Forty three population samples of mealy aphids from five European countries were collected from various winter and summer host plants (Table 1). The entire data set has been subdivided: 21 samples (bolded in Table 1) were used for canonical discrimination procedures and subsequent evaluation of the received discrimination functions was performed on remaining 22 samples.

For molecular analysis, a single aphid individual from one sampled plant was considered as a unique sample. Total genomic DNA was extracted from a single aphid using the DNeasy Blood & Tissue kit (Qiagen), which involved at least a 2 h digestion of tissue with proteinase K. Partial sequences of mitochondrial COIwere PCR-amplified using previously published primers (Turčinavičienė et al. 2006). PCR amplification was carried out in a thermal cycler (Eppendorf) in 50 µl volumes containing 1–2 µl genomic DNA, 5 µl of each primer (10 µM), 5 µl of PCR-reaction buffer, 5 µl of dNTP mix (2mM each), 4–8 µl of 25mM MgCl₂ and 1.25 U of AmpliTaq Gold 360 polymerase (5U/µl) and ddH₂O to 50 µl. The cycling parameters were as follows: denaturizing at 95°C for 10 min (1 cycle), denaturizing at 95°C for 30”, annealing at 49°C for 30” and extension at 72°C for 30” (32–37 cycles in total), and a final extension for 5 min (1 cycle). PCR products were subjected to electrophoresis on 2% TopVision agarose (Fermentas, Lithuania), stained with ethidium bromide and sized against a MassRuler Low Range DNA ladder (Fermentas, Lithuania) under UV light. PCR products were purified and sequenced at Macrogen Europe (Amsterdam, the Netherlands). The amplification primers were also used as sequencing primers. DNA sequences for each specimen were confirmed with both sense and anti-sense strands and aligned in the BioEdit Sequence Alignment Editor (Hall 1999). Partial sequences of COI gene were tested for stop codons and none were found. The sequence data have been submitted to the GenBank, Accession numbers JX943517-JX943559.

Forty three sequences of three Hyalopterus species were analyzed. Sequences of Aphis gossypii Glover, 1877 (Aphidini) and Nasonovia ribisnigri (Mosley, 1841) (Macrosiphini) were selected as outgroups for the phylogenetic analyses, which included Neighbor joining (NJ), Maximum parsimony (MP), Maximum likelihood (ML) and Bayesian inference in phylogeny (BI). NJ, MP and ML analyses were performed using MEGA 5 (Tamura et al. 2011). For NJ analysis Kimura 2-parameter (K2P) model of base substitution was used. Bootstrap values for NJ, MP and ML trees were generated from 1000 replicates. For ML analysis Tamura 3-parameter model with Gamma distribution (T92+G) was selected by MEGA 5 model selection option (Tamura et al. 2011). Bayesian analysis was conducted in MrBayes 3.2.1 (Ronquist and Huelsenbeck 2003) using General Time Reversible model with Gamma distribution (GTR+G), which was selected by jModeltest (Posada 2008). Four simultaneous chains, 3 heated and 1 “cold”, were run for 3 000 000 generations with tree sampling every 1000 generations. The topologies obtained by NJ, MP, ML and BI were similar, so only ML tree is shown with values of NJ/MP and ML/BI bootstrap support and posterior probabilities indicated above and below branches respectively.

Samples representing different clades in the molecular trees were used for canonical discrimination analysis: 2 samples from almond (Hyalopterus amygdali clade), 10 samples from cultivated plums (Hyalopterus pruni clade), and 9 samples from peaches (Hyalopterus persikonus clade) (Table 1).

Based on the earlier references (Poulios et al. 2007, Lozier et al. 2008), twenty two metric (in mm) characters were studied:

A2L – length of antennal segment 2; A2W – width of antennal segment 2; A3BW – basal width of antennal segment 3; A3L – length of antennal segment 3; A3SL – length of the longest hair on antennal segment 3; A4L – length of antennal segment 4; A5L – length of antennal segment 5; A6BL – length of basal part of antennal segment 6; A6TPL – length of terminal process of antennal segment 6; AT8SL – length of submedian hair on abdominal tergite 8; BL – body length (excluding cauda); CL – length of cauda; DT3L – length of the second segment of hind tarsus; F3L – length of hind femur; FSL – length of the frons hair; HW – width of the head across eyes; MDHSL – length of median dorsal head hair; MDHSW – distance between the bases of median dorsal head hairs. SL – length of siphunculus; T3L – length of hind tibia; URL – length of ultimate rostral segment; URW – basal width of ultimate rostral segment.

Measurements of the slide-mounted apterous viviparous females were performed by means of interactive measurement system Micro-Image (Olympus Optical Co. GmbH). STATISTICA 8 version software (Statsoft 2007) was exploited for data analysis. Pearson’s correlation coefficients were calculated to evaluate the correlation of morphometric characters with body length. Characters with strong (| r | ≥ 0.50) statistically significant (p<0.05) correlation with body length were removed from the further analysis: BL (r=1.00), F3L (r=0.58), T3L (r=0.59), A2L (r=0.57), HW (r=0.51). Remaining seventeen characters were used for forward stepwise discriminant analysis with host plant species as grouping variable followed by canonical analysis. Discriminant analysis was conducted in three steps. The first step was performed to discriminate between the all three mealy aphid species emerged in the COI dendrogram (Hyalopterus amygdali, Hyalopterus persikonus and Hyalopterus pruni). The second step was carried out to discriminate between Hyalopterus persikonus and non- Hyalopterus persikonus (Hyalopterus amygdali and Hyalopterus pruni) samples. The third step of the discriminant analysis was performed on Hyalopterus amygdali - Hyalopterus pruni data set (Hyalopterus persikonus samples excluded) to separate almond and plum mealy aphid species. Canonical scores were visualized as scatter plots. The morphological interrelationships among different samples were examined using hierarchical cluster analysis based on squared Mahalanobis distances (linkage method – UPGA).

Characters that contributed most in canonical discrimination functions were evaluated as having potential for species separation. The eventual species identification key based on these morphological characters and host plant information was constructed. Afterwards, it was applied on mealy aphid samples that were not used for the construction of the identification key (Table 1).

Lozier et al. (2008) reported partial COI sequences being the most variable in Hyalopterus aphids and suggested them as a possible tool for the identification of the mealy aphid species complex. Forty three partial COI sequences of 3 Hyalopterus species from 5 countries were included in analysis. The alignment contained 564 bases in final set with 79 variable sites, 35 of which appeared parsimony informative. The sequences were heavily biased towards A and T nucleotides. The average base composition was A = 34.3 %, C = 14.1 %, G = 12.0 % and T = 39.7 %. The overall transition/transversion ratio R = 2.805 for all sites.

The maximum parsimony (MP) analysis of partial COI sequences resulted in 425 equally parsimonious trees (length = 152, CI=0.76, RI=0.95). ML tree (T92+G model) showed similar topology, the same as NJ analysis (Kimura 2-parameter distances) and BI (GTR+G model) analyses. NJ, MP and ML bootstrap values over 50 % together with BI posterior probabilities over 0.50 are given at respective nodes of the same tree in Fig. 1. One can ensure that used Hyalopterus samples emerge as monophyletic relative to outgroup and form three major clades representing three host specific mealy aphid species. Hyalopterus pruni and Hyalopterus persikonus are placed as a sister species, whilst Hyalopterus amygdali is located basally.

The scatter plot of the first two canonical variates for samples from 18 different geographical localities representing three mealy aphid species (apterous viviparous females) is shown in Fig. 2. All individuals were reclassified correctly into their a priori specified groups. The following characters proved to be important predictors when separating between three Hyalopterus species: MDHSL, URW, T3L/CL (Table 2). The post hoc classification of samples gave 96.7 % correct identification of Hyalopterus persikonus (n=46), 100 % of Hyalopterus amygdali (n=10) and 99% of Hyalopterus pruni (n=94) specimens.

To discriminate between apterous viviparous females of Hyalopterus persikonus and non- Hyalopterus persikonus (Hyalopterus amygdali and Hyalopterus pruni) samples the following canonical function (for character acronyms see above) was obtained: 74.6150*URW-1.2696*T3L/CL+1. The values of canonical scores were >0 for Hyalopterus persikonus and <0 for Hyalopterus amygdali + Hyalopterus pruni. This combination of canonical variables separated 100 % of Hyalopterus persikonus (n=71) specimens involved in the analysis with a priori specified group membership. The post hoc classification gave 94.4 % correct identification of Hyalopterus persikonus (n=46) specimens.

To discriminate between apterous viviparous females of Hyalopterus amygdali and Hyalopterus pruni samples the following canonical function (for character acronyms see above) was obtained: -2.2645*SL-18.6609*MDHSL+1. The values of canonical scores were >0 for Hyalopterus amygdali and <0 for Hyalopterus pruni. This combination of canonical variables separated 94.5 % of Hyalopterus amygdali (n=18) and 100% of Hyalopterus pruni (n=67) specimens involved in the analysis with a priori specified group membership. The post hoc classification gave 100 % correct identification of Hyalopterus amygdali (n=10) and 94.7% of Hyalopterus pruni (n=94) specimens.

Out of eleven morphological characters included in the canonical function discriminating between sampled apterous viviparous females of mealy aphidspecies complex, the length of median dorsal head hair (MDHSL) enabled separation of 97.3 % Hyalopterus amygdali specimens. Namely, the lengths of median dorsal head hair from 0.026 to 0.039 mm were characteristic of Hyalopterus amygdali, whilst 0.036 – 0.067 mm – for other two species. Yet we failed to find any single character or ratio enabling satisfactory discrimination between apterous viviparous females of Hyalopterus pruni and Hyalopterus persikonus. For the present, the following morphological identification key might be suggested to identify apterous viviparous females of the mealy aphid species complex.