Evaluation of the potential role of glufosinate-tolerant rice in integrated pest management programs for rice water weevil (Coleoptera: Curculionidae)

The impact of a herbicide-tolerant rice, Oryza sativa L., variety was assessed for its resistance to rice water weevil, Lissorhoptrus oryzophilus Kuschel (Coleoptera: Curculionidae), and its place in current integrated pest management (IPM) programs. Greenhouse experiments were conducted to evaluate the resistance of a glufosinate-tolerant rice variety and its glufosinate-susceptible parent line Bengal to the rice water weevil in the presence and absence of glufosinate applications. The LC50 dose-response and behavioral effects of glufosinate on adult rice water weevils also were studied. Field studies investigated the impacts of glufosinate-tolerant rice on rice water weevil management in the presence and absence of glufosinate under early and delayed flood conditions. Greenhouse studies demonstrated that in the absence of glufosinate, oviposition was 30% higher on the glufosinate-tolerant rice line than on Bengal rice or on glufosinate-tolerant line treated with recommended rates of commercially formulated glufosinate. Applications of glufosinate to glufosinate-tolerant rice resulted in a 20% reduction in rice water weevil larval densities compared with nontreated glufosinate-tolerant rice. The LC50 of glufosinate against adult rice water weevil was nearly 2 times the concentration recommended for application to glufosinate-tolerant rice. There was no difference in the amount of leaf area consumed by adult rice water weevils on glufosinate-treated and nontreated foliage. The absence of direct toxicity of glufosinate to rice water weevil at recommended glufosinate use rates and lack of behavioral effects suggest that the reduction in rice water weevil densities observed after glufosinate applications resulted from herbicide-induced plant resistance. Field experiments showed that neither rice variety nor herbicide use affected larval densities; however, delaying flood and applying insecticide effectively reduced numbers of rice water weevil larvae.     

Identification of rice cultivars resistant to Lissorhoptrus oryzophilus (Coleoptera: Curculionidae), and their use in an integrated management program

The rice water weevil, Lissorhoptrus oryzophilus Kuschel, is the most destructive insect pest of rice in the United States and is a particularly severe pest in Louisiana. The current management program for this insect in Louisiana relies heavily on insecticides, most notably the seed treatment fipronil (Icon). Diversification of the management program by incorporation of alternative strategies is needed to improve the effectiveness and long-term stability of the program. In the three experiments reported here, three components of a diversified management program for the rice water weevil in Louisiana--host plant resistance, treatment of seeds with Icon, aid the cultural practice of delayed flooding--were investigated. Comparison of the densities of weevil larvae on the roots of several commercial cultivars indicated that the long-grain cultivar 'Jefferson' was more resistant to infestation by the rice water weevil than the other cultivars. The medium-grain cultivars 'Bengal', 'Earl' and 'Mars' and the long-grain variety 'Cocodrie' were the most susceptible to infestation. Comparison of yield data from untreated plots and plots treated with Icon indicated that the long-grain cultivars Cocodrie, Lemont, and Jefferson were more tolerant of weevil injury than the other cultivars. A 2-wk delay in flooding was associated with yield benefits in plots not treated with Icon. Treatment of seeds with Icon controlled weevils in all three screening experiments. The implications of these results for the development of an integrated management program for the rice water weevil are discussed.     

The influence of plant age on tolerance of rice to injury by the rice water weevil,Lissorhoptrus oryzophilus (Coleoptera: Curculionidae)

For most plant species, tolerance to many types of herbivory increases as plants age, but the applicability of this pattern to root herbivory has not been tested. Injury to roots of rice plants by larvae of the rice water weevil, Lissorhoptrus oryzophilus Kuschel, causes severe reductions in yields in the United States. It is generally thought that young rice plants, because their root systems are smaller, are less tolerant than older plants of root feeding by L. oryzophilus. Field experiments were conducted to test this hypothesis. Plots of rice (4.7 to 6.5 m2) were established and subjected to natural infestations of L. oryzophilus larvae. A soil insecticide was applied to plots at different times during the tillering phase of rice in order to manipulate the timing of weevil infestation. The impact of these treatments (timings of insecticide applications) was assessed by comparing relationships between yield loss and larval pressure for each treatment using analysis of covariance. Yield losses ranged from 13% to over 40% in plots not treated with insecticide. Patterns of yield losses from plots treated with insecticide at different times were best explained by the hypothesis that yield loss is determined both by the age of plants infested and by the size of larvae infesting plants. Young plants appear to be less tolerant than older plants, and feeding by large larvae appears to be more deleterious than feeding by smaller larvae. Management practices that delay infestation of rice by L. oryzophilus until plants are older may be an important component of management programmes for this pest.     

Identification and biological role of the endosymbionts Wolbachia in rice water weevil (Coleoptera: Curculionidae)

Wolbachia spp. are obligate intracellular bacteria present in reproductive tissues of many arthropod species. Wolbachia infection status and roles in host reproduction were studied in the rice water weevil, Lissorhoptrus oryzophilus Kuschel (Coleoptera, Curculionidae), an introduced species in China. We examined Wolbachia infection status in five populations in China where it reproduces parthenogenetically, and one native population in Southeast Texas, where it reproduces bisexually. All populations were infected by Wolbachia, and all specimens in each population were infected by Wolbachia of a single strain. Phylogenetic analyses based on multilocus sequence typing system indicated that Wolbachia in non-native L. oryzophilus weevils diverges evidently from those in native weevils. After treatments with tetracycline, parthenogenetic weevils reduced the fecundity significantly and eggs were not viable. Our results suggest that Wolbachia are necessary for oocyte production in L oryzophilus.     

Plant‐mediated interactions among above‐ground and below‐ground life stages of a root‐feeding weevil

1. Interactions among herbivores mediated by plant responses to herbivore injury may have large impacts on herbivore population densities. Responses may persist for weeks after injury and may affect not only the initial (inducing) herbivore, but also herbivores that are spatially or temporally separated from the initial attacker. 2. In many plant–insect interactions, multiple life stages of the insect may be associated with the same plant, and these various stages may interact indirectly with one another via induced responses. The rice water weevil (RWW), Lissorhoptrus oryzophilus, a serious global pest of rice, is one such insect. A series of experiments were performed with root‐feeding larvae and leaf‐feeding adults of the RWW using three conventional rice varieties. 3. The first objective of this study was to test whether RWW adult feeding on rice leaves resulted in altered oviposition by subsequent adults. The hypothesis for the first objective was that RWW adult feeding would decrease plant suitability, resulting in reduced oviposition by subsequent adults. 4. The second objective was to test whether injury by RWW larvae to rice roots resulted in altered oviposition by subsequent adults. The hypothesis for the second objective was that below‐ground RWW larval feeding would decrease plant suitability of rice to above‐ground RWW adults, resulting in decreased oviposition. 5. Results provided inconsistent support for the first hypothesis, indicating that responses differed among combinations of variety and injury level. Conversely, consistent support for the second hypothesis was found, indicating that larval feeding on roots decreased suitability of rice plants for oviposition.     

The influence of rice plant age on susceptibility to the rice water weevil,Lissorhoptrus oryzophilus

Plant age‐ and plant stage‐related changes in the resistance of rice, Oryza sativa, to its most important insect pest in the US, the rice water weevil (Lissorhoptrus oryzophilus), were investigated in a series of field and greenhouse choice and no‐choice studies. Rice plants were susceptible to infestation by rice water weevils over a broad range of plant ontogenetic stages, from at least the early vegetative stage to well into the reproductive stage. There was, however, a clear preference expressed by rice water weevils in both choice and no‐choice experiments for plants in (or nearly in) the tillering stage of development, with pre‐tillering and reproductive stage plants less preferred. The relationship between rice plant age and susceptibility to weevils is thus nonlinear. This study constitutes one of the most thorough studies to date of the relationship in a grass species between plant age and susceptibility to herbivores. The results provide a biological explanation for observed patterns of weevil infestations and a rationale for the cultural practice of delayed flooding.     

Deposition and longevity of oviposition-deterring pheromone in the cabbage seed weevil

Abstract The cabbage seed weevil ( Ceutorhynchus assimilis Payk.) lays eggs singly into pods of oilseed rape ( Brassica napus L.) through punctures bored with the mouthparts, preferring pods not recently used for oviposition. A simple new choice test has been used to test individual components of egg-laying behaviour for their effect on oviposition site selection. It is confirmed that an oviposition-deterring pheromone (ODP) is deposited during abdomen brushing of the pod which follows egg-laying. Neither pin punctures, weevil feeding punctures, oviposition punctures nor eggs had any deterrent effect. Pods walked on by female weevils were not avoided by those laying eggs. Observations suggest that the ODP is sensed by contact chemoreceptors on the antennae. The deterrent effect lasted only 1–2 h. The implications of these findings on the adaptive significance of the pheromone and its possible use in pest control are discussed.     

Oviposition and feeding behaviour by the vine weevil Otiorhynchus sulcatus on red raspberry: effects of cultivars and plant nutritional status

• 1 The vine weevil Otiorhynchus sulcatus is a major pest of horticultural crops worldwide, with root‐feeding larvae causing most damage. Adult oviposition aboveground may therefore influence levels of damage as the larvae are relatively immobile after oviposition.
• 2 The present study investigated feeding and oviposition behaviour on red raspberry Rubus idaeus using intact plants, ensuring that choices reflected the realistic differences in cultivar appearance and chemical composition. Previous studies investigating vine weevil feeding and oviposition on other crops have used excised plant material, which may inadvertently influence behaviour.
• 3 Adult weevils significantly preferred to feed on particular cultivars in the choice experiment (e.g. Tulameen), although they consumed significantly more foliage (0.22–1.03 cm²/day) on different raspberry cultivars (e.g. Glen Moy, Glen Rosa and a wild accession) in no‐choice situations.
• 4 In choice experiments, weevils tended to avoid laying eggs on some cultivars (e.g. Glen Moy and the wild accession). The number of eggs laid (1.91–4.32 eggs per day) did not, however, differ significantly between the cultivars in a no‐choice situation. Foliar nitrogen and magnesium concentrations were positively, although weakly, correlated with the total number of eggs laid.
• 5 The present study highlights the importance of considering both choice and no‐choice tests when assessing crop susceptibility to attack because weevils may avoid feeding on certain cultivars (e.g. Glen Moy) when given a choice, although this would cause significant damage to such cultivars if they were grown in monoculture (i.e. when there is no alternative).

     

Effect of arbuscular mycorrhizal fungi (Glomus intraradices) on the oviposition of rice water weevil (Lissorhoptrus oryzophilus)

Root-feeding insects are important drivers in ecosystems, and links between aboveground oviposition preference and belowground larval performance have been suggested. The root-colonizing arbuscular mycorrhizal fungi (AMF) play a central role in plant nutrition and are known to change host quality for root-feeding insects. However, it is not known if and how AMF affect the aboveground oviposition of insects whose offspring feed on roots. According to the preference–performance hypothesis, insect herbivores oviposit on plants that will maximize offspring performance. In a greenhouse experiment with rice (Oryza sativa), we investigated the effects of AMF (Glomus intraradices) on aboveground oviposition of rice water weevil (Lissorhoptrus oryzophilus), the larvae of which feed belowground on the roots. Oviposition (i.e., the numbers of eggs laid by weevil females in leaf sheaths) was enhanced when the plants were colonized by AMF. However, the leaf area consumed by adult weevils was not affected. Although AMF reduced plant biomass, it increased nitrogen (N) and phosphorus concentrations in leaves and N in roots. The results suggest that rice water weevil females are able to discriminate plants for oviposition depending on their mycorrhizal status. The discrimination is probably related to AMF-mediated changes in plant quality, i.e., the females choose to oviposit more on plants with higher nutrient concentrations to potentially optimize offspring performance. AMF-mediated change in plant host choice for chewing insect oviposition is a novel aspect of below- and aboveground interactions.     

Effects of defoliation on the resistance and tolerance of rice,Oryza sativa,to root injury by the rice water weevil,Lissorhoptrus oryzophilus

Understanding plant‐mediated interactions in agricultural systems may facilitate the development of novel and improved management practices, which is important, as management of these insects is currently heavily reliant on insecticides. The fall armyworm, Spodoptera frugiperda (JE Smith) (Lepidoptera: Noctuidae, Prodeniini), is a sporadic pest of rice fields in the southern USA. In southwestern Louisiana, this defoliating insect typically attacks rice early in the growth season, before fields are flooded. Defoliation by fall armyworm larvae may trigger increased expression of plant defenses, which may result in increased resistance to subsequent herbivores. The rice water weevil (RWW), Lissorhoptrus oryzophilus Kuschel (Coleoptera: Curculionidae, Stenopelmini), enters rice fields as an adult both before and after flooding, but oviposition and larval infestation occur only after fields are flooded. RWW may be affected by changes in plant resistance caused by fall armyworm defoliation before flooding. The objectives of this study were to investigate the plant‐mediated effects of natural and artificial defoliation on population densities of RWW larvae after flooding and on the ability of rice plants to compensate for root injury by RWW larvae. In the 2015 season, fall armyworm defoliation before flooding resulted in reduced RWW densities after flooding. However, in 2016 no significant effects of fall armyworm defoliation on densities of RWW larvae were detected. Similarly, mechanical defoliation of rice before flooding did not affect RWW densities after flooding. Although lowest yields were observed in plots subjected to both root injury and defoliation, there was little evidence of a greater than additive reduction in yields from simultaneous injury. These results suggest a lack of plant‐mediated interactions among these two pests in rice in the southern USA.     

Effects of nitrogen fertilizer applied before permanent flood on the interaction between rice and rice water weevil (Coleoptera: Curculionidae)

Nitrogen fertilizer was applied to rice, Oryza sativa L., before permanent flood to determine the interaction between rice and the rice water weevil, Lissorhoptrus oryzophilus Kuschel (Coleoptera: Curculionidae), through a series of experiments conducted over a 3-yr period in Texas and Louisiana. Both absolute and relative percentage of yield loss because of L. oryzophilus feeding was not affected by fertilizer rates in the Texas experiment. Absolute yield loss increased with nitrogen rates in 2001 and 2002 in the Louisiana experiment; however, percentage of yield loss was not affected. This finding suggests that nitrogen rate does not affect tolerance of rice to L. oryzophilus injury. In the Texas experiment, differences were detected for ratoon crop yield among nitrogen rates and insecticide that were applied at preflood on the main crop, indicating a carryover of fertilizer and insecticide effects from the main to the ratoon crop. L. oryzophilus populations tended to increase with nitrogen fertilizer in the Louisiana experiment. Our results show that farmers should not increase preflood nitrogen fertilizer to increase tolerance of rice to L. oryzophilus injury.     

Effect of nitrogen on resistance of sweet potato to sweetpotato weevil (Coleoptera: Curculionidae) and on storage root chemistry.

The effects of nitrogen fertilizer on sweet potato, Ipomoea batatas (L.) Poir., resistance to the sweetpotato weevil, Cylas formicarius elegantulus (Summers), was studied. Adult weevil feeding and oviposition preference, larval survival, and pupal weight were used as measures of sweet potato resistance. Sweet potato resin glycosides and caffeic acid concentrations in the periderm tissue of storage roots also were measured. Sweet potato genotypes (Beauregard, Excel, W-244, W-250, and Sumor) with varying levels of resistance to sweetpotato weevil were grown in the field under three nitrogen regimes (0, 45, and 135 kg N/ha). Harvested storage roots were evaluated in the laboratory for feeding and oviposition activity of sweetpotato weevil female adults under no-choice and choice test conditions. Larval survival rate and pupal weight were determined by rearing the insects individually on storage root sections. Nitrogen level had a significant effect on the number of eggs deposited, but not on the number of feeding punctures. Sweetpotato weevils laid fewer eggs on plants with the highest level of nitrogen. Nitrogen levels did not significantly affect larval survival and pupal weight. Genotype had a significant effect on feeding, oviposition, and larval survival. Beauregard had higher levels of feeding, oviposition, and larval survival compared with the other genotypes. No interaction effects between nitrogen and genotype were significant. Resin glycosides and caffeic acid concentrations were significantly different among genotypes and between years. Nitrogen levels significantly affected the concentrations of caffeic acid in 1997.     

Feeding and oviposition rates in the pine weevil Hylobius abietis (Coleoptera: Curculionidae)

Feeding and oviposition in the pine weevil Hylobius abietis (Linnaeus) were monitored under laboratory conditions in two long-term experiments lasting over an extended breeding season. Data were also collected from weevils under semi-natural conditions outdoors. In addition, the effects of crowding and starvation were studied in separate experiments. During the main peak oviposition period, female H. abietis consumed 50% more bark tissue than males. When oviposition ceased, the feeding rate of the females declined to the same level as in the males. The rates and spatial distribution patterns of oviposition and feeding were clearly affected by climatic conditions and the degree of crowding. Females were estimated to lay on average 0.8 eggs per day during the season under outdoor conditions. The realized fecundity of a female weevil during the first season was estimated to be approximately 70 eggs. The estimated average rate of feeding was 23 mm2 of Scots pine bark per weevil per day. This implies that planted seedlings can only constitute a minor part of the food resources needed to sustain H. abietispopulations of the size that usually appear on fresh clear-cuttings in northern Europe.     

稻水象甲(Lissorhoptrus oryzophilus Kuschel)滞育征候群中的飞行行为

１９９４－１９９６年对浙江省双季稻区稻水象甲（ＬｉｓｓｏｒｈｏｐｔｒｕｓｏｒｙｚｏｐｈｉｌｕｓＫｕｓｃｈｅｌ）飞行行为的研究表明，稻水象甲飞行扩散的行为特征是：（１）长时间的起飞准备和不高的起飞成功率，每晚的迁出率只有三分之一左右，（２）卵巢，飞行肌呈季节性消长而无局地飞行（ｔｒｉｖｉａｌｆｌｉｇｈｔ）；（３）飞行速度不高，飞行能力不强，且风力稍大便无法起飞，一般情况下不会形成远距离自然扩散，因此     

Role of root systems of eastern larch and white spruce in response to flooding

Abstract. Soil flooding causes rapid reductions in transpiration, stomatal conductance and photosynthesis of many woody plants, which can decrease growth and ultimately result in plant death. This study was conducted to determine the role of the root system in the flooding response. Eastern larch ( Larix laricina ) seedlings were grown in Plexiglas tubes in which water uptake by flooded and unflooded roots was measured independently. Further flooding studies were conducted with eastern larch and white spruce ( Picea glauca ) in which stems were girdled. Root hydraulic properties were analysed using pressure-flow relationships. Transpiration rates of partially flooded plants declined more slowly than fully-flooded plants. Water uptake by unflooded roots of partially flooded seedlings increased momentarily with flooding. After lOd, flooding caused little change in root hydraulic conductance, a decrease in root system reflection coefficient, and an increase in osmotic permeability. Stem girdling had little effect on stomatal conductance and transpiration in comparison to flooding effects. The response of plant tops to flooding appears to be xylem-mediated and in proportion to the amount of root system flooded. Root hydraulic conductance appears to be unaffected by flooding except for a possible temporary increase on the first day following flooding treatments.     

Effect of soil flooding on photosynthesis,carbohydrate partitioning and nutrient uptake in the invasive exotic Lepidium latifolium

Lepidium latifolium L. is an invasive exotic crucifer that has spread explosively in wetlands and riparian areas of the western United States. To understand the ecophysiological characteristics of L. latifolium that affect its ability to invade riparian areas and wetlands, we examined photosynthesis, chlorophyll concentration, carbohydrate partitioning and nutrient uptake in L. latifolium in response to soil flooding. Photosynthesis of flooded plants was about 60–70% of the rate of unflooded controls. Chlorophyll concentrations of flooded plants were about 60–70% of the unflooded plants during 15–50 days of flooding. Flooding resulted in an increase in leaf starch concentration, but root starch concentration was not significantly affected. However, concentrations of soluble sugar were significantly higher in both leaves and roots of flooded plants than unflooded controls. On day 50 after initial flooding, the concentrations of N, P, K and Zn in leaves of flooded plants were lower than in control plants. The concentrations of Mn and Fe in leaves of flooded plants were eight and two times those of control plants, respectively. In contrast, N, P, K and Zn concentrations of roots of flooded plants were slightly higher than in unflooded plants. The concentrations of Fe and Mn in roots of flooded plants were 15 and 150 times those of the control plants, respectively. The transport of P, K, and Zn to shoots decreased and that of Mn increased under flooding. The accumulation of N, K and Zn in roots decreased and that of Mn increased in response to flooding. The results suggested that the maintenance of relatively high photosynthesis and the accumulation of soluble sugar in roots of flooded plants are important adaptations for this species in flooded environments. Despite a reduction in photosynthesis and disruption in nutrient and photosynthate allocation in response to flooding, L. latifolium was able to survive 50 days of flooding stress. Overall, L. latifolium performed like a facultative hydrophyte species under flooding.     

Impact of azadirachtin on vine weevil (Coleoptera: Curculionidae) reproduction

Abstract 1 The dose–response of azadirachtin on vine weevil, Otiorhynchus sulcatus (Fabricius), reproduction is investigated by confining adults to feed on treated Taxus × media leaves, and by counting and evaluating development in the resulting eggs. 2 A dosage‐dependent reduction in oviposition is discovered for foliar surface residues of azadirachtin, with an EC₅₀ of 25–50 parts per million (p.p.m) and 99.2% inhibition of viable egg production with 100 p.p.m. 3 Switching weevils from treated to untreated foliage allows reproductive capability to be restored for weevils that cease egg laying after azadirachtin exposure of 50 p.p.m. Weevils that had already started laying eggs in untreated groups soon cease oviposition once switched to azadirachtin‐treated foliage. 4 A transovarial effect results in a decrease in the percentage of viable eggs as the azadirachtin concentration increases. 5 The amount of feeding on foliage does not appreciably decrease at these hormonally effective concentrations, and adult weevil mortality is only slightly greater in the azadirachtin‐treated groups. Therefore, the overall effect of azadirachtin on weevil populations in the field is difficult to assess, except by collecting weevils to determine whether they are able to lay viable eggs.     

Responses to flooding intensity in Leontodon taraxacoides

Natural flooding is one of the major factors affecting vegetation dynamics in many regions of the world. The Flooding Pampa Grasslands (Argentina) are frequently exposed to flooding events of diverse intensity and duration, some of which Leontodon taraxacoides , an exotic dicot. frequent in these grasslands, seems to survive. Its responses to four different water depths (0, 1, 7 and 13 cm) were studied. The results indicate that plants in conditions of total submergence (depth of 13 cm) did not survive. In less severe flood conditions, increases in the leaf insertion angle resulted in the maintenance of a large proportion of the total leaf area above the water. Differences in leaf length and a decrease in the width and the proportion of lobes per leaf were also found under partial submergence conditions (depth of 7 cm). Root and leaf aerenchyma, present in unflooded plants, showed a significant increase in flood conditions. In spite of the anatomical and morphological responses, total biomass and leaf area were severely affected by water depth. Control plants allocated more biomass to reproductive organs, while partly submerged plants allocated more to leaves and less to reproductive organs. Mature L. taraxacoides plants presented a wide range of plastic adjustment as a survival strategy in soil anaerobiosis, and appear to be able to survive short spring floods in a vegetative state; in contrast, they might not tolerate total submergence conditions imposed by more intense and long-lasting floods.     

Influence of Experimental Flooding on Litter Dynamics in a Rio Grande Riparian Forest,New Mexico

Flow regulation, which has largely eliminated flooding along the Rio Grande in central New Mexico, has substantially changed the riparian ecosystem. We investigated managed flooding as a means of restoring ecosystem function. After collecting baseline data during 1991 and 1992 in two riparian forest sites that had not flooded for about 50 years, we flooded an experimental site for 27–32 days during late spring of 1993, 1994, and 1995, leaving the reference site unflooded. During the final year of the study we compared these sites to two additional sites located within the riverside levee, one of which has been flooding naturally while the second remained largely unflooded. Three years of experimental flooding did not change the total biomass of either woody debris or forest-floor litter at the experimental flood site. Both woody debris and forest-floor litter, however, were significantly lower at the natural flood site than at the experimental flood site and two unflooded sites. Leaf and wood decomposition rates increased with flooding. The decay rate for cottonwood logs at the unflooded site (0.010 per year) predicted a half-life of 69.3 years, while the decay rate of 0.065 per year after 3 years of experimental flooding predicted a half-life of 10.6 years. This suggests that a decade of annual flooding may be used to restore the organic debris to pre-regulation levels. Flooding may also have caused an increase in litter production. These results suggest that experimental flooding has initiated a process of restoring ecosystem function within the riparian forest.     

The growth characteristics and yield potential of rice (Oryza sativa) under non‐flooded irrigation in arid region

The objectives of this field experiment were to study the growth characteristics and yield potential of rice plants under non‐flooded irrigation in arid area. Non‐flooded treatments included drip irrigation with plastic mulching treatments (DIs), furrow irrigation with plastic mulching treatment (FIM) and furrow irrigation with non‐mulching treatment (FIN). Conventional flooded cultivation (F) was check treatment (CK). The four drip irrigation treatments differed in the amount of water applied before and after panicle initiation. Root length density, leaf dry weight, shoot dry weight and root activity were generally higher in the non‐flood‐irrigated treatments (especially the drip‐irrigated treatments) than in the flood‐irrigated treatment at mid‐tillering. However, the growth and development of rice plants were limited after jointing in the non‐flooded irrigation treatments. Increasing the root/shoot ratio and root length density in the 20–40 cm depth and decreasing specific root length at 0–20 cm soil layer were important mechanisms for helping the rice plants to adapt to the non‐flooded environmental stresses. Finally, the grain yield in the non‐flooded irrigation treatments was lower than that in the F treatment. These low yields were mainly attributed to the low root length density at 0–20 cm depth and root activity. Generally speaking, the restricted degrees in the DIs were smaller than that in the FIM and FIN treatments. Among the DIs, both the highest grain yield (8223–8900 kg ha^?1) and the highest water use efficiency (WUE) (0.63) were observed when the soil water content was kept at ?30 kPa before panicle initiation and at ?15 kPa after panicle initiation (referred to as the DI2 treatment). The yield in the DI2 treatment was not significantly different than that in the flood‐irrigated treatment. However, WUE was 2.5 times higher in the DI2 treatment than in the F treatment. These results suggest that drip irrigation technology can be considered as a better water‐saving cultivation of rice plants in arid region.