INTRODUCTION
The term weedy rice generally includes all the species of genus Oryza which behave as rice and which crop in rotation with rice weeds. Weedy rice populations have been reported in many rice- growing areas in the world where the crop is directly seeded (Parker and Dean, 1976; Ferrero and Finassi, 1995). Even though weedy rice belongs to different species and subspecies, all these plants share the ability to disseminate their grains before rice harvesting. Weedy plants can also adapt to a wide range of environmental conditions. Weedy rice grains frequently have a red pigmented pericarp and it is for this reason that the term red rice is commonly adopted in international literature to identify these wild plants. This term, however, does not seem very appropriate as red-coat grains are also present in some cultivated varieties, but also absent in various weedy forms (FAO, 1999).
In most rice areas the spread of weedy rice became significant mainly after the shift from rice transplanting to direct seeding, and has started to become very severe over the last 15 years, particularly in European countries, after the cultivation of weak, semi-dwarf indica-type rice varieties (Tarditi and Vercesi, 1993). The spread has generally been favoured by the planting of commercial rice seeds that contain grains of the weed.
Weedy rice infestations are reported for 40-75 percent of the rice area in European countries (personal communication), 40 percent in Brazil (De Souza, 1989), 55 percent in Senegal (Diallo, 1999), 80 percent in Cuba (Garcia de la Osa and Rivero, 1999) and 60 percent in Costa Rica (Fletes, 1999).
ORIGIN AND DESCRIPTION
The phylogenetic origin of the weedy forms is closely related to that of cultivated rice. Many weedy plants share most of the features of the two cultivated species Oryza sativa and O. glaberrima (Khush, 1997). O. sativa, which is also known as Asian, comprises the varietal groups indica, japonica and javanica, and is grown worldwide (Olofsdotter, 1999). O. glaberrima is also named African rice and is mainly cultivated in West Africa. The genus Oryza includes more than 20 wild species most of which are diploid. Based on the morphological, physiological, biochemical features and crossing relationships, eight different genomes have been identified in the genus Oryza (Aggarval et al. 1997).
Wild species like O. perennis, O. nivara, O. rufipogon and O. longistaminata share the same genome and can easily be crossed with the cultivated O. sativa species (Olofsdotter, 1999). The wild O. barthii species (= O. breviligulata) is considered to be the progenitor after mutation of African rice.
O. glumaepatula is a wild endemic Central and South American species that is conventionally considered to be a subtype of O. rufipogon, but, according to recent genetic analysis, it has been determined to be closer to African forms.
In addition to these species, also O. latifolia, O. punctata, O. officinalis (perennial) and Zizianiopsis miliacea (perennial) are also problem weeds sometimes, both in cultivated rice, along the edges of reservoirs, or in watercourses where they impede the water flow. O. latifolia is a weedy rice species that is widespread in Central America where it is normally named Arrozon or Arroz pato (Castro Espitia, 1999). It has a height of about 2 m and produces seeds with a white pericarp.
In areas where rice differentiation has occurred, several species of Oryza are frequently present together in the spontaneous flora; while in areas where the crop has been imported only O. sativa plants are weed problems.
Weedy plants show a wide variability of anatomical, biological and physiological features (Craigmiles, 1978; Kwon et al. 1992; Tang et al. 1997, Vaughan et al. 2001). A study carried out on 26 Uruguayan weedy accessions revealed two main groups of samples. One group included plants with a black hull, purple apex and long awn, showing evident wild traits, while the other group had straw hull and apex and no awn mimicking cultivated varieties (Federici et al. 2001).
At seedling stage, weedy plants are difficult to distinguish from the crop (Hoaghland and Paul, 1978), while the identification of the weed is possible after tillering, thanks to many gross morphological differences with the rice varieties: more numerous, longer and more slender tillers, leaves which are often hispid on both surfaces, tall plants, pigmentation of several plant parts, easy seed dispersal after their formation in the panicle (Diarra et al. 1985a; Coppo and Sarasso, 1990; Kwon et al. 1992, Suh et al. 1997).
The seeds of most weedy biotypes of O. sativa and O. glaberrima have a pigmented pericarp resulting from the presence of a variable content of different antocyanins, cathekins and cathekolic tannins (Baldi, 1971).
The red pigmentation is a dominant character and is controlled by more than one gene (Leitao et al. 1972; Wirjahardja et al. 1983)
The red layer of the weed grains harvested with the crop should be removed with an extra milling but this operation results in broken grains and grade reduction (Smith, 1981; Diarra et al. 1985a, 1985b).
Weedy biotypes of O. sativa have been differentiated into indica or japonica types, on the basis of the morphological and physiological traits, isozymes, RFLP (Restriction Fragment Length Polymorphism), RAPD (Random Amplified Polymorphic DNA) and AFLP (Amplied Fragment Length Polymorphism) markers.
According to a study funded by the European Community, weeds collected in Mediterranean rice fields belonging to the japonica group and weeds from Brazil were close to the indica group (Ghesquière, 1999). In this study no specific allele of weeds were found which can serve as a diagnostic marker to easily determine the varietal origin of the weedy forms. Nevertheless, a great deal of evidence would seem to show that the primary origin of red rice can come from distant crosses between indica and japonica varieties.
Vaughan et al. (2001) pointed out that the several samples of weedy biotypes collected in the United States belong not only to the indica and japonica subspecies, but also to the O. rufipogon and O. nivara species.
BIOLOGY
Dormancy and seed longevity
Unlike cultivated varieties, weedy rice seeds show a variable degree of dormancy. The duration of the dormancy varies according to the biotype and the storage conditions of the seeds after shattering. The length of the dormancy has been investigated in several countries in natural conditions. O. punctata was dormant for more than one year in Swaziland (Armstrong, 1968) and up to five years in East Africa (Majisu, 1970). Viable weedy rice seeds with red pericarp, remained dormant for up to two years in the United States (Klosterboer, 1978) and three years in Brazil (Leitao et al. 1972).
Environmental conditions during seed formation, moisture and the storage temperature are considered to be the main factors that can affect the length of dormancy (Delatorre, 1999; Leopold et al. 1988; Ferrero, 1984). According to Leopold et al. (1988) the weedy seeds of the strawhulled biotype (straw colour pericarp), kept at -15°C, showed a variable duration of the dormancy in relation to the humidity content of the seeds after ripening. The breaking of dormancy was quicker at grain humidity ranging from 6-14 percent and very low at water content lower than 5 percent or higher than 18 percent.
A significant reduction of the dormancy usually already occurs two months after ripening (Cohn and Hughes, 1981). Even though numerous studies have been carried out at a biochemical level, to define the physiological and genetic bases of dormancy (Footitt and Cohn, 1995; Cohn, 1996; Delatorre, 1999; Cai and Morishima, 2000), the mechanisms of the onset and breaking of this phenomenon have not yet been completely clarified. Dormancy regulation can most likely be attributed to factors present in glumella and in embryo. (Delatorre, 1999).
Dehulled seeds which were stored at -15°C maintained their dormancy and were able to germinate when kept at 5°C. (Cohn and Hughes, 1981). The breaking of weedy rice dormancy obtained with substances such as sodium nitrite, propionic acid, propionate-methyl, cytokinin, n-propanol resulted to be usually accompanied by a pH reduction of the embryo tissues (Footitt and Cohn, 1992).
Seed longevity has been investigated in several studies that led to contrasting results. In a research carried out in the United States, seeds from different populations of weedy rice remained vital by 90 percent after two years of burial and up to 20 percent after seven years (Goss and Brown, 1939).
According to Diarra et al. (1985a) the longevity of the weedy rice seeds can last for up to 12 years.
In a study conducted in Italy the viability of weedy rice seeds taken at a depth by ploughing in loamy soil decreased to six percent after one year and to five percent after two years of burial (Ferrero and Vidotto, 1998a). The non-viable seeds appeared empty, without embryos and reserve matter. It was presumed that most seeds can germinate under favourable environmental conditions (high temperature and oxygen content following tillage operations), but cannot emerge from the soil. The germination percentage of the viable seeds varied in time, decreasing from 91 percent at the beginning of the experiment to 73 percent after one or two years of burial. This behaviour was explained by the fact that many of the seeds that were dormant at the beginning of the experiment did not germinate and remained dormant for over two years. The seeds dug up after one year required, on average, less time to germinate than those buried for two years.
Emergence
The emergence of weedy rice is greatly influenced by the soil texture, the presence of water on the field and the depth of seed burial, which in turn is strictly related to the tillage that has been adopted for seedbed preparation (Ferrero and Finassi, 1995; Ferrero and Vidotto, 1997a; Saldain et al. 1996; Gealy et al. 2000).
Katayama (1969), categorized 20 Oryza species into five groups, according to their behaviour during germination, as follows:
1. Which included O. stapfii and O. subulata, with germination occurring within two and three days.
2. Which included O. barthii, O. minuta, O. latifolia, O. punctata, with germination peaking at six days and total germination greater than 50 percent.
3. Which included two species with behaviour similar to that of group two, but germination less than 33 percent.
4. Which included O. perennis, O. officinalis and other four species, with germination occurring up to nine days and greater than 60 percent.
5. Including O. breviligulata, with germination similar to that of group four but less than 50 percent.
The seedlings that emerge before rice planting are the main contributors to the total emergence from the seed bank. Almost all the plants freely grown in undisturbed soil are able to emerge from mid-April to mid-May, after reaching an accumulation of 200 growing day-degrees (Ferrero et al. 1996) (Figure. 1). The minimum temperature for weedy rice germination is considered to be 10 °C, the same as that of cultivated varieties.
If harrowing and ploughing are taken into consideration, the emergence of red rice seedlings in relation to the 0-10 cm seed bank is remarkably influenced by the type of tillage. On average, the emergence percentages in harrowed and ploughed plots is 7.2 percent and 2.5 percent, respectively (Ferrero and Vidotto, 1999) (Figure 2). These different values of emergence are most likely owing to the movement of the red rice seeds in the soil and this is determined by ploughing. Inversion of the top soil layer buries newly-shed seeds and stimulates their dormancy. At the same time, ploughing brings up the seeds buried in the previous season near to the soil surface, but many of these seeds lose their ability to germinate.
Figure 1. Emergence of weedy rice in an undisturbed soil in relation to the growing degrees (from Ferrero et al. 1996)
Seed age, burial depth, flooding conditions and heavy soil all have a negative influence on weed germination and emergence (Eastin, 1978; Ferrero and Finassi, 1995).
In clayey soil, germination of weedy rice grains situated in the upper 5 cm layer and covered with 6-8 cm of water was, on average, less than one-third that recorded in the same soil but which is only kept moist (Vidotto and Ferrero, 2000).
In contrast with the cultivated varieties, weedy rices are not able to emerge in field conditions just after shattering during autumn, though temperatures are favourable for germination. According to the studies carried out in Mediterranean conditions (Vidotto and Ferrero, 2000) the time required for red rice to germinate depends on the storage conditions and proves to be inversely related to the storage duration. Shattered seeds required at least 70 days in favourable temperature and moisture conditions before germination started. In the same studies, weedy rice seeds placed in the upper soil layer (0-1 cm) showed an 80-90 percent germination in clayey and loamy soils, respectively, when the soil was kept continuously moist, and between 60 and 80 percent when submerged below 2-3 cm of water. An emergence reduction resulted with an increase in depth for both the moist and flooded soils. The emergence of the seeds placed at 4-5 cm depth was between 20-40 percent (in heavy and loamy soils, respectively) in the moist soil and between 5-20 percent in the flooded soil. In both soils, no emergence occurred from the seeds buried at more than 10 cm. The seeds that are not able to emerge are mostly in a position to germinate but not to form viable seedlings. Emergence from the 0-1 cm layer was completed in 14 days in the moist soil and in 18 days in the flooded soil (Ferrero and Finassi, 1995). The seeds buried at a depth of 4-5 cm showed a delay in germination of 15 days in comparison to the seeds that were placed near the soil surface. This behaviour could be one of the reasons for the continuous emergence of these weeds in rice fields.
Figure 2. Red rice emergence from a seed bank in relation to the soil tillage for seed bed preparation (from Ferrero and Vidotto, 1999).
Flowering
In wild and cultivated varieties, flowering begins in the top florets of the panicle and proceeds downwards to the lowest florets (Roy, 1921). In weedy rice florets opening begins between 08: 00 and 09: 00 am and continues at least one hour longer than that of the cultivated varieties. For this reason even though all rice species are self-pollinated plants, cross-pollination is higher in weedy rice than in cultivated varieties. The likelihood of out-crossing of red-coat weedy plants with cultivated varieties was investigated by Langevin et al. (1990). The percentage of crossing ranged between 1.08 percent in the Lemont variety and 52.18 percent in the Nortai variety. The high degree of hybridization with Nortai has been attributed to the prolonged duration of the floret opening in this variety. Due to heterosys, hybrids were generally taller and more vigorous and began flowering 20-30 days later than the parent weedy plants.
Flowering is induced by the length of the day (short photoperiods enhance flowering) the plants age (higher for younger plants) the biotype (higher in biotypes from higher latitudes) (Katayama, 1974).
The pigmentation of the hull and seed starts developing in the terminal spikelet a few days after anthesis and continues in the remainder of the spikelet as it matures (Holm et al. 1997).
Shattering
Early seed shattering is a specific characteristic of weedy rice. This behaviour is controlled by the gene Sh which shows the shattering character in conditions of dominant homozygosys (Sh Sh) or heterozygosys (sh Sh) (Sastry and Seetharaman, 1973). The seed drop results from a formation of an abscission tissue formed by three layers of cells between the spikelet and the pedicel (Nagao and Takahashi, 1963). This layer of cells is not fully formed in cultivated varieties and bands of lignified tissue provide the bind of the spikelet to the pedicel.
Ferrero and Vidotto, (1998a) found that seed shattering in weedy rice started nine after flowering and increased gradually for 30 days until complete development of the panicle (Figure 3). At this stage shattering concerned 65 percent of the total grains and did not appear to be remarkably influenced by different N supplies.
Shattered and non-shattered seeds, considered on the whole, started to become viable at about nine days from the beginning of flowering, with a germinability of about 20 percent. This value increased quickly and had already reached about 85 percent at 12 days after flowering. In general, the shattered grains showed a lower germinability until 24 days after flowering, in comparison to that of non-shattered seeds. From this time on, the germinative capacity of the two groups of seeds was different. The germinability of the shattered seeds was very low during the first 15 days after flowering, with a maximum value of about 5 percent. This behaviour can most likely be explained by the incomplete development of the early shattered grains, which broke off mainly because of environmental causes (wind). The seeds that shattered after 15 days from flowering contained nearly filled and physiologically mature grains.
Figure 3. A. Shattered seed percentage of total seeds ( observed data; - fitted function) and filled seed percentage of shattered seeds (O observed data).
Figure 3. B. Germinability evolution of the shattered () and non-shattered () seeds and mean germinability of filled seeds (). (from Ferrero and Vidotto, 1998a)
Competitive ability
Weedy rice can cause severe yield losses to cultivated rice in relation to the density, type of weedy plants and cultivated varieties (Diarra et al. 1985a; Diarra et al. 1985b; Fisher and Ramirez, 1993; Eleftherohorinos et al. 2002). Short varieties are usually more susceptible to weedy rice competition than tall ones (Kwon et al. 1991a). Several studies have been carried out to assess the effects of different weed densities. With 11 weedy rice plants m-2, Abud, (1989) observed a yield loss of about 43 percent. In studies conducted in Arkansas, the yield of semidwarf cultivar Lemont was already affected at a weed density as low as two plants m-2 (Kwon et al. 1991b). Five and 20 plants m-2 of weedy rice caused a yield loss of 40 and 60 percent, respectively, in cultivar Oryzica 1 (Fischer and Ramirez, 1993). Some studies pointed out that competition effects are also closely related to the interference duration (Kwon et al. 1991a). Combining the effects of weedy rice density and duration of competition, Fischer and Ramirez (1993) observed a yield reduction of 50 percent when 24 weedy rice plants m-2 competed with the crop during the first 40 days after emergence. With the same initial density, the yield loss reached 75 percent in the case of season-long competition. In a greenhouse experiment, significant effects on rice plant growth were recorded only when the competition was longer than 70 days, starting from the emergence (Estorninos et al. 2000). In studies of competition using cultivar Mars, intervarietal competition resulted as being important than intravarietal competition, with the weedy rice acting as the dominant competitor (Pantone and Baker, 1991a; Pantone and Baker, 1991b). Considering the yield components, the effect of plant density seems to be significant on the number of rice panicles per plant and florets per panicle, while the percent of filled florets and the grain weight do not seem to be influenced by this parameter (Pantone et al. 1992). Eleftherohorinos et al. (2002) pointed out that interference between rice and weedy rice began three weeks after rice emergence, but was not affected by an increase of the nitrogen rate from 100 to 150 kg ha-1. According to this study a density of 40 weedy rice plants m-2 resulted in a reduction of 46 and 58 percent in Ariete and Thaibonnet rice varieties, respectively.
CONTROL
Weedy rice control methods that can be applied in rice crops are expensive, time-consuming and usually do not lead to a total eradication of the weed infestation. Incomplete control of the weed for a given year could lead to eliminating the results of several years of good control. Weedy rice escapes of 5 percent or less can produce enough seeds to restore original soil seed bank population levels (Goss and Brown, 1939; Rao and Harger, 1981).
Control of weedy rice plants is much more difficult than that carried out on other weeds because of the great morphological variability, particular growth behaviour, and high biological affinity with cultivated varieties. Chemical weed control with herbicides selective to rice is usually not effective on the weedy forms, with the only exception of transgenic varieties that have appropriately been transformed to tolerate herbicides which are selective to cultivated rice and with a wide spectrum of activity. For this reason chemical weed control cannot be applied to cultivated rice during its growth unless the herbicides are applied with wick or wipe spraying systems in combination with short varieties. Control is also complicated by spaced-out germination over a long period of rice-growth. The high elasticity of the germination process can favour competition activity of weedy plants that are able to germinate earlier than cultivated plants, or allow the weed to escape the control treatments carried out in rice pre-planting (Ferrero and Vidotto, 1997)
Weedy rice control can be managed by applying preventative, cultural, mechanical and genetic practices (Table 1).
Prevention
Prevention is the basic means of reducing weed infestation and can be obtained mainly by planting seed that is free from weedy rice grains. This measure is however not easy to apply as hulled grains are similar to those of cultivated varieties regardless of the colour of the pericarp. Both white and red pigmented grains are difficult to recognise, as the colour of the pericarp can only be detected after hulling. The only possibility of obtaining rice seed that is free from weedy rice grains is that of inspecting the fields where seed production is carried out and removing the weedy plants before harvesting (even manually).
Another important preventative system is that of the accurate cleaning of the equipment that is used during rice harvesting, to avoid spreading of the weed to non-infested fields.
Table 1. Main weedy rice control strategies and methods
Control strategy |
Control method |
Preventative |
Certified seed |
|
Cleaning of machinery |
Cultural |
Rotation |
|
Soil tillage |
|
Stale seed bed preparation |
|
Water management |
|
Rice variety |
|
Hand weeding |
Mechanical |
Before rice planting |
|
After rice planting |
Chemical |
Before rice planting |
|
After rice planting |
Genetic |
Rice varieties tolerant to total herbicides |
Cultural methods
The best control of weedy rice can be obtained with rotation, but this practice cannot be applied in particular environmental conditions, such as saline and hydromorphic soils (Català, 1995, Sagarra, 1987). Crops that are normally rotated with rice in temperate climate areas include soybean, maize, wheat, sunflower, sorghum and other crops. The introduction of mungbean cropping in Vietnam resulted in a huge decrease of the weedy rice plants and other species (Watanabe et al. 1998). In these conditions many weeds can commonly emerge but do not complete their cycle because of the insufficient soil moisture during the mungbean season.
Several studies carried out in Italy have shown that weedy rice control in soybean is usually better than in maize (Ferrero and Vidotto, 1997b). This result can be attributed to the lower emergence of the weed seeds in maize and the greater efficacy of the herbicides in soybean. The lower emergence in maize is likely a result of the burial of the weed grains during soil tillage in deep soil layers, which prevents them from germinating. In these conditions seeds buried at more than 10 cm are not capable of emerging from the soil. One year of soybean cultivation led to a reduction of the seed bank in the 0-10 cm soil layer by about 97 percent (Ferrero and Vidotto, 1997b). The reduction in the same layer was still higher (98.5 percent) when soybean was planted at the end of May after the spring flush of weed emergence.
Post-emergence application of herbicides selective to soybean and maize were less effective than the combination of pre-emergence followed by post-emergence treatments for both rotational crops. The best results (99 percent of control) on soybean were obtained with the pre-emergence application of pendimethalin followed by post-emergence treatment with propaquizafop.
Similar results with rotation have also been reported in the southern part of the United States, where a one or two-year rotation with soybean is frequently adopted to control severe infestations of weedy rice plants (Barrentine et al. 1984; Khodayari et al. 1987; Minton et al. 1989; Griffin and Harger, 1990; Noldin et al. 1998). The use of antigerminative herbicides, such as metolachlor, at 3.5 kg ai ha-1, alachlor, at 3.5 kg ai ha-1, applied in soybean pre-emergence resulted in a weedy rice control of about 90 percent. Graminicides such as clethodim, fluazifop-P, quizalofop-P and sethoxydim also proved to be effective in suppressing the weed seedhead, in soybean post-emergence. The best results are usually obtained when herbicide application is delayed until the four-leaf stage (Askew et al. 2000).
As previously seen (Figure 2), weedy rice emergence greatly depends on the soil tillage that has been adopted for the seed bed preparation and the water contents during weed germination (Ferrero et al. 1996; Ferrero and Vidotto, 1999). Minimum tillage, performed at no more than 10 cm in depth and good soil moisture conditions create the best conditions for weedy rice emergence, while ploughing and soil flooding remarkably affect weed germination. Seed bed preparation with mouldboard ploughing can be considered as a helpful means of agronomic control when the weed infestation is low and no chemical control measures are planned.
A cultural strategy of weedy rice control also includes the use of weed-suppressing varieties and submergence tolerant varieties. Tall and long cycle varieties usually show a greater competitiveness than modern early and semi-dwarf varieties.
The planting of a tall variety such as Kilombero in Tanzania has resulted in the suppression of the growth of O. barthii, while the cultivation of the short stature Katrin resulted in the overgrowth of the weed (Jonhson et al. 1999).
The competitiveness of cultivated rice can also be improved by increasing the seeding rate, but this practise frequently results in an enhanced crop lodging and higher pest incidence (Sonnier, 1978).
Stale seed bed, also named the false seeding technique, is a cultural method commonly applied in rice monoculture. After seedbed preparation the area is left idle, to allow weedy rice and other weeds to grow. The rice can then either be drilled or water-seeded after the weeds are destroyed by either mechanical (harrows) or chemical (non-selective herbicides) means. This technique is aimed at reducing the weed infestation in the same season in which it is applied and gradually decreasing its seed bank. The success of the stale seed bed method depends on the way the soil is prepared, the water management and its duration. As previously shown, minimum tillage results in a higher percentage of germination of the seeds that are present in the upper soil layer, compared with mouldboard ploughing. In Rio Grande do Sul (Brazil) about 250 000 ha are cropped every cropping season using minimum tillage (Nolding and Cobucci, 1999). Soil flooding during the application of the stale seed bed reduces emergence from the soil in comparison to dry or moist soil, but favours the evenness of the germination that in turn make the control easier. The duration of the stale seed bed must be a compromise between the necessity of obtaining the greatest number of seedlings at the 2-3 leaf stage and that of not delaying rice planting too long. The duration of this technique in temperate climate conditions is about 25-30 days.
Water management can play an important role in weedy rice control. As previously reported, flooding in well-levelled soils limits weedy rice germination (Diarra et al. 1995c; Vidotto and Ferrero, 2000). Puddling combined with the presence of a thin layer of water over the well-levelled soil maintains the anaerobiotic conditions in the top soil and prevents weedy plants from becoming established (Fisher, 1999). Planting pre-germinated rice seeds on soil which has been flooded for 20 days after puddling has resulted in a good suppression of these weeds in Central America (Armenta and Coulombe, 1993). The combination of water-seeding and the use of weedy rice-free seed has led to the virtual disappearance of the weed in California (Fisher, 1999).
The control of weedy rice plants is sometimes carried out manually, but this practice is costly and time consuming. Hand weeding is quite impractical up to 30-40 days after crop emergence as it is very difficult to distinguish in the early stages the cultivated varieties from the weedy rice. The practice is then done when much of the competition damage has already occurred.
Hand weeding of weedy rice plants is sometimes carried out for light infestations and frequently it is used together with another means of control (e.g. chemical) when the latter has given poor results, so as to avoid grain dispersal. The manual control method is also of great interest in fields where yields are intended for seed production in order to get a weed-free production.
Mechanical
Several techniques using mechanical instruments can be applied to control weedy rice. Most of these can be applied in crop pre-planting, after weedy rice emergence which can be stimulated by the tillage that is carried out to prepare the seed bed. Weed germination can also be enhanced by watering the field or from seasonal rainfall. Weed seedlings can then destroyed by blade or rotary harrowing carried out on both dried and flooded soils, just before the planting of the rice. The weed control obtained with this practice is satisfactory but more time consuming and usually lower than that achieved with chemical destruction (Ferrero et al. 1999). Mechanical control can also favour new flushes of weed emergence after the interventions because of the germination stimulation of the seeds brought to the soil surface by the machinery (Finassi et al. 1996).
Weedy rice can also be mechanically controlled in rice planted in lines. In the Mekong delta (Vietnam) this method also resulted in a saving of more than 100 kg of rice seeds and a reduction of the damage resulting from insects, diseases and lodging (Chin et al. 1999). With this seeding technique it is possible for farmers to remove the weeds grown between the rows using mechanical tools and to raise fish and shrimps which can grow better than in broadcast water-seeding (Quan, 1999).
Mechanical control can also be applied after rice planting when weedy rice is taller than the crop. This practice is aimed at preventing the spread of the weed and is mainly carried out by cutting weed panicles before they set seeds.
In Colombia, the panicle is cut by a machete, while in Europe this operation is usually performed with a combine harvester cutting device that is mounted onto the front of a tractor (Ferrero and Vidotto, 1999). Cutting equipment is usually fitted with a roll-crusher made up from two contra-rotating rollers. The European experience has shown that at least 94 percent of the panicles can be cut down using this equipment in two phases, the first at the beginning of the flowering and the second 15 days later.
Chemical
The close anatomical and physiological similarity to the crop makes the control of weedy rice plants with selective post-emergence herbicides very difficult. The most successful management technique is based on herbicide application before crop planting, both before and after emergence of these weeds.
Several antigerminative herbicides such as chloroacetamides, thiocarbamates and dinitroanilines applied alone or in mixtures with other herbicides proved to be effective on weedy rice before its emergence (Khodayari et al. 1987; Griffin and Harger, 1990; Noldin et al. 1998,). Good control of these weeds (often higher than 75 percent) can be obtained in European rice conditions with pretilachlor and dimethenamid used alone or in combination at 1.5 kg ai ha-1 and 0.48 kg ha-1, respectively, (Ferrero and Vidotto, 1999). To avoid any phytotoxicity risks, both herbicides need to be applied at least 25 days before rice planting.
The main thiocarbamate herbicides that are used to control weedy plants are molinate and butylate (Smith, 1981; Fisher, 1999; Garcia de la Osa and Rivero, 1999). Both products are applied in pre-planting and need to immediately be incorporated into the soil to avoid volatilisation. According to the experiments carried out by CIAT in Central and South America, the best results can be achieved by applying molinate at 7.2 kg ai ha-1 and butylate at 4.2 kg ai ha-1 with seed protectants such as oxabetrinil at 1.5 g ai kg-1 and flurazole at 2.5 g ai kg-1 (Smith, 1992).
In continued flooded monocultures, an effective management of weedy rice is often achieved through the application of the stale seed bed technique followed by spraying of the graminicides or total herbicides once the weeds have reached the 2-3-leaf stage at least (Vidotto et al. 1998). The most frequently applied graminicides are dalapon (about 12 kg a.i. ha-1), clethodim (0.2 kg ai ha-1) and cycloxydim (0.6-0.8 kg ai ha-1). Other wide spectrum herbicides are glyphosate (1-1.5 kg ai ha-1), glufosinate ammonium (0.5-0.7 kg ai ha-1), paraquat (0.8 kg ai ha-1) and oxyfluorfen (0.8 kg ai ha-1). Graminicides are highly effective even at early stages of the weeds while total herbicides have to be applied on more developed plants. Delaying the treatment to a more advanced growth phase of the weeds implies the planting of very early and sometimes low-yielding varieties.
Chemical control in crop post-planting should only be considered as a salvage operation and mainly relies on difference in size or growth stage between weedy rice and commercial rice. This practice prevents the infestation from becoming worse thanks to the grain shattering, but has no influence on the weed-crop competitive relationships.
Weedy rice that has grown taller than rice can be treated with foliar systemic herbicides such as glyphosate or cycloxydim, at 20 and 5 percent concentrations, respectively, by using wick/wiper applicators. This equipment wipes the herbicide over the top of the weeds and, owing to the difference in height between these plants and the crop, prevents contact with the desirable vegetation. Wick/wiper applicators are usually made up of a frame with a rope, sponge or carpet which can absorb the herbicide solution and wipe it onto the weed (Stroud and Kempen, 1989). They can be mounted on self-moving machines, the front of a tractor or hand-held equipment. The results of the treatments carried out with this equipment on semi-dwarf varieties at the beginning of the weedy plant flowering showed a higher than 90 percent germinability reduction of the weed seeds (Balsari and Tabacchi, 1997; Ferrero and Vidotto, 1999). This percentage concerned only the seeds of the weed panicle that come in contact with the wiping equipment. About one-third of the panicles in the experimental field escaped the treatment as they were equal to or lower in height than the crop. The seeds of the escaped panicles, on one hand, can feed the soil seed bank, but on the other, can select short biotypes for the following years that can no longer be controlled with this equipment.
The seed viability of weedy rice can be affected by spraying maleic hydrazide at the heading stage of these plants (Noldin and Cobucci, 1999). To avoid negative effects on the yield and seed viability, commercial rice plants have to be earlier and to have reached the milky-stage. The use of this growth regulator has been approved in Brazil and is being tested in several countries in South America.
Genetic and biotechnological
The genetic and biotechnological approach is largely being adopted to face abiotic and biotic issues in rice such as water scarceness, low and high temperatures, pests, diseases and weed control (Fujimoto et al. 1993; Rathore et al. 1993; Christou, 1994).
The problem of weedy rice can be tackled by the introduction of herbicide-tolerant varieties which allow the selective post-emergence control of this plant (Linscombe et al. 1996; Wheeler et al. 1997).
Many traditional breeding works and, in particular biotechnological research have been carried out to obtain varieties that are resistant to glyphosate, glufosinate-ammonium, imidazolinones and wide spectrum herbicides which are non selective for traditional rice varieties. Several rice accessions that are tolerant to glyphosate and sulfosate were identified, out of more than 14,000 originating from Colombia, Brazil, India and the United States (Dilday et al. 1995).
The control of weedy rice plants using herbicide-tolerant varieties can lead to different results depending on the variety, timing and cultural conditions.
The glufosinate-resistant variety Gulfmont showed injuries when subjected to sequential applications of glyfosate at 0.42 kg ai ha-1 (Wheeler et al. 1998). Glufosinate can be safely applied to transgenic varieties at the 3-4-leaf or at the tillering stage (Sankula et al. 1997a). Glufosinate applied at the 3-4-leaf stage of the weedy rice (red rice) resulted in a better control (91 percent) than at panicle initiation (74 percent) or boot stage (77 percent) (Sankula et al. 1997b). Better weedy rice control was obtained by applying glufosinate to drained rice fields (Sankula et al. 1997a). Soil flooding reduced the herbicide activity proportionally to the water depth.
Imazethapyr can be selectively applied to imidazolinone-resistant varieties (IMI rice). This herbicide has proved to be effective against weedy rice and other rice weeds when applied as a soil or foliar treatment at 70 kg ai ha-1 (Olofsdotter et al. 1999).
The introduction of herbicide-resistant varieties often raises concern from ethical, sanitary, social, environmental and biological points of view. The ethical and sanitary issues mainly concern the question of whether man has the right to manipulate the natural genome of a living being with genetic engineering technologies and the supposed health risks to human beings of the product obtained from transgenic plants.
The social problems could be related to the dependence of rice farmers on the resistant-seed producers. This problem is considered higher in the developing world where rice growers are accustomed to saving seeds from season to season.
The environmental and biological constraints are mainly associated with the risk of spreading the resistance gene from the crop to other Oryza species, the growth of volunteer resistant rice or the selection, in the long run, of uncontrolled plants (Langevin et al. 1990; Oard et al. 2000).
The transfer of resistance gene to weedy species is likely to occur as the incidence of natural hybridization has been reported as ranging between 1-52 percent in early and late flowering varieties, respectively (Langevin et al. 1990). Field studies carried out in Spain have shown that the average gene flow from the transgenic Senia variety (tolerant to glufosinate) to red weedy rice, considering all the wind directions, was 0.082 percent (Messeguer, 2002). These findings suggest that within a few generations the advantages of the herbicide resistance gene could partly disappear.
The continuous cultivation of transgenic or IMI rice varieties could also lead to the selection of uncontrolled plants. This constraint could be overcome by turning to rotational crops (e.g. soybean) and using herbicides with different action mechanisms or with mechanical weed control means.
Volunteer rice could become a real problem mainly in the production of non-transgenic certified seeds. For this reason the cultivation of common rice for seed production should not succeed that of a transgenic rice variety.
CONCLUSIONS
Several species of the genus Oryza behave like a weed even though they share most of the features of the cultivated varieties. They are undesirable, above all, because their seeds can easily shatter before crop threshing and remain dormant in the soil for a long period of time. Weedy rice varieties are usually very similar to commercial varieties, both as regardsplant morphology and tolerance to herbicides. Because of their high competitive ability, these weeds can remarkably affect rice yields.
The effective control of weedy rice cannot be based on one single practice, but should rely on complex management programme based on an appropriate combination of preventative, cultural, mechanical, chemical and genetic means (Vidotto et al. 2001). Preventative practices, which include the use of weedy rice-free seed and clean equipment, are the starting point for a successful application of other means of control. Of the cultural practices, rotation is frequently the best way of reducing severe weedy rice infestation. In continuous rice cultivation, an effective control of the weed can be obtained by applying the stale seed bed method to stimulate weed germination and by destroying the seedlings through harrowing or with herbicides.
The spread of weedy rice seeds can be successfully prevented in crop post-planting both by panicle cutting or the localised application of systemic herbicides, but these measures should be aimed more at preventing the infestations from becoming worse, rather than reducing them.
The introduction of herbicide-resistant varieties offers rice growers a good opportunity to manage weedy rice and other weeds, even though its success depends on how well the cultivation strategies can avoid the transfer of resistance genes to weeds.
BIBLIOGRAPHY
Aggarval, R.K., Brar, D.S. & Klush, G.S. 1997. Two new genomes in the Oryza complex identified on the basis of molecular divergence analysis, using total genomic DNA hybridization. Molecular General Genetics 254, 1-12.
Armenta, S.J. & Coulombe, J. 1993. Highlights of the Caribbean Rice Improvement Network Activities (1986-1992), Bonao, Dominican Republic: CRIN/CIAT/IRRI/IICA/SEA: 53-65.
Armstrong, K. 1968. Weed control on a Swaziland rice and sugar cane estate. Proc. 9th British Weed Control Conference-9, 687-693.
Askew, S.D, Shaw, D.R. & Street, J.E. 2000. Graminicide application timing influences red rice (Oryza sativa) control and seedhead reduction in soybean (Glycine max). Weed Tech. 14: 176-181.
Baldi, G. 1971 Presenza del carattere Pericarpo rosso in varietà di riso coltivate (O. sativa). Il riso 20: 299-302.
Balsari, P. & Tabacchi, M. 1997. Lotta meccanica di soccorso al riso crodo. Linformatore Agrari, 53 (14), 56-60
Barrentine, W.L., Street, J.E. & Kurtz, M.E 1984. Post-emergence control of red rice (O. sativa). Weed Sci. 32: 832-834.
Cai, H.W. & Morishima, H. 2000. Genomic regions affecting seed shattering and seed dormancy in rice. Theoretical and Applied Genetics 100: 840-846.
Castro Espitia, H.A. 1999. Manejo de arroces contaminates en las areas productoras de arroz comercial de Costa Rica. Report of the Global Workshop on Red Rice Control, 30 August-3 September, Varadero, Cuba, 19-24.
Català, M. 1995. Chemical and cultural practices for red rice control in rice fields in Ebro Delta, Spain. Crop Protection 5: 405-408.
Chin, D.V., Hach, C.V., Thanh, N.C. & Tai, N.T. 1999. Weedy rice situation in Vietnam. Report of the Global Workshop on Red Rice Control, 30 August-3 September, Varadero, Cuba, 67-74.
Cohn, M.A. & Hughes, J.A. 1981. Seed dormancy in red rice (Oryza sativa L.) I. Effect of temperature on dry after-ripening red rice. Weed Sci. 29, 402-404.
Cohn, M.A. 1996. Chemical mechanisms of breaking seed dormancy. Seed Science Research 6: 95-99.
Coppo, B. & Sarasso, G. 1990. Il riso crodo. In Quaderno Agricolo, Istituto Federale di Credito Agrario, Piemonte, Liguria e Valle dAosta 22: pp.15-29.
Craigmiles, J.P. 1978. Introduction. In Red rice research and control, ed. Eastin, E.F. pp.5-6. Texas Agric. Exp. Stn. Bull, pp.1270
Christou, P. 1994. Biotechnology of food crops - Rice biotechnology and genetic engineering. Technomic Publishing Company, Lancaster, USA, pp.201
Delatorre, C.A.1999. Dormencia em sementes de arroz vermelho. Ciencia Rural 29: 565-571.
De Souza, P.R. 1989. Arroz vermelho: um grande problema. Lavoura arrozeira 42: 30-31.
Diallo, S. 1999. Problème posé par le riz rouge en riziculture au Sénégal. Report of the global workshop on red rice control, 30 August-3 September, Varadero, Cuba, 45-49.
Diarra, A.R.J., Smith, R.J. & Talbert, R.E. 1985a. Growth and morphological characteristics of red rice (Oryza sativa) biotypes. Weed Sci. 33: 310-314.
Diarra, A.R.J., Smith, R.J. & Talbert, R.E 1985b. Interference of red rice (Oryza sativa) with rice (O. sativa). Weed Sci. 33: 644-649.
Diarra, A.R.J., Smith, RJ. & Talbert, R.E. 1985c. Red rice (Oryza sativa) control in drill-seeded rice (Oryza sativa). Weed Sci. 33: 703-707.
Dilday, R.H., Jalaluddin, M. & Price, M. 1995. Tolerance in rice to glyphosate and sulfosate. Proc. Int. Symposium on Weed and Crop Resistance to Herbicides. Cordoba, Spain, 192-193.
Eastin, E. 1978. Additional red rice research in Texas. In Red rice research and control. Texas Agriculture Experimental Station Bulletin No. 1270.
Elefhterohorinos, I. G., Dhima, K.V. & Vasilakoglou, I.B. 2002. Interference of red rice in rice grown in Greece. Weed Sci. 50: 167-172.
Estorninos, L.E Jr., Gealy, D.R. & Talbert, R.E. 2000. Interference between red rice and rice in a replacement series studies. Research Series Arkansas Agricultural Experiment Station, pp. 463-468.
FAO. 1999. Report of the Global workshop on red rice control. Varadero, Cuba, 30 August-3 September, pp. 55.
Federici, M.T., Vaughan, D., Tomooka, N. Kaga, A. Wang, X.W., Doi, K., Francis, M., Zorrilla, G. & Saldain, N. 2001. Analysis of Uruguayan weedy rice genetic diversity using AFLP molecular markers. EJB Electronic Journal of Biotechnology. (available at http://www.ejb.org/content/vol4/issue3/full/3)
Ferrero, A. & Finassi, A. 1995. Viability and soil distribution of red rice (Oryza sativa L. var. sylvatica) seeds. In Med. Fac. Landbouw., Rijksunv. Gent. pp. 205-211.
Ferrero. A., Finassi, A. & Vidotto, F. 1996. Prediction of red rice seedling densities from seed bank. In Med. Fac. Landbouw., Rijksuniv. Gent. pp. 1181-1187.
Ferrero, A. & Vidotto, F. 1997a. Influence of soil tillage on red rice emergence. In Med Fac. Landbouw. En Toegepaste Biologische Wetenschappen, Universiteit Gent, 62: 785-789.
Ferrero, A. & Vidotto, F. 1997b. Influence of the rotation on seed bank evolution of red rice (Oryza sativa L. var. sylvatica). Proc.of the Int. Symposium on Rice Quality, Quality and Competitiveness of European Rices, Concerted Action-EC-DG VI (AIR3-PL93-2518). Nottingham UK, November 24-27.
Ferrero, A. Vidotto, F. 1998a. Germinability after flowering, shattering ability and longevity of red rice seeds. 6th EWRS Mediterranean Symposium 1998, Montpellier, 205-211.
Ferrero, A. & Vidotto, F. 1998b. Shattering ability of red rice seeds in cultural conditions. Proc. 50th Int. Symposium on Crop Protection, Gent, Belgium, 839-843.
Ferrero, A. & Vidotto, F. 1999. Red rice control in rice pre- and post-planting. In FAO Report of the Global workshop on red rice control. Varadero, Cuba, 30 August-3 September, 95-107.
Ferrero, A., Vidotto, F., Balsari, P. & Airoldi, G. 1999. Mechanical and chemical control of red rice (Oryza sativa L. var. sylvatica) in rice (Oryza sativa) pre-planting. Crop Protection, 18: 245-251.
Finassi, A. Airoldi, G., Balsari, P. & Ferrero, A. 1996. Contenimento del riso crodo con interventi meccanici. Proc. Giornate Fitopatologiche, Numana, 1996. 1: 413-420.
Fischer, A.J & Ramirez, A. 1993. Red rice (Oryza sativa): competition studies for management decisions. Int. J. Pest Management 39: 133-138.
Fischer, A.J. 1999. Problems and opportunities for managing red rice in Latin America. Report of the global workshop on red rice control. 30 August-3 September, Varadero, Cuba. 77-85.
Fletes, M.S. 1999. Evaluation de la maleza arroz rojo (Oryza sativa) en las principales zonas arroceras de Nicaragua. Report of the global workshop on red rice control. 30 August-3 September, Varadero, Cuba. 41-44.
Footitt, S. & Cohn, M.A. 1992. Seed dormancy in red rice. VII Embrio acidification during dormancy-braking and subsequent germination. Plant Physiolog, 100, 1196-1202.
Footit, S. & Cohn, M.A. 1995. Seed dormancy in red rice (Oryza sativa) IX. Embrio fructose-2,6-bisphosphate during dormancy breaking and subsequent germination. Plant Physiology 107, 1365-1370.
Fujimoto, H., Itoh, K., Yamamoto, M., Kyozuka, J. & Shimamoto, K. 1993. Insect-resistant rice generated by introduction of a modified delta-endotoxin gene of Bacillus thuringiensis. Biol. Technology, 11: 1151-1155.
Garcia de la Osa, J. & Rivero, L.E. 1999. El arroz rojo. Estudios y perspectivas de su manejo en la produccion arrocera cubana. Report of the Global Workshop on Red Rice Control, 30 August-3 September, Varadero, Cuba. 25-31.
Gealy, D.R, Saldain, N.E, Talbert, R.E. 2000. Emergence of red rice (Oryza sativa) ecotypes under dry-seeded rice (Oryza sativa) culture. Weed Tech. 14: 406-412.
Ghesquière, A. 1999. Report to European Commission of the research project biology and control of red rice, FAIR CT 1496, coordinated by Ferrero, A.
Goss, W.L., & Brown, E. 1939. Buried red rice. J. of American Society of Agronomy 31: 633-637.
Griffin, J.L. & Harger, T.R. 1990. Red rice (Oryza sativa) control options in soybeans (Glycine max). Weed Tech. 4: 35-38.
Hoagland, R.E. & Paul, R.N. 1978. A comparative study of red rice and several commercial rice (Oryza sativa) varieties. Weed Sci. 6: 619-625.
Holm, L.R. Doll, J., Holm, E., Pancho, J. & Herberger, J. 1997. The Wild rices. Oryza sativa L., Oryza punctata Kotschy ex Steud., Oryza rufipogon Griff., Oryza barthii A. Chev. (syn. O. breviligulata A. Chev. Et Roehr) and Oryza officinalis Wall ex Watt. In World weeds. Natural histories and distribution. New York, John Wiley and Sons, Inc. 531-547.
Katayama, T. 1969. Botanical studies in the genus Oryza. Part II: Germination behaviour. Memoirs Faculty Agriculture, Kagoshima University. 7, 89-119.
Katayama, T. 1974. Photoperiodism in the genus Oryza. IV. Combinations of plant age day length and number of treatments. Proc. Crop Science Society Japan, 43, 224-236.
Khodayari, KR., Smith, J.R., Jr. & Black, H.L 1987. Red rice (Oryza sativa) control with herbicide treatments in soyabeans (Glycine max). Weed Sci. 35: 127-129.
Khush, G.S. 1997. Origin, dispersal cultivation and variation of rice. Plant molecular biology, 35: 25-34.
Klosterboer, A. 1978. Red rice control in Texas. In Red rice: Research and Control. Texas Agriculture Experimental Station, Bulletin 1270.
Kwon, S.L., Smith, R.J., Jr. & Talbert, R.E. 1991a. Interference of red rice (Oryza sativa L.) densities in rice (Oryza sativa L.).Weed Sci. 39: 197-174.
Kwon, S.L., Smith, R.J., Jr. & Talbert, R.E. 1991b. Interference and duration of red rice (Oryza sativa L.) in rice (Oryza sativa). Weed Sci. 39: 363-368.
Kwon, S.L., Smith, R.J., Jr. & Talbert, R.E. 1992. Comparative growth and development of red rice (Oryza sativa) and rice (O. sativa). Weed Sci. 40: 57-62.
Langevin, A.S., Clay, K. & Grace, B.J. 1990. The incidence and effect of hybridization between cultivated rice and its related weed, red rice (Oryza sativa L.). Evolution 44: 1000-1008.
Leitao, H.N., Banzato, N. & Azzini, L. 1972. Estudio de competicao entre o arroz vermelho e o arroz cultivado. Bragantia 31: 249-258.
Leopold, A.C., Glenister, R. & Cohn, M.A. 1988. Relationship between water content and after-ripening in red rice. Physiologia Plantarum 74: 659-662.
Linscombe, S.D., Jodary, F., Christou, P., Braverman, M.P., Oard, J.H. & Sanders, D.E. 1996. Potential for the use of transgenic rice for the control of Oryza sativa and other rice weeds. Proc. 2nd Int. Weed Control Congress, Copenhagen, 435-439.
Johnson, D.E., Riches, C.R., Kayeke, J., Sarra, S. & Tuor, F.A. 1999. Wild rice in sub-Saharan Africa: its incidence and scope for improved management. Report of the global workshop on red rice control, 30 August-3 September, Varadero, Cuba. 87-93.
Majisu, B. 1970. A potential dangerous weed of rice in East Africa. East African Agriculture and Forestry Research Organization., Newsletter 60. Nairobi.
Messeguer, J. 2002. Field assessment of gene flow from transgenic rice to cultivated and red rice. Proc. of Riceuconf Dissemination Conference of current European Research on Rice. Turin, 16-17 bis.
Minton, B.W., Shaw, D.R. & Kurtz, M.E. 1989. Postemergence grass and broadleaf herbicide interactions for red rice (Oryza sativa) control in soybeans (Glycine max). Weed Tech. 3: 329-334.
Nagao, S. & Takahashi, M. 1963. Trial construction of twelve linkage group in Japanese rice. J. Fac. Agric. Hokkaido Univ., 53: 72-130.
Noldin, J.A., Chandler, J.M., McCauley, G.N. & Sij, J.W. 1998. Red rice (Oryza sativa) and Echinochloa spp. control in Texas Gulf coast soybean (Glycine max). Weed Tech. 12: 677-683.
Noldin, J.A. & Cobucci, T. 1999. Red rice infestation and management in Brasil. Report of the global workshop on red rice control, 30 August-3 September, Varadero, Cuba. 9-13.
Oard, J., Cohn, M.A., Linscombe, S., Gealy, D. & Gravois, K. 2000. Field evaluation of seed production, shattering and dormancy in hybrid populations of transgenic rice (Oryza sativa) and the weed, red rice (Oryza sativa). Plant Science 155: 13-22.
Olofdotter, M., Valverde, B.E. & Madesen, K.H. 1999. Herbicide-resistant rice (Oryza sativa L.) - A threat or a solution. Report of the Global Workshop on Red Rice Control, 30 August-3 September, Varadero, Cuba. 9-3, pp. 123-145.
Pantone, D.J. & Baker, J.B. 1991a. Weed-crop competition models and response-surface analysis of red rice competition in cultivated rice: a review. Crop Science 31 (5): 1105-1110.
Pantone, D.J. & Baker, J.B. 1991b. Reciprocal yield analysis of red rice (Oryza sativa) competition in cultivated rice. Weed Sci. 39: 42-47.
Pantone, D.J., Baker, J.B. & Jordan, P.W. 1992. Path-analysis of red rice (Oryza sativa L.) competition with cultivated rice. Weed Sci. 40, 313-319.
Parker, C. & Dean, M.L. 1976. Control of wild rice in rice. Pesticide Science 7: 403-416.
Quan, H.Q. 1999. Improve rice yield potential on intensive integrated and and mechanized cultivation in Song Hau state farm. Fourth Rice variety improvement in Mekong delta, Canton. Cited from Chin et al, 1999.
Rao, S.R. & Harger, T.R. 1981. Mefluidide-bentazon interactions on soybeans (Glycine max) and red rice (Oryza sativa). Weed Sci. 29: 208-212.
Rathore, K.S., Chowdhury, V.K. & Hodges, T.K. 1993. Use of bar as a selective marker gene and for the production of herbicide-resistant rice plants from protoplasts. Plant Molecular Biology 21: 871-884.
Roy, S. 1921. A preliminary classification of the wild rices of the Central Provinces and Berar. Agricultural J. India 16: 365-380.
Sagarra, J. 1987. Importancia del conreu de larròs a Catalunya. Produccions i aprofitaments. In Larròs tecniques, pp. 7-10. Fundacio La Caixa, Amposta.
Saldain, N.E, Talbert, R.E., Gealy, D.R. & Guy, C.B. 1996. Emergence of red rice ecotypes under dry- and water-seeded rice conditions. Research Series Arkansas Agricultural Experiments Station, 70-76.
Sankula, S., Braverman, M.P., Jodari, F., Linscombe, S.D. & Oard, J.H. 1997a Evaluation of Gufosinate on rice (Oryza sativa) transformed with BAR gene and red rice (Oryza sativa). Weed Tech. 11: 70-75.
Sankula, S., Braverman, M.P., Jodari, F., Linscombe, S.D. & Oard J.H. 1997b. Response of BAR-Transformed rice (Oryza sativa) and red rice (Oryza sativa) to glufosinate application timing. Weed Tech. 11: 303-307.
Sastry, M.V.S., Seetharaman, R. 1973. Inheritance of grain shattering and lazy habit and their interrelationship in rice. Indian J. Gen. Plant. Breed, 38: 318-321.
Smith, R.J., Jr. 1981. Control of red rice (Oryza sativa L.) in water seeded rice (Oryza sativa L.). Weed Sci. 29: 61-62.
Smith, R.J., Jr. 1992. Integrated red rice management. In Rice in Latin America: improvement, management and marketing, pp. 143-158. ed. Cuevas-Pérez, F. Cali Colombia. Centro International de Agricultura Tropical (CIAT) and International Rice Research Institute (IRRI).
Sonnier, E.A. 1978. Cultural control of red rice. Proc. Red Rice Symposium, Texas. A. and M. University, College Station, Texas.
Stroud, D. & Kempen, H.M. 1989. Wick/Wiper. In Principles of weed control in California. California Weed Conference. Thomson Publication, Fresno, CA, pp. 148-150.
Suh, H.S., Sato, Y.I. & Morishima, H. 1997. Genetic characterization of weedy rice (Oryza sativa L.) based on morpho-physiology, isozymes and RAPD markers. Theoretical applied Genetics, 94: 316-321.
Tang, L., Morishima, H. & Tang, L.H. 1997. Genetic characterization of weedy rices and the interference on their origins. Breeding Science 47: 153-160.
Tarditi, N. & Vercesi, B. 1993. Il riso crodo: un problema sempre più attuale in risicoltura. LInformatore Agrario 11: 91-95.
Vaughan, L.K., Ottis, B.V., Prazak-Havey, A.M., Bormas, C.A., Sneller, C. & Chandler, J.M. 2001. Is all red rice found in commercial rice really Oryza sativa? Weed Sci. 49: 468-476.
Vidotto, F., Ferrero A. & Tabacchi, M. 1998. Lotta al riso crodo (Oryza sativa L. var. sylvatica) con la tecnica della falsa semina. Proc. Giornate Fitopatologiche, Scicli e Ragusa, 1998. 369-374.
Vidotto, F, Ferrero A. 2000. Germination behaviour of red rice (Oryza sativa L.) seeds in field and laboratory conditions. Agronomie 20: 375-382.
Vidotto, F., Ferrero, A. & Ducco, G. 2001. A mathematical model to predict the population dynamics of Oriza sativa var. sylvatica, Weed Res. 41: 407-420.
Watanabe, H., Vaughan, D.A. & Tomooka, N. 1998. Weedy rice complexes: Case studies from Malaysia, Vietnam and Suriname. Int. Symposium on Wild and Weedy Rices in Agro-ecosystems, 10-11 August. Ho Chi Minh city.
Wheeler, C.C., Baldwin, F.L., Gealy, D., & Gravois, K. 1997. Weed control in Liberty-tolerant rice. Research Series Arkansas Agricultural Experiment Station, 64-66.
Wheeler, C.C., Baldwin, F.L., Talbert, R.E. & Webster, E.P. 1998. Efficacy of Liberty (glufosinate) in Liberty-tolerant rice. Research Series Arkansas Agricultural Experiment Station, 330-335.
Wirjahardja, S., Guhardja, E. & Wiroatmodjo, J. 1983. Wild rice and its control. Proc. Weed Control Rice Conference, IRRI, Philippines.