Introduction

Eriophyoid mites are obligate plant feeders (Lindquist and Oldfield 1996). They are host-specific and mostly restricted to a narrow range of acceptable host plant species. The majority of eriophyoid species have been reported from a single host species or species within a single genus (Oldfield 1996). Studies on host specificity and plant–eriophyoid mite relationships are scarce and mostly concerned with species which can be used for biological control of weeds (Skoracka and Kucyznski 2006).

The genus Aceria includes over 900 species and is known to be taxonomically problematic, it may not be a natural assemblage of closely related species (Oldfield 1999; Amrine et al. 2003). Despite the description of so many species of Aceria, little has been published on relationships of intraspecific taxa (Oldfield 1999). Few papers have dealt with morphometric analysis of Aceria species inhabiting different hosts. The quantitative approach is available only for Aceria tosichella Keifer (Shevtchenko et al. 1970) and Aceria tenuis (Nalepa) (Sukhareva 1992), which live on monocots of the family Poaceae. Recently, 15 morphological features of 35 Aceria species living on Asteraceae were analyzed statistically to obtain information on the morphological variability of species (Sukhareva 2001).

The eriophyid mite Aceria anthocoptes (Nalepa) has been noted on Cirsium arvense (L.) Scop. (creeping, California, or Canada thistle) in several European countries (Davis et al. 1982). Discovered in Maryland in 1998, A. anthocoptes is now known to be present in 29 counties and seven states in the USA (Ochoa et al. 2001). A. anthocoptes is the only eriophyid mite that has been recorded on C. arvense (Petanović et al. 1997). Bull thistle, Cirsium vulgare (Savi) Tenore, has been mentioned as a host for A. anthocoptes; however no published data confirm this association (Ochoa et al. 2001). One immature A. anthocoptes found in spring on Cirsium discolor (Muhl. ex Willd.) Spreng from Maryland (USA) probably represents an ‘accidental’ occurrence (Ochoa et al. 2001). Recently, A. anthocoptes was found in Poland as accidental on Lolium perenne L. (Poaceae) (Skoracka 2004).

Cirsium. arvense is considered to be one of the most harmful weeds in the temperate world (Holm et al. 1977; Skinner et al. 2000). The species is highly variable; on the basis of hairiness and leaf shape, several infraspecific taxa, usually treated as varieties, are recognized. Four varieties—horridum, mite, argenteum, and incanum—are distributed in the flora of Central Europe (Hegi 1954). In Serbia, three varieties are found (Gajić 1975): vestitum (with hairy leaf undersurface and soft thistles), arvense (with both surfaces of leaves green and smooth, and weak thistles), and horridum (like arvense, but with hard thistles). The varieties vestitum and arvense both have two forms. Infraspecific taxa of C. arvense morphologically differ in the hairy cover of the leaf epidermis, leaf size, stem length, number of capitula, etc. They also exhibit different levels of resistance to herbicides (Zand et al. 2002).

Host specificity tests carried out with the A. anthocoptes population from the Belgrade area (Serbia) on five Cirsium, four Carduus, three Centaurea, two Carlina L., and one Onopordum L. species revealed that after 30 days the mite was found only on two varieties of C. arvense (Gassmann et al. 2004). It is widespread on C. arvense in Serbia (Rančić 2003). With its apparent host specificity and because of the damage it causes to its host plant, A. anthocoptes is being studied as a potential candidate for the biological control of C. arvense (Ochoa et al. 2001; Gassmann et al. 2004; Rančić et al. 2006).

Previous observations showed that A. anthocoptes lives and feeds between leaf hairs and that its population is more numerous on the infraspecific taxa of C. arvense with hairy abaxial sides of leaves. A hairy undersurface can provide a protective microhabitat and may have an effect on mite morphology (Petanović et al. 1997). On the other hand, based on the conclusions of Sukhareva (2001), it seems probable that different asteraceous hosts do not affect phenotypic variation of the mite.

The aim of this study was to describe phenotypic variation quantitatively among populations of the Canada thistle eriophyid mite A. anthocoptes originating from two varieties of host plants with hairy and smooth abaxial leaf surface, respectively and from two geographically separate areas of Serbia in order to test the hypothesis that different host varieties and geographical position of the locality do not have an impact on the phenotype of A. anthocoptes.

Materials and methods

The main criteria for the selection of samples were hairiness of leaves of the host plants (C. a. var. vestitum and arvense) and the presence of at least two varieties of the same locality. The samples of two varieties of C. a.—vestitum (hairy undersurface) and arvense (smooth undersurface)—were found and checked for mite presence at two localities in Serbia: Loc. Bežanijska Kosa (UTM—DQ56), 116.75 m above sea level (a.s.l.), collected on June 15, 2004: population I BK (from C. a. var. vestitum) and population II BK (from C. a. var. arvense), and Loc. Ivanjica (UTM—DP32), 468 m a.s.l., collected on September 25, 2003: population I IV (from C. a. var. vestitum) and population II IV (from C. a. var. arvense).

The climate of Belgrade can be described as moderate-continental. Belgrade is located at the altitude of 132 m (latitude—44°48″, longitude—20°28″). The average annual air temperature is 11.9°C. The average air temperature for the period April–October is 17.6°C. Average annual precipitation sums are 678 mm. The average precipitation sums for the period April–October are 429 mm. The average humidity for the period April–October is 65%.

The climate of Ivanjica can be described as moderate-continental. Ivanjica is an area with the altitude of 465 m (latitude—43°35″, longitude—20°14″). The average annual air temperature is 9.3°C. The average air temperature for the period April–October is 14.7°C. The average annual precipitation sums are 845 mm. The average precipitation sums for the period April–October are 554 m. Average humidity for the period April–October is 77%.

Mite specimens were collected by direct examination with a stereo microscope and/or extracted using the method described by De Lillo (2001). Twenty-four to 30 mites from each sample were mounted in a dorso-ventral position on slides in Kiefer’s F medium and identified (Amrine and Manson 1996). Protogyne females randomly selected from each population were examined with a phase-contrast microscope (LEICA DMLS). Deutogyne females and males were excluded from the analysis. Twenty-three traits were measured on each individual with the IM 1000 (Leica, Wetzlar, Germany) software package (Fig. 1). The morphometric traits selected are important and were commonly used for eriophyoid mite identification.

Fig. 1
figure 1

Measurements of A. anthocoptes female morphology used in statistical analysis. Explanation of abbreviations: A—length of body, B—length of prodorsal shield, C—width of prodorsal shield, D—length of scapular setae sc, E—scapular tubercles apart from sc, F—no. of dorsal annuli, G—no. of ventral annuli, H—length of lateral setae c2, I—length of I ventral setae d, J—length of II ventral setae e, K—length of III ventral setae f, L—length of genitalia, M—width of genitalia, N—length of setae 3a, O—spacing of tubercles 3a, P—spacing of tubercles 1b of coxa I, R—spacing of tubercles 1a of coxa I, S—spacing of tubercles 2a of coxa l, T—length of setae 2a, U—length of tibia I, W—length of tarsus I, Y—length of tibia II, Z—length of tarsus II

Full size image

All variables that entered the analyses presented normal distribution, as well as homogeneity of variance. The data were tested for normality using Kolmogorov–Smirnov and Shapiro–Wilk tests. When there are several dependent variables, there is a potential for inflated Type I errors (Zar 1999) due to multiple tests of (likely) correlated dependent variables (Tabachnik and Fidell 1996). A multivariate analysis of variance (MANOVA) allows for the comparison of the population means of all variables of interest at the same time (multivariate response), rather than considering multiple responses as a suite of univariate responses (Zar 1999). This reduces the magnitude of Type I errors. The statistical significance of the MANOVA can be determined in a variety of ways. The most often used statistic test Wilks’ Lambda was applied (Zar 1999). A two-way MANOVA was used to examine the effects of host-plant and locality on Canada thistle eriophyid mite A. anthocoptes morphological variation. The fixed factors were the variety of C. arvense (hairy versus smooth leaf undersurface) and two localities. When the MANOVA were statistically significant, subsequent univariate ANOVAs were performed to elucidate which responses contributed to the significant multivariate response. For this purpose Unequal N HSD post-hoc test was used. Additionally, to describe and interpret effects from MANOVA, a multivariate discriminant analysis (DA) was used as a useful post method to employ following a MANOVA. Discriminant analysis was employed on all data, in order to determine the relative importance of characters as discriminators between a priori groups and the relative positions of the centroids of those groups (Manly 1986). In addition, canonical variables were computed. All statistical analyses were conducted using the Statistica 6 software package (StatSoft 2001).

Results

Descriptive statistics of the quantitative traits of A. anthocoptes are given in Table 1. The two-way MANOVA of A. anthocoptes from two varieties of C. arvense at Bežanijska Kosa (I BK and II BK) and Ivanjica (I IV and II IV) revealed significant effects of host-plant, locality and host-plant × locality interaction on A. anthocoptes morphological variation (P < 0.001 in all cases, Table 2).

Table 1 Basic statistical data for 23 morphological traits of nine Aceria anthocoptes populations from two varieties of Cirsium arvense at Bežanijska Kosa (I and II BK) and Ivanjica (I and II IV). n = number of specimens, SD = standard deviation
Full size table
Table 2 Results of two-way MANOVA on the considered traits of Aceria anthocoptes from two varieties of Cirsium arvense (hairy and smooth undersurface) (Host-plant as factor) at Bežanijska Kosa and Ivanjica (Locality as factor)
Full size table

Some of the most significant results from Unequal N HSD post-hoc test are shown in Table 3.

Table 3 Results of Unequal N HSD post-hoc test for the most significant traits (C, F, U, V) of Aceria anthocoptes from 2 × 2 host plants and localities: 1—I BK; 2—I IV; 3—II BK; 4—II IV
Full size table

The result of the discriminant analysis showed the most important and distinct discrimination to be between the populations from Ivanjica (I IV and II IV) and Bežanijska Kosa (I BK and II BK) based on the first canonical axis (function) (Fig. 2). The total correct percent of Classification matrix of all four groups was high (89.189%).

Fig. 2
figure 2

Plots of scores of the first two canonical axes (Root 1 and Root 2) of four populations of Aceria anthocoptes from two host-plants and two geographically distant areas in Serbia. Abbreviations of populations are described in “Materials and methods” Section

Full size image

From the standardized canonical discriminant function coefficients (Table 4) it is evident that the first canonical function describes 93.34% of the total variability; the first and second together, 98.63%; and all three roots with 100% of the total variability. It should be stressed that the first and the second canonical functions describe most of the variability. The number of dorsal annuli (F) contributes most to this discrimination. The following two characters contribute, but to a lesser extent: E—the distance between tubercles (sc) and U—length of tibia of the first legs based on the first canonical function (Table 2). Width of prodorsal shield (C), length of the tibia of the second leg (V), and length of the third ventral (f) setae (K), has the most distinct discriminative power based on the second canonical function. Bearing in mind that the second canonical function describes only 5.29% of the total variability, it could be inferred that its discriminative power is significantly lower in comparison with the first canonical function.

Table 4 Standardized coefficients for canonical variables on first (Root 1) and second (Root 2) canonical axes in discriminant function analysis (DA)
Full size table

Aceria anthocoptes mites from two host plants (C. a. var. vestitum and arvense) and the same geographical location (Ivanjica, Southwest Serbia) are characterized by the smallest number of dorsal annuli, the smallest distance between scapular setae and the longest first leg tibia. The results obtained in this work indicate that the A. anthocoptes populations from Ivanjica differ from those originating from Bežanijska Kosa (Belgrade region, Central Serbia).

A host-plant-dependent morphological difference exists among Bežanijska Kosa populations, but does not in case of the Ivanjica populations (Fig. 2). The Bežanijska Kosa population of A. anthocoptes from hairy undersurface host-plant is characterized by narrower prodorsal shield, longer second leg tibia, and shorter length of the third ventral (f) setae, compared with the population inhabiting smooth undersurface host-plant on the same locality.

Discussion

Results of our study done at two widely separated localities revealed non-significant differences of certain morphological traits between A. anthocoptes populations inhabiting two different varieties of the host-plant. However, mites from two different varieties of the host-plant at Bežanijska Kosa had significant differences in width of prodorsal shield, length of the tibia of the second leg, and length of the third ventral (f) setae (K). Differences in these traits can be explained to some extent as adaptation sensu lato to protected microhabitat.

According to Kozlowski (1998) and Skoracka et al. (2002), free-living eriophyoids inhabiting protected microhabitats (narrow and deep grooves or a more hirsute surface) are elongated in body shape and have short setae and legs as a result of adaptation to such conditions. Absence of significant differences between A. anthocoptes inhabiting two varieties of the host plant at Ivanjica locality is possible to ascribe to higher average relative humidity which can neutralize the more exposed microhabitat such as variety of the host plant with smooth undersurface of leaves.

It has been reported that morphology of eriophyoid mites may be closely related to the structure of their habitat. Boczek et al. (1984) found certain differences of many morphometric features in females of Aculus fockeui (Nalepa & Trouessart) collected from various species of Prunus. According to Kozlowski (1998), protogyne females of Aculus schlechtendali (Nalepa) living on several varieties of apple trees display differences in body size and length of the ventral setae. Skoracka et al. (2002) found host-dependent morphological variation in three populations of the cereal rust mite Abacarus histrix (Nalepa). The authors interpreted the morphological variation as being on adaptation to specific environmental conditions created by the host.

On the other hand, Lamb (1953) did not observe differences between measurements of Aculops lycopersici (Massee) from Lycopersicum esculentum Miller and Physalis peruviana L. (Solanaceae). Morphometric analysis of 40 traits in four widely separated Aculops euphorbiae (Petanović) populations inhabiting Euphorbia seguierana Neck. subsp. seguierana and E. seguierana Neck. subsp. niciciana (Borbás ex Novák) Rech.f. (Euphorbiaceae) showed variation in measurable characteristics. The four populations exhibited a significant difference in only one trait, width of the dorsal shield. No impact of the different host plant subspecies was proven (Petanović and Dimitrijević 1995).

So far, one possible explanation of the results on morphological variation in different populations of A. anthocoptes obtained in the present study agrees with the hypothesis of Sukhareva (2001). In her paper dealing with mites of the genus Aceria on asteraceous plants, it was inferred that no distinct groups of species connected with phylogenetic groups of hosts can be singled out. The Aceria group of species from plants of the tribe Carduae is homogeneous with respect to quantitative features. This may be explained in the high degree of variability of the host plants themselves and in rapid evolution of this group of plants, accompanied by development of new forms. According to this author, one of the special features of evolution of the Asteraceae is its biochemical direction, which is associated with the synthesis of sesquiterpenes. It results in the development of physiologically specific new forms of plants. Among species of the genus Aceria inhabiting phylogenetically remote hosts that are sometimes different ecologically, there are similar, almost identical species of mites.

The results of our study support the hypothesis that varieties of plants within the family Asteraceae do not affect the phenotype of mite species.

At the other hand the most important and distinct discrimination was shown between the two distant geographical populations from Ivanjica and Bežanijska Kosa. The number of dorsal annuli (F) contributes most to this discrimination. Considering that samples of mites from two localities were taken in June and September respectively, it is also possible to ascribe these differences to season influences of mite morphology. The obtained results to be confirmed needed in further experimental work. In studying morphologically similar species (or subspecific taxa) of mites, we should consider their biochemical and genomic features.