first report was on a "wild rose". The original source - (Conners, 1941) is included at the end of this chapter so you can judge for yourself. (Also see chapter 6 of this e-book) Even a cursory review of the Canadian Plant Disease Survey would have indicated the problem not only persisted at Morden (Conners, 1942,43) but, "it is destructive and is spreading" (Conners, 1944). Conners' reports tended to be very low key, for him, the above quote is an expression of great alarm. We can document from multiple sources that it remained at Morden at least through 1969! |
| There are two other papers that demand attention and both from states neighboring Iowa. Both papers provide evidence of the RRD susceptibility of cultivated roses that is contrary to the percieved results of Epstein and Hill's field tests; Allington, W.B., Staples, R. and Viehmeyer, Glenn, 1968, Transmission of Rose Rosette Virus by the Eriophyid Mite 'Phyllocoptes fructiphilus', Journal of Economic Entomology, 6:1137-1140 and Crowe, F.J. 1983, Witches' Broom of Rose: A New Outbreak in Several Central States. Plant Disease 67(5)544-546. |
| Allington et all reported on “a rose breeding program that was maintained at the University of Nebraska substation at North Platte. The large nursery, consisted of 4 or 5 acres of rose-breeding stock. By 1959 so many plants in this nursery were afflicted with a condition causing rosetting and virtually eliminating flowering that their breeding program and existence of the nursery were threatened by the necessity of rogueing out so many affected plants.” The results reported in Allington et al. indicate that cultivated roses are susceptible to witches' broom/rose rosette. |
| Crowe, F.J. 1983, Witches' Broom of Rose: A New Outbreak in Several Central States. Plant Disease 67(5)544-546 said. “All previous reports emphasize the rarity of infection of cultivated rose hybrids, especially in urban settings. Recently, however, a large number of roses with these same witches' broom symptoms have been observed in Kansas and some nearby states.” “In surveys in 1980,1981, and 1982, I found a real increase in incidence of the disease rather than simply a greater awareness of a long standing, heretofore unrecognized problem. ...By September 1982, as many as 35% of the cultivated hybrid roses in a few plantings had become infected and witches' broom was found in numerous gardens where no plants have been affected in 1981. "In our current situation , cultivated hybrid garden roses in highly urban areas have become infected, even in the absence of adjacent wild roses". Only recently (Rohozinski,2001) was this paper given it's due. If one accepts the hypothesis presented in Chapter 9, Crowe's work is even more important because it could be used establish a natural rate of spread for RRD from North Platte, compared to intentional spread. |
| Epstein and Hill (1998:17) commented on the above work in American Rose Magazine, “Unfortunately, earlier reports were based on observations made during full-blown epidemic outbreaks of the disease, and this was presented as the typical pattern of how rose rosette disease develops. These views have been shown to be erroneous.” |
| Shown by what? The field tests mentioned above? Perhaps one can quibble about the definition of “full-blown epidemic” as opposed to a partly-blown epidemic, but the so-called "erroneous" papers have stood the test of time. Crowe's paper described what we have seen in East Tennessee, are hearing about in Missouri and now starting to see in Georgia and Virginia. |
| In addition to the above works, they simply ignore the report by Ken G. Purdy to the Missouri Department of Conservation entitled "1981 Multiflora Rose Control Pilot Program Evaluation" which tested farmer's perceptions of the Multiflora problem Vs the reality. In the summary of his report he states: “Results of this survey indicated the average farmer participating in the Department's multiflora rose control pilot program believed the majority of land on his farm was infested by multiflora rose. Severity of infestation generally was perceived as slight to moderate. Infestation estimates by the survey evaluator indicated, however, that the majority of land inspected was not infested by multiflora rose.” |
| The Iowa State papers give percentages of multiflora coverage, but don't address sampling technique. Perhaps a grid was put over an aerial photo, but the numbers in the papers compared to the pictures, seemed to me that the Field Test authors were also greatly overstating the multiflora problem. The Purdy report also stated: “Although survey respondents generally indicated an intense dislike for multiflora rose and appeared anxious to control or eliminate it on most farm areas, they nevertheless suggested that rose presented a problem of only moderate importance to the overall farm operation. The evaluator assessed the rose problem to be less significant than did respondents.” |
| “The sincerity of the landowner's concern about multiflora rose was evident to the evaluator, but the overall problem presented by rose did not appear to be significantly greater than that of other woody plant invaders, such as honey locust and red cedar. Furthermore, survey results indicated that multiflora rose infestations could be controlled by the same management used to control other woody plant invaders.” |
| Does Missouri simply have better farmers than Iowa? I doubt it. It's more likely the Iowa farmers are also overstating the problem. Ann and I drove completely across southern Iowa on US Route 34 in June of 2001. The multiflora problem was not great, compared to what we have in the Valley and Ridge of the eastern USA. Ann saw multiflora on two west facing hill slopes and on NO flat land. On a trip to Clifty Falls State Park in Indiana, we visited the County Agent for Jefferson Co., Indiana, and asked him about his multiflora problem. It was way down his list of problems and even down his list of undesirable invasive alien plants. While at Clifty Falls, the site of some important studies of the NATURAL SPREAD of RRD, we learned something which pertains to the Iowa State Field Tests - Deer love RRD infected roses. |
| Ignoring or Dismissing One's Own Data: The reason I mention deer liking RRD infected rose is the fact that one of their risk assessment studies was reportedly completely wiped out by deer (Epstein, Abraham & Anderson, Susan, 7/15/ 96, "Searching for Biological Control of Multiflora Rose, a press release, available 10/3/2001 at web site http://www.ag.iastate.edu/aginfo/news/past/rose.html). That would lead me to believe, based on what I've seen at Clifty Falls, that the beds were heavily infected with RRD at the time (or the deer were starving). Additionally, deer's preference for the soft succulent growth of RRD infected roses can result in underreporting of infection due to removal of the soft symptomatic parts of the plants. A later Iowa State study incorporated deer fences. Two additional study sites were wiped out by floods which makes the reported results resemble a salvage operation more than a complete study. |
| Iowa State workers reported “Increase in mite populations seems strongly influenced by the occurrence of extended periods of temperatures in excess of 28 degrees C during May, June, July and August.” (Epstein and Hill). That temperature (about 85F) is commonly reached in areas downwind of Iowa, and where there are roses on which airborne Iowa mites can land and infect. They also state that "Over wintering of the mite victor is sporadic in Iowa." This too is not true of areas to the south of Iowa where infection is spread more effectively. Iowa is not an island. The release of a deadly, infectious pathogen that would spread into other areas down wind without first testing it under those conditions was inexcusable. |
| Conclusion: Hort farm conditions are not appropriate for tests on ornamental roses. An experiment is supposed to emulate "real world conditions" and these tests did not. The “transmissibility trials” of the RRD Field Tests conducted at Iowa State University in the 1990's were an ancillary part of a multipurpose test, the main purpose of which was to test “augmentation" of RRD as a weed control. The phrase “Risk to ornamental rose seems to be low” was widely repeated by Dr. Epstein and has been widely quoted. The phrase “under conditions of this study” was not as widely repeated, especially in second hand reports. This had the effect of misrepresenting the study as a legitimate risk assessment that would apply to a wide variety of ornamental roses under natural conditions. |
| Unfortunately, professionals all over the country have heard about the tests and assumed that the results are applicable over a wide geographic area with little chance of the disease moving to cultivated roses. (See the Maryland Report from Plant Disease in Chapter 7, part 2). They are wrong. |
| My wife and I have seen almost a thousand, infected ornamental roses in sixteen states and provinces and heard of hundreds more in these and other states in the last three and a half years. The vast majority of cases of RRD are never reported and worse yet, never diagnosed. When results of a field test do not match the real world, it is reasonable to assume the test, rather than the world, is wrong. |
| The safe distance prescribed for ornamental roses as a result of the above field tests whether it is 100 feet ( ), 100 yards ( ), 100 meters (E&H, 1997) , one forth of a mile (E&H, Feb. 98, Am Rose), half a mile (Proceedings of the Symposium Planning for a Sustainable Future The Case of the North American Great Plains 1995) or 20 meters ( ) - all used at various times and attributed to the same tests, fly in the face of logic. Think about it for a minute. Dust particles, much larger than phyllocoptes fructiphilus were carried from the great plains to New York City during the Dust Bowl Days of the 1930's. That's over a thousand miles and pictures can be found in almost any freshman geology book. This past year, there was great excitement as dust storm debris from China was tracked by satellite across California (Darmenova, 2002). A recent article (by David Ballingrud, St Petersburg Times, 6/14/01) described Sahara Desert dust carrying possible pathogens to Florida. Another recent article detailed how Sahara sands were harming the sky slopes in Switzerland (Henley, The Gardian, 4/11/02). |
| How does one reconcile these and thousands of other examples of aeolian (wind) transport of particles over vast distance (both in and out of the jet stream) with the assertion that an infected mite won't blow onto an uninfected rose more than _____ (fill in the blank using the safe distance they were using that day) away? It simply doesn't make since! |
| I don't doubt the Iowa State data. I doubt the conclusions drawn from their data. Perhaps, rather than demonstrating risk, they demonstrated that when unwatered and neglected, a wild rose well suited for that climate and soil performs better than unwatered, weed encrusted and neglected roses that were bred to be pampered in California. Or, that the rose that grows bigger and has more succulent growth and least winter die back (multiflora or any other big winter hardy rose) stands a better chance of snagging a mite and infection. To me, the tests demonstrate that spread of RRD within an infected garden are greatly influenced by plant density (thus their "lapse rate") as well as other factors addressed above. They do nothing to address the initial infection that occurs in nature when the first mite happens to land on a rose bush, and it is that infection rate, miles away, that augmentation unquestionably greatly increases. The field tests conducted at Iowa State, are of very limited applicability, but the many constraints on their practicality have never been addressed. |
| ************************************************* |
| Were the HTs grafted plants? |
| Were they spring planted? |
| Was the soil amended specifically for the roses? |
| Were they on an irrigation system? |
| If so, how many gallons per week and for what time, spring to fall, summer only? |
| Were they fertilized ? |
| If so, with what N-P-K and how often? |
| Were they sprayed for blackspot? Insects? |
| Were they deadheaded? |
| What winter protection was used, when was it applied? |
| Was there a fall pruning before winter, or were winter killed canes removed in spring? If so, when? |
| Was any pruning done in mid summer? |
| Were the roses mulched through summer & weeded, or were they grown in field conditions ? |
| What were last and first frost dates at the test farms those years? The effective killing temps for roses without Rose Mosaic Virus is assumed to be the low 20s-those dates would be helpful as well. |
| Did you use virus (rose mosaic) free roses? If so, how did you determine they were RMV free? |
| What criteria did you use to select the roses to include in your field study? |
| The questions were never answered. She then contacted Dr. John Hill with the same questions, and except for confirming, in a very round about way, that they did nothing to protect against using virused test plants, he didn't answer the questions either. Most of these are simple yes-no questions. Most scientists writing plant pathology papers go through great pains to include such information in their papers so there can at least be an attempt at reproducing results. We are surprised that we have been unable to obtain cultural information and we were told not to expect reproducability of results from field tests. It was only after she asked the above questions and received no answers that I decided to post this chapter. If they choose to answer the questions, we will be glad to add their answers, as we really want to know more about their growing conditions. The above work is based on the best information (see Bibiliography) we could obtain. |
| ******************************************************************************************** |
| Landowners indicated the greatest utilization problem resulted from rose infestations in pastures. Rose in pastures received the greatest amount of Tordon treatment. Wooded areas and draws generally were believed to be infested although treatment of these areas was limited. Hayfields were least affected by rose, indicating that annual mowing will keep the problem to a minimum. However, those landowners who mowed pastures and hayfields annually did not perceive rose as any less of a problem than those who mowed infrequently or not at all. |
| The majority of landowners appeared to have made an adequate attempt to control their rose problem and seemed to have experienced successful multiflora rose treatment with Tordon. However, variable results were reported by several of the landowners. Individuals experiencing these variable treatment results suggested that improper placement of the chemical or use of inadequate amounts was responsible for the decreased effectiveness of the treatment. |
| Slightly over one-third of the respondents reported off-site movement of Tordon. Evaluator observation of this off-site movement indicated that most plants affected were damaged due to probable extension of their root zones into the treatment site. Non-target plant treatment frequently was reported by farmers and observed by the evaluator. Most of this non-target treatment appeared to be the result of experimentation and was directed primarily towards plants perceived as undesirable to the farming operation, e.g. honey locust, hawthorne, and prickly pear cactus. |
| Overall, destruction of wildlife habitat in treatment areas was limited, however, results indicated a slight decrease in the quality of wildlife habitats viewed due to the destruction of multiflora rose. The most notable amounts of wildlife habitat destruction were observed in draws; farm areas which commonly provide valuable sources of food and cover for wildlife. This observation suggested that a rose control program of increased scope would have a substantial negative impact on the quality of wildlife habitats in the Ozarks. |
| Although survey respondents generally indicated an intense dislike for multiflora rose and appeared |
| anxious to control or eliminate it on most farm areas, they neverthess suggested that rose presented a problem of only moderate importance to the overall farm operation. The evaluator assessed the rose problem to be less significant than did respondents. |
| The sincerity of the landowner's concern about multiflora rose was evident to the evaluator, but the |
| overall problem presented by rose did not appear to be significantly greater than that of other woody |
| plant invaders, such as honey locust and red cedar. Furthermore, survey results indicated that multiflora rose infestations could be controlled by the same management used to control other woody plant invaders. |
| ********************** |
| Part 4 - Origional Source Material THE FIRST REPORT |
| Twentieth Annual Report of the Canadian Plant Disease Survey 1940 |
| Below is page 98 of the Twentieth Annual Report of the Canadian Plant Disease Survey 1940 - the first known reference to what is now called Rose Rosette. Please note, it doesn't tell what kind of rose was infected. |
| 98 Rose |
| were covered with sori, while this year’s were clean to all appearances. |
| Rust was common on Rosa rugosa in a bed at Ste. Anne do la Poca— |
| tiere, Que. Rust (P. americanum) was moderate to severe on hybrid roses at |
| Kentville, N.S. (R. J. Baylis). Rust was occasionally severe in Queens Co. |
| P.E.I. |
| Powdery Mildew (Sphaerotheca pannosa) was quite prevalent on a few |
| susceptible varieties at the Station, Summerland, B.C.; most varieties were |
| free from infection. The disease was moderate on Etoile de France at Winnipeg, |
| Man. Powdery mildew was very prevalent and destructive particularly to cer- |
| tain Rambler varieties in many parts of Oint. The disease was observed |
| causing slight damage in Three Rivers, Two Mountains, Terrebonne, and Ber— |
| thier counties, Que. Powdery mildew was quite common and destructive in |
| P.E.I., particularly on Crimson Rambler, Hybrid Perpetuals and Dorothy |
| Perkins; dusting with extremely fine sulphur has given good results. |
| Witches’ Broom (?virus) was observed affecting some canes at |
| Morden, Man. the number of spines was greatly increased on affected canes. |
| RUDBECKIA — Golden Glow |
| Powdery Mildew (Erysiphe cichoracearum) was reported at Summer— |
| land, B.C. |
| SCLIDAGO - Goldenrod |
| Rust (Coleosporium Solidaginis) severely infected and destroyed |
| the leaves of some ornamental goldenrods at Morden, Man. |
| SYRIWGA — Lilac |
| Powdery Mildew (Microsphaera Alni) caused slight damage in gardens |
| at Fredericton, N. B. |
| Blight (Phvtophthora Svringae) slightly infected leaves and more |
| particularly young shoots as they emerged from the ground near the main |
| branches at Ste. Anne de la Pocatiere, Que. This is the first time the |
| disease has been observed at Ste. Anne. The wet spring was probably very |
| favourable for the disease. (R. 0. Laohance) |
| Mosaic (?virus). About 2/ of the lilac bushes at the Station, |
| Fredenicton, N.B. show a definite mosaic. A diffuse veinal mottle is present |
| in most leaves of the affected bushes. (D. J. MecLeod) |
| TAGETES - Marigold |
| Yellows (virus). Scattered plants of Crown of Gold were affected |
| at Morden, Man. |
| Top of Page |