Genetic Structure of Cochliobolus sativus Populations Sampled from Roots and Leaves of Barley and Wheat in North Dakota

Common root rot (CRR) and spot blotch, caused by Cochliobolus sativus (Ito and Kurib.) Drechsl. ex Dast., are important diseases of barley (Hordeum vulgare L.) and wheat (Triticum aestivum L.) worldwide. However, the population biology of C. sativus is still poorly understood. In this study, the genetic structure of three C. sativus populations, consisting of isolates sampled respectively from barley leaves (BL), barley roots (BR) and wheat roots (WR) in North Dakota, was analysed with amplified fragment length polymorphism (AFLP) markers. A total of 127 AFLP loci were generated among 208 C. sativus isolates analysed with three primer combinations. Gene diversity (H = 0.277–0.335) were high in all three populations. Genetic variation among C. sativus individuals within population accounted for 74%, whereas 26% of the genetic variation was explained among populations. Genetic differentiation was high (ØPT = 0.261, corrected = 0.39), whereas gene flow (Nm) ranged from 1.27 to 1.56 among the three populations analysed. The multilocus linkage disequilibrium (LD) (= 0.076–0.117) was moderate in C  sativus populations. Cluster analyses indicate that C. sativus populations differentiated according to the hosts (barley and wheat) and tissues (root and leaf) although generalists also exist in North Dakota. Crop breeding may benefit from combining genes for resistance against both specialists and generalists of C. sativus.     

Nematicidal activity of culture filtrate of <Emphasis Type="Italic">Beauveria bassiana</Emphasis> against <Emphasis Type="Italic">Meloidogyne hapla</Emphasis>

Culture filtrates of Beauveria bassiana at different concentrations were evaluated for nematicidal activity against the northern root knot nematode (Meloidogyne hapla); bioassays included egg hatching, mortality and infectivity on tomato plants in pots under glasshouse conditions. The percentage mortality and inhibition of hatching of root-knot nematode were directly proportional to the concentration of culture filtrates of B. bassiana. Soil drenching with culture filtrate of B. bassiana significantly reduced nematode population densities in soil and in the roots and subsequent gall formation and egg-mass production by M. hapla under glasshouse conditions.     

Observations on the occurrence, pathogenicity and control of Pratylenchus vulnus, P. thornei and Xiphinema diversicaudatum associated with glasshouse roses

About half the nurseries and glasshouses in the Lea Valley of Hertfordshire were found to be infested with either or both Pratylenchus vulnus and Xiphinema diversicaudatum. The latter nematode probably occurred naturally in the soil on which the glasshouses were built but P. vulnus is thought to have been imported with rose rootstocks and is not known to occur outdoors in Britain. Both nematode species can cause decline of roses and even small numbers seem harmful. Numbers of P. vulnus can increase greatly, especially on rootstocks of Rosa canina, although numbers were often smaller when roses were severely damaged than on healthier crops. Numbers of X. diversicaudatum increased more slowly and R. canina was a better host for it than R. chinensis. X. diversicaudatum seemed to respond less quickly than P. vulnus to reduced host-plant vigour. The two rootstocks exhibited different host-status for the two species of nematode and cuttings of various rose cultivars showed different host-status to P. thornei, another species of lesion nematode which is not known to be pathogenic to roses. Two applications of dibromochloropropane liquid at the rate of 70 l/ha in a large volume of water maintained nematode densities at an acceptably low level, and growers who adopted this treatment as a supplement to pre-planting steam sterilisation and/or DD (dichloro-propane: dichloropropene) prolonged the productive life of their crops for several years.     

Differential Reaction of Alfalfa Cultivars to Meloidogyne hapla and M. chitwoodi Populations

Meloidogyne hapla reproduced and suppressed growth (P < 0.05) of susceptible Lahontan and Moapa alfalfa at 15, 20, and 25 C. At 30 C, resistant Nevada Syn XX lost resistance to M. hapla. M. hapla invaded and reproduced on Rhizobium meliloti nodules of Lahontan and Moapa, inducing giant cell formation and structural disorder of vascular bundles of nodules without disrupting bacteroids. At 15, 20, and 25 C a M. chitwoodi population from Utah reproduced on Lahontan, Moapa, and Nevada Syn XX alfalfa, suppressing growth (P < 0.05). Final densities of the Utah M. chitwoodi population were greater (P < 0.05) than those of Idaho and Washington State populations on Lahontan at 15 and 25 C and on Nevada Syn XX at 15 C, but were less consistent and smaller (P < 0.05) than those of M. hapla on Lahontan and Moapa at 20 and 25 C. Inconsistent reproduction of the Utah M. chitwoodi population on alfalfa suggests the possible existence of nematode strains revealed by variability in alfalfa resistance. No reproduction or inconsistent final nematode population densities with no damage were observed on Lahontan, Moapa, and Nevada Syn XX plants grown in soil infested with Idaho and Washington State M. chitwoodi populations.     

Interrelationships of Meloidogyne Species with Flue-cured Tobacco

Microplot and field experiments were conducted to determine relationships of population densities of Meloidogyne spp. to performance of flue-cured tobacco. A 3-yr microplot study of these interactions involved varying initial nematode numbers (P_i).and use of ethoprop to re-establish ranges of nematode densities. Field experiments included various nematicides at different locations. Regression analyses of microplot data from a loamy sand showed that cured-leaf yield losses on ''Coker 319'' for each 10-fold increase in P_i were as follows: M. javanica and M. arenaria—-13-19%; M. incognita—5-10%; M. hapla—3.4-5%; and 3% for M. incognita on resistant ''Speight G-28'' tobacco. A P_i of 750 eggs and larvae/500 cm³ of soil of all species except M. hapla caused a significant yield loss; only large numbers of M. hapla effected a loss. M. arenaria was the most tolerant species to ethoprop. Root-gall indices for microplot and most field-nematicide tests also were correlated negatively with yield. Relationships of P_i(s) and necrosis indices to yield were best characterized by linear regression models, whereas midseason numbers of eggs plus larvae (P_m) and sometimes gall indices vs. yield were better characterized by quadratic models. The relation of field P_m and yield was also adequately described by the Seinhorst model. Degrees of root galling, root necrosis, yield losses, and basic rates of reproduction on tobacco generally increased from M. hapla to M. incognita to M. arenaria to M. javanica.     

Influence of Meloidogyne hapla on Alfalfa Yield and Host Population Dynamics

Self-thinning in alfalfa, a dynamic process involving the progressive elimination of the weakest plants, was enhanced by Meloidogyne hapla. Alfalfa stand densities decreased exponentially with time and were reduced 62% (P = 0.05) in the presence of M. hapla. As stand densities decreased over time, mean plant weights increased at a rate 2.59 times faster in the absence of M. hapla. In a stepwise multiple regression analysis, 65% of the total variation in yield could be explained by changes in stand density and 85% by average weight of individual stems. Alfalfa yields were suppressed (P = 0.05) by M. hapla, with suppression generally increasing with time and as the nematode population density increased. Yield suppression was attributable primarily to the decline in plant numbers and to suppression in individual plant weights.     

Assessment of yield loss of wheat cultivars caused by Heterodera filipjevi under field conditions

Among the cereal cyst nematodes, Heterodera filipjevi is the dominant species of Cereal cyst nematodes (CCNs) in most cereal growing areas of Iran. To evaluate the impact of H. filipjevi on wheat cultivars, a field trial was performed in two infested fields in Isfahan province, Iran 2014 and 2015. The trials were conducted in a factorial experiment based on the complete randomized block design. The treatments consisted of three winter wheat cultivars (Back‐cross Rowshan, Pishtaz and Parsi) which were planted with and without applying nematicide aldicarb Each cultivar was planted in 6 m² (2 × 3 m) plots which were replicated five times. The nematode reproduction factor was calculated after determining the initial and final population of H. filipjevi in each plot before sowing the seeds and after harvesting. The grain yields and growth parameters were recorded and the variables of 2 years experiment evaluated by linear regression analysis. Cultivar × nematicide treatment combinations in the first and second years showed that H. filipjevi significantly affected grain yield and growth parameters in all three cultivars. The results revealed significant reduction of grain yield by 24.8%, 24.8% and 20.4% in Back‐cross Rowshan, Pishtaz and Parsi cultivars, respectively. The nematode reproduction factor ranged from 0.32 to 1.76 in plots plus and minus nematicide application, respectively. The analysis showed linear inverse relationships between the initial population (Pi) and the yields of wheat cultivars in check plots without aldicarb application.     

Host Suitability of Twelve Leguminosae Species to Populations of Meloidogyne hapla and M. chitwoodi

Legumes of the genera Astragalus (milkvetch), Coronilla (crownvetch), Lathyrus (pea vine), Lotus (birdsfoot trefoil), Medicago (alfalfa), Melilotus (clover), Trifolium (clover), and Vicia (common vetch) were inoculated with a population of Melaidogyne chitwoodi from Utah or with one of three M. hapla populations from California, Utah, and Wyoming.Thirty-nine percent to 86% of alfalfa (M. scutellata) and 10% to 55% of red clover (T. pratense) plants survived inoculation with the nematode populations at a greenhouse temperature of 24 ± 3°C. All plants of the other legume species survived all nematode populations, except 4% of the white clover (T. repens) plants inoculated with the California M. hapla population. Entries were usually more susceptible to the M. hapla populations than to M. chitwoodi. Galling of host roots differed between nematode populations and species. Root-galling indices (1 = none, 6 = severely galled) ranged from 1 on pea vine inoculated with the California population of M. hapla to 6 on yellow sweet clover inoculated with the Wyoming population of M. hapla. The nematode reproductive factor (Rf = final nematode population/initial nematode population) ranged from 0 for all nematode populations on pea vine to 35 for the Wyoming population of M. hapla on alfalfa (M. sativa).     

Effects of high grafting on tomato plants infected by Meloidogyne incognita and Ralstonia solanacearum

The root‐knot nematode Meloidogyne incognita is known to increase the severity of bacterial wilt in many solanaceous crops. In Japan, several bacterial wilt‐resistant rootstocks that have the M. incognita resistance (Mi) gene in their genome have been developed for tomatoes. In this study, we aimed to examine whether the presence of Mi gene‐breaking M. incognita population affects the development of bacterial wilt in bacterial‐wilt‐resistant tomato rootstocks with Mi in their genetic background. We also aimed to examine the possibility of using high‐grafted tomatoes to control bacterial wilt in plants infected by M. incognita. Our results indicate that the resistance to bacterial wilt was easy to break in usual‐grafted tomato plants infected with M. incognita and that M. incognita enhanced the vertical movement of Ralstonia solanacearum in the bacterial‐wilt‐resistant tomato rootstocks. In addition, our results suggest that high grafting led to significantly less wilting in the plants infected by M. incognita than did usual grafting.     

Effect of Inoculum Density on Verticillium Wilt Incidence in Commercial Olive Orchards

Verticillium wilt, caused by Verticillium dahliae Kleb., is presently the most destructive disease of olive, particularly in Andalucía (southern Spain). ‘Picual’ and ‘Arbequina’ are the dominant cultivars being planted in Spain. Both cultivars are highly susceptible to the defoliating pathotype of V. dahliae when artificially inoculated by root‐dipping or stem injection. Conversely, ‘Arbequina’ is considered more resistant than ‘Picual’ based on field observations and farmer's experience. In this study, the differential reaction between of cultivars was confirmed by surveys of naturally infested orchards with different inoculum densities of the pathogen. The average percentage of affected olive trees of ‘Picual’ was 60.2%, while only 13.1% of trees of ‘Arbequina’ showed disease symptoms. Overall, the pathogen caused extensive wilting of branches and defoliation on the trees of ‘Picual’, whereas ‘Arbequina’‐infected trees showed chlorotic symptoms and slight defoliation. The relationship between inoculum density and disease incidence fit a logarithmic function for both cultivars. The percentage of affected trees of ‘Arbequina’ per year increased linearly (y = 0.3559x, R² = 0.5652, and P = 0.0195) with the inoculum density in the soil, whereas this relationship was not observed for the ‘Picual’. Planting density had no effect on disease incidence for any of the two cultivars.     

Integrative taxonomy of the Rhinolophus macrotis complex (Chiroptera,Rhinolophidae) in Vietnam and nearby regions

The taxonomic status of Rhinolophus macrotis sensu lato (s.l.) in Vietnam and adjacent territories remains problematic. To address this issue, we performed an integrated study of morphological, acoustic, and genetic characters of R. macrotis s.l. specimens and compared these with sympatric species within the philippinensis group (R. marshalli, R. paradoxolophus, and R. rex). Our results reveal that in addition to a cryptic species of R. macrotis previously found in Jiangxi and Jingmen, China, R. macrotis s.l. in continental Asia includes three further species, namely R. cf. siamensis, R. cf. macrotis, and R. cf. macrotis “Phia Oac.” These four taxa are distinguished from genuine R. macrotis in Nepal and R. siamensis in Thailand by their morphological and/or genetic features. Further taxonomic evaluation of the subspecies of R. macrotis s.l. is needed to determine their affinities with recently recognized cryptic species and to possibly describe new taxa. Our results also show that interspecific divergences in mitochondrial DNA sequences (Cytb and COI genes) among taxa within the philippinensis group (particularly between R. cf. siamensis/R. cf. macrotis and R. rex/R. paradoxolophus) are significantly lower than those of other morphological groups in the genus. These phylogenetic patterns might be explained by recent allopatric speciation or ancient introgression events among ancestors of the taxa during the Pleistocene. However, further investigations including genetic analyses of nuclear genes are needed to test the latter hypothesis.     

Long term dynamics of Rosa multiflora in a successional system

Long term studies of invasion dynamics are critical in developing a more complete understanding of the factors that influence species spread. To address this issue, the dynamics of the non-native invasive plant, Rosa multiflora, were examined using a 40-year record of successional change. The roles of biotic and abiotic factors in regulating R. multiflora invasion were also assessed. The invasion showed an initial 9-year time lag, followed by a 20-year period of population expansion and an ultimate decline as succession progressed. During all phases of R. multiflora’s invasion, there was continuous turnover within plots. Rainfall during the previous season was found to increase R. multiflora colonization during population expansion while tree species inhibited the invader’s growth. During expansion and decline of R. multiflora, common associated species were often positively or negatively correlated with changes in R. multiflora cover. Though early population dynamics were regulated by propagule pressure, the major influence on R. multiflora late in succession was canopy closure. Although the invasion of this species was largely self-limiting in this system, the species is likely to persist within late successional systems and may require management intervention.     

Role of Nematodes,Nematicides, and Crop Rotation on the Productivity and Quality of Potato,Sweet Potato,Peanut, and Grain Sorghum

The objective of this experiment was to determine the effects of fenamiphos 15G and short-cycle potato (PO)-sweet potato (SP) grown continuously and in rotation with peanut (PE)-grain sorghum (GS) on yield, crop quality, and mixed nematode population densities of Meloidogyne arenaria, M. hapla, M. incognita, and Mesocriconema ornatum. Greater root-gall indices and damage by M. hapla and M. incognita occurred on potato than other crops. Most crop yields were higher and root-gall indices lower from fenamiphos-treated plots than untreated plots. The total yield of potato in the PO-SP and PO-SP-PE-GS sequences increased from 1983 to 1985 in plots infested with M. hapla or M. arenaria and M. incognita in combination and decreased in 1986 to 1987 when root-knot nematode populations shifted to M. incognita. The total yields of sweet potato in the PO-SP-PE-GS sequence were similar in 1983 and 1985, and declined each year in the PO-SP sequence as a consequence of M. incognita population density increase in the soil. Yield of peanut from soil infested with M. hapla increased 82% in fenamiphos-treated plots compared to untreated plots. Fenamiphos treatment increased yield of grain sorghum from 5% to 45% over untreated controls. The declining yields of potato and sweet potato observed with both the PO-SP and PO-SP-PE-GS sequences indicate that these crop systems should not be used longer than 3 years in soil infested with M. incognita, M. arenaria, or M. hapla. Under these conditions, these two cropping systems promote a population shift in favor of M. incognita, which is more damaging to potato and sweet potato than M. arenaria and M. hapla.     

Efficacy of commercial formulations of entomopathogenic nematodes against tropical sod webworm,Herpetogramma phaeopteralis (Lepidoptera: Crambidae)

Large quantities of insecticides are used on warm season turfgrasses to combat pest infestations. To investigate the potential for microbial control, we screened commercially available entomopathogenic nematode products against Herpetogramma phaeopteralis Guenée, an economically injurious pest in the south‐eastern United States and Caribbean islands. All tested products, based on Steinernema carpocapsae, S. feltiae, Heterorhabditis bacteriophora, H. megidis and H. indica, were pathogenic to H. phaeopteralis larvae in the laboratory, but S. carpocapsae caused the highest mortality. Amongst nematode species, median lethal concentration (LC₅₀) was not different for three different larval sizes (based on 95% CL) with the exception of H. indica, which had higher LC₅₀ for small larvae. The number of infective juvenile stages (IJs) produced per White trap was significantly greater from larvae infected by H. bacteriophora and least for those infected by H. indica. A proprietary formulation of S. carpocapsae ‘Millenium^®’ was chosen for further greenhouse experiments. Overall, the neonicotinoid insecticide clothianidin provided the best control, but greenhouse experiments also revealed that the label rate of Millenium (10⁶ IJ/l at 2500 l/ha) reduced webworm populations by 83–93% and was as effective as clothianidin against larger‐size larvae. Our data suggest that commercial formulations of S. carpocapsae can be a good option for H. phaeopteralis biocontrol, but further field studies are warranted to confirm effectiveness under different environmental scenarios.     

Effects of Population Densities of Meloidogyne hapla on Growth and Yield of Tomato

Growth and yield of ''Veebrite'' tomato were studied in 20-cm (i.d.) clay-tile microplots containing initially 260, 1,840, 6,120, or 27,950 Meloidogyne hapla larvae/kg of soil. Low nematode numbers stimulated, and the highest nematode population suppressed, vegetative plant growth. More tomatoes, with a higher total weight, were harvested from plants infested with 260 and 1,840 nematode larvae at planting than from those with initial densities of 6,120 and 27,950 larvae. At the two highest densities, the cumulative fruit production (weight) was suppressed by 10% and 40%, respectively. The increase in growth and yield at the lower densities appeared to be due to an increase in the size of the root systent. However, at the higher densities, yield was no longer directly related to root weight. The reproduction factor of M. hapla was negatively correlated with initial density; for the lowest and highest initial densities, it was 96X and 7X at midseason, and 354X and 3X at harvest, respectively. The equilibrium density was 63,000 larvae/kg of soil; initial densities larger than 2,000 larvae/kg of soil may require control.     

Damage functions and thermal requirements of Meloidogyne javanica and Meloidogyne incognita on watermelon

The relationship between the initial (P_i) and final (P_f) population densities of Meloidogyne javanica and yield of watermelon, Citrullus lanatus, cv. Sugar Baby were determined in pot and field experiments. In the pots, the maximum reproduction rate of the nematode was 14, and the equilibrium density was 49 400 eggs/100 cm³ of soil. Yield data represented as fresh top weight fitted the Seinhorst damage function (P < 0.001, R² = 0.7), and the minimum relative yield (m) was 0.65 at P_i ≥ 3200 eggs/100 cm³ of soil and the tolerance limit (T) 74 eggs/100 cm³. In the field experiments (2011 and 2012), the maximum reproduction rate was 73 and 70, and the equilibrium density 32 and 35 second‐stage juveniles (J2)/100 cm³ soil. Yield data represented as fruit weight fitted the Seinhorst damage function in 2011 (P < 0.001, R² = 0.92) and the m‐ and T‐values were 0.63 and 20 J2/100 cm³ of soil, respectively. Meloidogyne incognita and M. javanica needed similar length of time for development to egg‐laying females and life cycle completion at 24.4°C.     

Effect of the rhg1 Gene on Population Development of Heterodera glycines

The effect of the rhg1 gene on equilibrium population densities (E) and reproduction factors (Rf) of Heterodera glycines was studied by comparing the nematode population development on two near-isogenic soybean lines (NIL), differing at the rhg1 locus. The NIL were inoculated with a series of initial egg densities (Pi) in the greenhouse. The relationships between final population densities (Pf = females per plant or eggs per plant) or Rf (final egg density/Pi) on both NIL and Pi were adequately described by quadratic models. The rhg1 gene suppressed Pf and Rf at all Pi of a population of H. glycines race 3 (HG Type 0-); E and maximum Rf were higher on the NIL-S line than on the NIL-R line. After two generations of culture of the race 3 population on the NIL-R line, the population selected by the rhg1 gene (R-eggs) had higher Pf and Rf on the NIL-R line than the population cultured on the NIL-S line (S-eggs) at all Pi. Both R-eggs and S-eggs produced similar egg numbers on the NIL-S line, which was higher than the egg number of either population on the NIL-R line at all Pi. The ratio of E in female numbers on the NIL-R line to E on the NIL-S line increased from 29% for the original race 3 population (S-eggs) to 46% for the rhg1-selected population (R-eggs). Regardless of different egg sources, a trend of increase in the number of eggs per female with the rise of Pi was observed on the NIL-S line. In contrast, female fecundity of both populations declined with the increase of Pi on the NIL-R line. At most inoculum densities, the highest number of eggs per female was observed on the NIL-S line inoculated with the R-eggs, whereas the lowest number of eggs per female was detected on the NIL-R line inoculated with the S-eggs. This study demonstrated that the E and maximum Rf determined by the quadratic models are useful measurements of plant resistance to nematodes.     

Influence of Pratylenchus vulnus and Meloidogyne hapla on the Growth of Rootstocks of Rose

Pratylenchus vulnus is involved in a desease of Rosa noisettiana ''Manetti'' rose rootstock characterized by darkening of roots, death of feeder roots, and stunting of entire plants. The disease is more severe when plants are grown in silt loam soil than when they are grown in sandy loam soil. The nematodes reproduce best in silt loam soil at 20 C. Meloidogyne hapla did not affect the growh of Manetti. Rosa sp. ''Dr. Huey'', Manetti, and R. odorata rose rootstocks were found to be goos hosts for P. vulnus whereas R. multiflora was less suitable. M. hapla reproduced well on R. odorata, Dr. Huey, and R. multiflora, but not on Manetti.     

Influence of Glomus fasciculatum and Meloidogyne hapla on Allium cepa in Organic Soils

The influence of Meloidogyne hapla and Glomus fasciculatum on Allium cepa (onion) grown in organic soil was evaluated under greenhouse conditions. In the absence of G. fasciculatum, M. hapla significantly retarded the growth of A. cepa cv. Krummery Special and MSU 8155 × 826, but had no detrimental influence on Downing Yellow Globe, Spartan Banner, or Spartan Sleeper. All five cultivars maintained populations of M. hapla, Final root population densities of M. hapla associated with Spartan Banner, Krummery Special, MSU 8155 × 826, and Spartan Sleeper were significantly greater than those recovered from Downing Yellow Globe. Final root population densities of M. hapla were directly proportional to the initial population densities. Root colonization of onion by G. fasciculatum significantly enhanced the growth and development of Downing Yellow Globe. The rate of increase of A. cepa growth and the final spore density were directly proportional to the initial spore density of G. fasciculatum. Final population densities of M. hapla in the presence of G. fasciculatum were generally greater than in the absence of the fungus. After 15 weeks, A. cepa plants grown in the presence of both M. hapla and G. fasciculatum were significantly larger than those grown in the presence of only M. hapla.     

Valuable New Leaf or Inflorescence Resistances Ensure Improved Management of White Rust (Albugo candida) in Mustard (Brassica juncea) Crops

Field resistances/susceptibilities against Albugo candida race 2V were determined for 29 Indian Brassica juncea varieties and compared with resistant varieties from China (6) and Australia (7). ‘Basanti’ (AUDPC incidence 46.7; AUDPC severity 29.2) represents the first high‐level resistance to race 2V in Indian varieties. Several others showed lower but still useful levels of resistance, including Narendra Ageti Rai‐4 (AUDPC incidence 150.6; AUDPC severity 66.8) and JM1 (AUDPC incidence 167.1; AUDPC severity 83.7). Highly susceptible Indian varieties had AUDPC incidence values >200 and severity >100. ‘Basanti’ had least stagheads/plant (0.32), while Narendra Ageti Rai‐4 had lowest % plants with stagheads (2.48). In contrast, almost half of Indian varieties had stagheads/plant >1 and % plants with stagheads >4, and >26 for ‘Kranti’. The resistance in ‘Basanti’ paves the way forward towards significantly improved white rust management in mustard in India. JM06011, JM06021, JR049 from Australia and CJB‐003 from China had zero leaf incidence. There were significant (P < 0.001) relationships between disease incidence with severity (R² 0.92), stagheads/plant (R² 0.69) and also % plants with stagheads (R² 0.60); between disease severity with stagheads/plant (R² 0.68) and also % plants with stagheads (R² 0.69); and between stagheads/plant with % plants with stagheads (R² 0.59).