Comparative life table analysis of Tetranychus urticae Koch (Acari: Tetranychidae) on ten rose cultivars

Life history parameters of plant feeders are useful tools to evaluate resistance or susceptibility of host plants including different cultivars. This study compared population growth parameters of Tetranychus urticae Koch on 10 rose cultivars, including Bella Vita, Cool Water, Dolce Vita, Maroussia, Orange Juice, Pink Promise, Roulette, Tea, Valentine, and Persian yellow in laboratory conditions at 24±1 °C, 65±5% relative humidity, and a photoperiod of 16:8 (L:D) h. The results revealed that mite survival rate varied from 66.5% on Bella Vita to 85.9% on Persian yellow. The immature development time was different among the tested rose cultivars and ranged from 9.35 days on Orange Juice to 12.30 days on Bella Vita. The highest fecundity rate was recorded on Pink Promise. Consequently, population growth parameters were also significantly affected. The lowest intrinsic rate of increase (rm) was recorded on Roulette and this parameter was relatively higher on Cool Water, Orange Juice, and Persian yellow. In addition, the highest net reproductive rate (R0) was observed on Pink Promise, which was significantly higher than Roulette, Tea, and Valentine cultivars. The longest mean generation time (T) was calculated on Roulette and the shortest on Cool Water, Tea, and Orange Juice. The lowest performance of the two-spotted mite on Roulette could indicate that this is a suitable cultivar against mite infestation. Differences in mite susceptibility of tested rose cultivars here highlighted have the potential to be used for integrated pest management of T. urticae in ornamental rosa cultivations.


INTRODUCTION
The cultivation of ornamental plants that are used as cut flowers, flowering, and non-flowering potted plants is an important component in agriculture (Peronti and Sousa-Silva, 2002). Among the ornamental plants, rose plants of Rosa spp. (Rosaceae) with their showy and often fragrant blooms are important ornamental shrubs and an exceptionally valuable decorative element of green areas in the cities (Jaskiewicz, 2006;Bidarmania et al., 2015). Dif-ferent rose species and cultivars are cultivated because of their habits, effective coloring of the leaves in different seasons, as well as their decorative fruit. In addition to ornamental aspects of rose plants, valuable rose oils distilled from rose petals are used for the production of medicinal agents, perfumes, cosmetics, and other aromatherapy products utilized in the industry (Seneta and Dolatowski, 2003;Jaskiewicz, 2006;Demirozer, 2012).
Rose is the most popular perennial ornamental http://www1.montpellier.inra.fr/CBGP/acarologia/ ISSN 0044-586X (print). ISSN 2107-7207 (electronic) 607 plant all over the world (Bidarmania et al., 2015). Rose plants are cultivated in most countries in the northern hemisphere such as in China, France, Bulgaria, Egypt, India, Morocco, Turkey, and Iran (Demirozer, 2012). Roses are susceptible to several pests and diseases that reduce flower growth and quality. The phytophagous two-spotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae), has a broad host-plant range and is a serious pest of several crops worldwide (Tsagkarakou et al., 2002;Tehri, 2014). This mite is a generalist species that can feed on hundreds of host plants and causes considerable damage to field, greenhouse and horticultural crops, as well as ornamental and fruit-bearing trees including Rosa (Helle and Sabelis, 1985;Sedaratian et al., 2010;Jafari et al., 2012;Jalalvandi et al., 2015).
Feeding activity of spider mite leads to the appearance of typical yellow chlorotic spots on leaves with profuse webbing on the underside of leaves. In severe infestations, leaves may fall and the flowering may be considerably reduced (Khodayari et al., 2008). The short life span and high reproductive potential causes a fast-growing population that allows the mite to achieve economic injury level in appropriate conditions, resulting in a rapid decline of host plant yield and quality (Carey and Bradley, 1982;Fathipour et al., 2006).
The control of this mite has relied on conventional chemical control (Badii et al., 2004), with frequent pesticide applications to achieve effective control. However, due to fast development rate, high fecundity and ability to develop resistance to most available pesticides (Cranham and Helle, 1985;Hoyt et al., 1985), this mite is difficult to control with pesticides. Therefore, the use of alternative control measures is needed against this mite due to the difficulties associated with its control and the huge economic losses it causes on many crops. Host plant resistance is an alternative, safe procedure for limiting herbivores because it is economically and environmentally secure (Wermilinger et al., 1991;Panda and Khush, 1995;Li et al., 2004;Zehnder et al., 2007). Three mechanisms, called antixenosis, antibiosis, tolerance, or some combinations are known as being involved in plant resis-tance to pests (Painter, 1951;Kogan and Ortman, 1978;Smith, 2005). Among them, antibiosis is the most important. It has a direct effect on the survival and developmental rates, adult longevity, and fecundity (Sedaratian et al., 2010;Khanamani et al., 2012).
At present, T. urticae is one of the most important pests of rose shrub all over the world (Grbic et al., 2011;Bidarmania et al., 2015). Despite the economic importance of this mite, little information is available concerning the development and reproduction of this pest on various rose cultivars. Hence, in current research, it was planned to investigate the development, longevity, and reproductive potential of T. urticae in relation to different rose cultivars affecting population growth. Results here presented provide complementary knowledge on resistant rose cultivars which could be useful in predicting the spider mite population growth. It can then contribute to enhance integrated pest management programs for this mite in urbanized areas and commercial Rosa potting breedings.

Stock cultures
The shrubs of ten rose cultivars used in this study were obtained from Rose Farm Incorporation in Tehran, Iran. They are the most commonly cultivated rose cultivars in commercial floricultures and urbanized areas in Iran. All the rose cultivars were imported ones, including Bella Vita, Cool Water, Dolce Vita, Maroussia, Orange Juice, Pink Promise, Roulette, Tea, Valentine, and Persian yellow. All the selected cultivars had similar external characteristics such as leaf color and density of leaf pubescence, except for flower size and color. All the plants were kept in a greenhouse at 25±5°C, 50-70% relative humidity, and natural light conditions (14-16 hours of light during plant growing season in Ardabil, northwest of Iran). All rose cultivars were grown in 30 cm pots filled with suitable soil, which was composed of 60% loam soil with 40% additives such as peat, sand, sawdust, and manure to improve physical soil traits. In order to establish spider mite colonies, two shrubs of each rose cultivar were kept in large wood-framed cubic cages (2×2×1 m) and were infested with T. urticae. The mites used in the experiments were originally collected from rose shrubs grown in university campus in Ardabil, Iran, during May 2012 and were transferred onto related host plants in the greenhouse. Spider mites were reared for at least four generations before they were used for the experiments.

Development and mortality
All the experiments were conducted in a growth chamber at 24±1°C, 60±5% relative humidity, and a photoperiod of 16:8 (L:D) h. To evaluate the developmental time and immature survival rate, adult mites were randomly selected from the rearing colonies and placed on the leaf discs of host plant using a fine-hair brush. Leaf discs of host plants were cut and placed upside down on a cotton layer in petri dishes (8 cm diameter × 2 cm height) with an opening in the center of the lid for ventilation. The discs were surrounded with wet cotton to keep the leaf discs fresh and avoid the escape of mites. Three terminal branches of rose bushes were selected for testing. The last three leaves from apical part on each of the selected branches were cut and each leaf was confined inside a petri dish.
In order to do the experiments, pairs of female and male mites were randomly selected from the colony and transferred onto each leaf disc. The mites were allowed to mate and lay eggs for 24 hours. After this period, a newly laid egg was selected to remain in the cage and all the other adults and eggs were removed. Overall, there were five potted plants for each rose cultivar to evaluate the development and survival rate of 60 individualized eggs per treatment. These leaf discs were checked daily under a stereomicroscope and mortality and survivorship of different stages were recorded. The observations continued until the emergence of adults.

Reproduction and life table parameters
After maturity and adult appearance, the females were coupled with males obtained in the same experiment on respective cultivars. Each pair of female and male adults in a petri dish was considered as a replication unit and there were a total of 20 replicates for each examined cultivar. Daily observations were made to determine female fecundity and survivorship of individuals until the death of the last female mite. The reproduction period, adult longevity, and fecundity rate were determined. The following population growth parameters were calculated from fertility and survivorship schedules (Birch, 1948;Carey, 1993): intrinsic rate of increase (r m ), mean generation time (T), finite rate of increase (ń), net reproduction rate (R 0 ), gross reproduction rate GRR), and doubling time (DT).

Statistical analysis
Effect of rose cultivars on developmental time, reproduction period, adult longevity, and fecundity was analyzed using one-way ANOVA. If significant differences were detected, multiple comparisons were made using Student-Newman-Keuls (SNK) method (P < 0.05). Statistical analysis was performed using SPSS statistical software version 16.0 (SPSS, 2007). Differences in R 0 , T, ń, DT, and r m values were tested for significance by estimation of variances through Jackknife procedure (Maia et al., 2000).

Development and mortality
Tetranychus urticae successfully developed to adulthood on all rose cultivars examined and the results of its biological attributes, including immature survival rate and developmental times, are presented in Table 1. The survivorship of immature stages (from egg to adult) was ranged from 66.50% on Bella Vita to 85.92% on Persian yellow. Development times of all immature stages, including incubation period (F = 6.261; df = 9,528; P < 0.01), larval period (F = 7.671; df = 9,499; P < 0.01), and nymphal period (F =9.285; df = 9,453; P < 0.01) were affected by the rose cultivar. Moreover, total development time (from egg to adult) of T. urticae was significantly affected by host plants (F = 32.136; df = 9,453; P < 0.01) and this mite developed faster on Orange Juice cultivar. Total developmental  Figure 1. Age-specific survival rate generated similar curves among the cultivars. However, the declining trend of survival rate was slower on Pink Promise than the other cultivars.

Adult longevity and fecundity
Different rose cultivars significantly affected the longevity and fecundity of T. urticae ( Table 2). The longevity of female T. urticae adults feeding on rose cultivars was different (F = 13.099; df = 9,190; P < 0.01). The adult females that spent their immaturity period on Pink Promise lived for a long time. Similarly, oviposition period of the spider mite varied between the rose cultivars (F = 8.912; df = 9,190; P < 0.01) and ranged from 13.35±0.47 days on Bella Vita cultivar to 23.05±1.64 days on Pink Promise cultivar ( Table 2). The egg hatching rate on different rose cultivars varied from 70.0% on Bella Vita and Valentine to 86.0% on Persian yellow cultivar. Sex ratio of offspring was influenced by rose cultivars and ranged from 65.0% on Roulette to 75.0% on Cool Water cultivar. There was a significant difference in the total fecundity rate of T. urticae on the tested cultivars (F = 8.979; df = 6, 190; P < 0.01) and the highest value was 125.2±10.1 (eggs/female) on Pink Promise cultivar. Figure 1 shows the age-specific fecundity (mx) curves of T. urticae on different rose cultivars. These curves revealed that the age-specific fecundity schedule fluctuated throughout the oviposition period; however, a clear peak is evident at the beginning of the oviposition period on most cultivars. These curves showed an asymmetrical pattern, which skewed toward older individuals.

Life table parameters
The values of life table parameters of T. urticae on different rose cultivars are listed in Table 3. The gross reproduction rate (GRR) of spider mite was different between rose cultivars (F = 2.674; df = 9,190; P < 0.01) and the mite reared on Pink Promise cultivar showed the highest GRR value (75.15±7.59 eggs) but those on Cool Water cultivar had the lowest GRR value (49.22±3.90 eggs). The net reproductive rate (R 0 ) of T. urticae showed a significant difference (F = 2.988, df = 9,190, P < 0.01) between the rose cultivars tested. The cohort reared on Pink Promise had the highest value of R 0 and the lowest values of this parameter were calculated on Roulette, Tea, and Valentine cultivars. The intrinsic rate of increase (r m ) values differed between the ten rose cultivars (F = 20.488; df = 9,190; P < 0.01). The highest and lowest r m values were observed on Cool Water (0.253±0.005) and Roulette (0.169±0.004), respectively. The mean generation time (T) was also found to be significantly different on the rose cultivars (F = 22.216; df = 9,190; P < 0.01) and our results revealed that the shortest generation time was on Cool Water and the longest on Roulette cultivar. A similar trend was displayed for doubling time (DT), which exhibited a significant difference on rose cultivars (F = 120.179; df = 9,190; P < 0.01). In addition, the finite rate of increase (ń) of T. urticae was significantly different between the rose cultivars tested (F = 20.473; df = 9,190; P < 0.01) and was higher on Cool Water.

DISCUSSION
Host plant resistance has been effectively used in developing pest management programs for several crop pests (Wilson, 1994;Dent, 2000;Sarfraz et al., 2007). It is well recognized that the host plant quality can affect life history traits of their herbivore insects by impairing growth, increasing mortality, and reducing the fecundity rate (Price et al., 1980). Host plants are different in chemical and morphological characteristics that influence their suitability as hosts for different herbivore insects (Ave and Tingey, 1986;Storer and van Emden, 1995;Awmack and Leather, 2002;Golizadeh et al., 2016).
Our study showed that the performance of T. urticae varied considerably depending upon rose cultivars. The lower immature survival rate of spider mite was observed on Bella Vita. Similarly, a lower immature survival rate was reported for rose aphid, Macrosiphum rosae on Bella Vita cultivar (Golizadeh et al., 2016). This result could indicate that Bella Vita is a cultivar with poor nutritional quality or with secondary plant substances for these two important pests of rose plants. Moreover, T. urticae showed the longest developmental time on Bella Vita cultivar. On the other hand, the population maintained on Orange Juice cultivar had a comparatively faster developmental rate. Faster developmental rate on a particular host cultivar may allow a short life cycle and rapid population growth, which would be subsequently reflected in final population size. Shorter developmental times and higher reproduction of herbivore insects on a host plant indicate higher susceptibility of a host plant (van Lenteren andNoldus, 1990, Liu et al., 2004;Golizadeh and Razmjou, 2010;Golizadeh et al., 2016).
In this study, the females reared on Pink Promise cultivar showed the highest fecundity rate and longevity. The higher fecundity rate on Pink Promise cultivar suggested that the quantity and/or quality of nutritional contents were more appropriate for T. urticae than the other rose cultivars. Different fecundity rates have been reported for spider mite on various soybean cultivars (Sedaratian et al., 2010), eggplant cultivars (Khanamani et al., 2012), cucumber accessions (Shoorooei et al., 2012), ornamental crop Gerbera (Krips et al., 1998), bean varieties (Fathipour et al., 2006;Ahmadi et al., 2007) and various leguminous plants (Razmjou et al., 2009). Presumably, larval and nymphal feedings on nutritionally poor plants can reflect lower female fecundity on a plant (Verkerk and Wright, 1996;Sarfraz et al., 2007). The differences between findings of multiple researches may be due to the differences in quantity and quality of nutrients and secondary compounds in host plants, rearing techniques, conditions of experiments, hairiness and structure of leaf epidermis, as well as the source of T. urticae population.
The life table parameters, especially intrinsic rate of increase (r m ), have been used as indicators of pest population performance to assess the level of plant resistance to herbivorous pests (Sabelis, 1985;Sedaratian et al., 2010;Golizadeh et al., 2016). Several factors, such as development time, survivorship, and fecundity rate can affect the intrinsic rate of increase, so this parameter adequately summarizes the physiological qualities of an insect in relation to its capacity to increase. Hence, r m would be the most appropriate index to evaluate the performance of an insect on different host plants, as well as their susceptibility (Kocourek et al., 1994;Southwood and Henderson, 2000). The significant differences in our r m values showed extensive variation among rose cultivars in terms of suitability for spider mites. The estimated r m values of T. urticae on all rose cultivars (except for Roulette cultivar) in current research obviously fell in the range of r m value (0.212 to 0.480 day -1 ) reported in literature (Sabelis, 1985;Bounfour and Tanigoshi, 2001;Kafil et al., 2007;Razmjou et al., 2009;Sedaratian et al., 2010). In our study, r m value on Roulette (as the least suitable cultivar) was 0.169 day -1 . Susceptibility of rose cultivar for T. urticae is indicated by both net reproduction rate and mean generation time, which are summarized in intrinsic rate of increase (r m ). The relatively lower net reproductive rate on Roulette is a major factor affecting r m value on this cultivar. In addition, the longest mean generation time was calculated on this host plant (21.20±0.28 days), which can effectively result in a lower intrinsic rate of increase on Roulette cultivar. The relatively longer generation time on Pink Promise had caused r m value not to be the highest on this cultivar despite the highest fecundity and reproductive rate on this host plant. The variations in life table parameters were probably a function of different food sources (host plants) taken up by the adults during larval and nymphal stages. The lower performance of some host cultivars may be due to the absence of primary essential nutrients for the growth and development of this mite or the presence of secondary metabolites that directly affect potential herbivore development and fecundity (Potter and Anderson, 1982;Sabelis, 1985;Awmack and Leather, 2002;Pietrosiuk et al., 2003;Razmjou et al., 2009;Sedaratian et al., 2010;Khanamani et al., 2012).
In conclusion, three rose cultivars, i.e. Cool Water, Orange Juice, and Persian yellow, were the most susceptible host plants and the Roulette cultivar showed relatively positive characteristics com-pared with others and was a relatively resistant cultivar. Other cultivars showed intermediate susceptibility. Host plant resistance is in many ways an efficient pest management strategy and it has been one of the most important elements of integrated pest management programs (Zehnder et al., 2007). It is easy to use, economical to the producer, cumulative in effects, and mostly compatible with other strategies, especially biological control methods (Wiseman, 1994, Desneux et al., 2007. Our findings provide some basic information describing resistance of ten rose cultivars to spider mite. Thus, the use of partially resistant cultivars can enhance biological and chemical control methods. When this information is used in association with additional ecological characteristics (such as temperature, humidity and other environmental conditions), it could be useful in developing and planning spider mite integrated pest management programs on rose shrubs in urbanized areas, as well as in commercial rose breeding. Additional studies to determine the type and quantity of nutrients, as well as insect growth deterrent compounds of each cultivar are needed for a comprehensive discussion.