Occurrence of inter-species hybrids in natural populations might be evidence of gene flow between species. In the present study we found evidence of gene flow between two sympatric, genetically related scale insect species – the citrus mealybug Planococcus citri (Risso) and the vine mealybug Planococcus ficus (Signoret). These species can be distinguished by morphological, behavioral, and molecular traits. We employed the sex pheromones of the two respective species to study their different patterns of male attraction. We also used nuclear ITS2 (internal transcribed spacer 2) and mitochondrial COI (Cytochrome c oxidase sub unit 1) DNA sequences to characterize populations of the two species, in order to demonstrate the outcome of a possible gene flow between feral populations of the two species. Our results showed attraction to P. ficus pheromones of all tested populations of P. citri males but not vice versa. Furthermore, ITS2 sequences revealed the presence of ‘hybrid females’ among P. citri populations but not among those of P. ficus. ‘hybrid females’ from P. citri populations identified as P. citri females according to COI sequences. We offer two hypotheses for these results. 1) The occurrence of phenotypic and genotypic traits of P. ficus in P. citri populations may be attributed to both ancient and contemporary gene flow between their populations; and 2) we cannot rule out that an ancient sympatric speciation by which P. ficus emerged from P. citri might have led to the present situation of shared traits between these species. In light of these findings we also discuss the origin of the studied species and the importance of the pherotype phenomenon as a tool with which to study genetic relationships between congener scale insects.
Citation: Kol-Maimon H, Ghanim M, Franco JC, Mendel Z (2014) Evidence for Gene Flow between Two Sympatric Mealybug Species (Insecta; Coccoidea; Pseudococcidae). PLoS ONE 9(2): e88433. https://doi.org/10.1371/journal.pone.0088433
Editor: Norman Johnson, University of Massachusetts, United States of America
Received: July 8, 2013; Accepted: January 7, 2014; Published: February 11, 2014
Copyright: © 2014 Kol-Maimon et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: Murad Ghanim declares that he is a PLOS ONE editorial board member and that this does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Genetic isolation between species in areas of sympatry may occur via three types of sex barriers . The first and most common type is ‘ecological isolation’, which includes adaptation to different habitats and different seasonal phenotypic features, leading to isolation between potential mates in space and time. There is evidence that ecological isolations can be breached under laboratory conditions following interbreeding , , . The second type - ‘postzygotic isolation’ is created by copulation and fertilization with an outcome of non-fertile offspring . The third type - ‘behavioral isolation’ refers to cases where potential mates meet but do not copulate because courtship behavior patterns do not mesh , , .
Hybrid speciation is one form of sympatric speciation, defined as the occurrence of new species as an outcome of interbreeding between two or more species . Hybrid speciation begins with the creation of a hybrid zone, in which genetically distinct groups meet and produce individuals of mixed ancestry , , . Hybrid individuals are restricted to the hybrid zone as long as they display low fitness and survival incapability . When the habitat or the population gene pool changes, hybrid zone can be breached and hybrid speciation may begin , .
Two mealybug species, the citrus mealybug Planococcus citri (Risso) and the vine mealybug Planococcus ficus (Signoret) are key pests of a wide range of agricultural crops, and they share many host plants and habitats. Both mealybug species are known to transmit plant viruses and to secrete honeydew on which sooty mold develops, thereby causing severe economic losses worldwide , , , . The taxonomic status of these species has been disputed for almost two centuries. In 1857, Signoret taxonomically differentiated P. ficus from P. citri, in light of their development on different host-plant species. In 1915 both were 1 reclassified as one species, but in 1956 they were again reclassified as two species , , ; Cox  separated them according to adult female morphology, and later their separation was further confirmed by the striking differences between the molecular structures of the female sex pheromones , , . The P. citri pheromone consists of a single chemical component, i.e. (S+)-cis-(1R)-3-isopropenyl-2,2-imethylcyclobutanemethanol acetate , whereas P. ficus occurs in populations whose females release one pheromone compound, i.e. lavandulyl senecioate ( = LS) as well as other populations whose females release two pheromone compounds, i.e. LS and lavandulyl isovalerate ( = LI) , , . Planococcus ficus and P. citri also differ in other properties (see Table 1), for example, development rate , , . Recently molecular tools based on sequencing of the mitochondrial Cytochrome Oxidase I (COI) and the nuclear Internal Transcribed Spacer 2 (ITS2) genes , ,  have been used to differentiate P. citri from P. ficus. Sequencing the ITS2 segments of suspected hybrid specimens is frequently used to demonstrate potential hybridization and gene flow between species , , , . On the other hand, sequencing of maternally inherited COI segments is considered a better tool for phylogeny and taxonomy even though introgression of these segments from one species to another sometimes occurs , , , , .
The natural occurrence of inter-species hybrids in insects is well documented , , , , , . Among the scale insects (Coccoidea), laboratory-induced inter-species hybrids have been observed in the mealybug family (Pseudococcidae) , , , , .
However, evidence is yet lacking for natural gene flow between feral populations of sympatric mealybug species, or in other scale insect families. Hybrids of P. citri and P. ficus were documented in laboratory experiments by Tranfaglia and Tremblay , who found that the hybrid females displayed intermediate morphological features. By using crude solutions of pheromone effluvia of the two species, Rotundo and Tremblay  found that the laboratory hybrid males of each species displayed reciprocal attraction to the other congener pheromone in addition to that of conspecific females.
The present study aimed at revealing possible gene flow between feral populations of P. citri and P. ficus. We used sex pheromone male attraction and sequencing and GenBank references comparison of ITS2 and COI segments to show gene flow between those species. So far there is no evidence of field occurrence of natural hybrids of P. citri and P. ficus that could indicate occurrence of gene transfer between these species. However, in a controlled environment in the laboratory, these species are easily hybridized (20, Kol-Maimon et al., unpublished data). In many areas feral populations of these species share habitats, therefore cross-mating should be expected, although mealybug F1 hybrids may suffer marked mortality . Rotundo and Tremblay 1982  and Kol-Maimon et al. (unpublished data) documented high mortality among hybrid male crawlers of P. citri and P. ficus. Furthermore, no information about fertility of these F1 hybrids is available.
The objectives of this study were addressed by examining populations of P. citri and P. ficus from different locations and different host plants. Each population was characterized by its male pherotype behavior  and its genetic identity, e.g., COI, and by using molecular markers for detecting potential female hybrids that might account for pheromone cross-attraction of their male offspring. Furthermore, the ITS2 segments of suspected crossbred specimens were sequenced. This approach is frequently used to demonstrate potential hybridization and gene flow between 22 species, among both animals and plants , , , , .
Materials and Methods
Unless otherwise specified, the mealybug populations were reared on potato sprouts washed with 95% ethanol in darkness, at 25o29 C and 50% relative humidity. Gravid females were transferred to clean potato sprouts, , and placed on tissue paper in sealed plastic containers measuring 15×10 cm in diameter and height, respectively, and covered with thin, polyethylene sheets which allow ventilation but prevent movement of crawlers between samples. Every 3 days the seal was removed for 5–10 min for ventilation, and the tissue papers were replaced to prevent development 1 of excess humidity and mold.
Male pherotype characterization
The male pherotypes were characterized according to their specific responses to P. ficus and P. citri pheromones. Individual males were exposed to these compounds in no-choice tests, in which the compounds were presented in a random succession of three arenas . The pheromone solutions for the bioassay were prepared by dissolving the appropriate pheromone component in n-hexane (1 ng/µl). The males were exposed in 10–cm-diameter glass Petri dish arenas, each containing a 5-mm-diameter filter paper disk (double-layer Whatman No 1) impregnated with 6 ng of the tested pheromone, and two untreated paper disks that served as controls. Pherotype identity was assigned to every male as follows: F – attracted to one or both P. ficus pheromones , ; C – attracted only to P. citri pheromone; FC – attracted to pheromones of both species; N – no attraction to any of the tested pheromones. Among those labeled as F and FC, we distinguish between males that responded to lavandulyl senecioate (LS) and those that responded to lavandulyl isovalerate (LI) or both. Further pherotypes that were identified: S20 attracted only to LS; I- attracted only to LI; SI- attracted to both LI and LS; CS21 attracted to both P. citri pheromone and LS; CI- attracted to both P. citri pheromone and LI; CSI- attracted to P. citri pheromone, LS and LI.
Distribution of male pherotypes among the studied populations
Young and gravid mealybug females were collected in several different locations from different host plants (Table 2) and were placed on sprouted potatoes for further rearing. The male pherotype of the first laboratory-reared population was characterized as described above.
Females were individually homogenized in 100 µl of CTAB (cetyl-trimethyl-ammonium bromide) buffer (1 M Tris; 5 M NaCl; 0.5 M EDTA; CTAB 2%; β-mercaptoethanol). The extract was incubated at 65°C for 2 h, after which DNA was extracted with chloroform (100%), and precipitated overnight with 8% ammonium acetate and 60% isopropanol. After maximum speed centrifugation for 15 min, the DNA pellet was washed with 70% ethanol, air-dried and resuspended in double-distilled water.
Polymerase Chain Reaction (PCR) on COI and ITS2 gene segments
DNA at 100–500 ng/µl from each extraction was used for PCR. The PCR reaction (total volume of 50 µl) included: 42 µl of double-distilled water, 5 µl of Taq polymerase buffer, 0.6 µl of 20 pmole (0.24 pmole final concentration) primer 1 (COI-TL2-N-3014 or ITS2-M-R-454847), 0.6 µl of 20 pmole (0.24 pmole final concentration) primer 2 (COI- CJ-J-2183 or ITS2-M-F-454845), 0.4 µl of 25 mM dNTPs (0.2 mM final concentration), and 0.4 µl of 5 U/µl Taq DNA polymerase (0.08 U/µl final concentration). The cycling conditions were as follows: 95°C for 3 min, 35 cycles of 95°C for 1 min, 55°C for 1 min and 72°C for 1 min; and a final elongation cycle at 72°C for 10 min. The primers were: ITS2-M-F-454845: 5′ CTC GTG ACC AAA GAG TCC TG 3′. ITS2-M-R- 454847 5′ TGC TTA AGT TCA GCG GGT AG 3′. COI- CJ-J-2183: 5′ CAA CAT TTA TTT TGA TTT TTT 3′. COI- TL2-N-3014: 5′ TCC ATT GCA CTA ATC TGC CAT 3′. The primer sequences and PCR procedure were according to Malausa et al. 2010 . Total size of fragments amplified was ∼800 bp for COI and ITS2.
Cloning and sequencing of PCR products
The number of single nucleotide polymorphisms (SNPs) between P.citri and P.ficus on the ITS2 segment that was amplified in this study using BioEdit and ClustalW software is 58. ITS2 segment that was used to identify P.citri from P.ficus was ∼800 bp in length. 58 SNPs out of 800 bp is approximately 8%, which means 92% identity between those species of the ITS2 segment was used in this study. Comparison of ITS2 and COI sequences between different P. citri specimens and between different P. ficus specimens from GenBank data base show 98% identity and more within each species.
ITS2 sequences of P. ficus COI-verified females that showed more than 98% identity with the P. ficus GenBank reference , ,  were considered as type A; ITS2 sequences of P. citri COI-verified females that showed more than 98% identity with the P. citri GenBank reference  were considered as type B; and those with less than 92% identity to P. citri or P. ficus GenBank references were considered as type H (Fig. 1). H type specimens were not detected among COI-verified P. ficus females.
A- Individuals with more than 98% identity with P. citri GenBank references, considered as P. citri. B- Individuals with more than 98% identity with P. ficus GenBank references, considered as P. ficus. H- Individuals with less than 92% identity with P. citri and P. ficus GenBank references, confirmed as hybrids of the two species by cloning sequencing. Black arrows mark double-peak signals indicating the existence of heterozygosity in this region. ITS2 GenBank references: P. ficus: GU134677, JQ085574, HQ852471; P. citri: JF714195. COI GenBank references: P. ficus: JN120845, EU250573, DQ238220; P. citri: AB439517, AF483204.
To determine whether an ITS2 PCR product with double pick signals lengthwise (type H figure 1) was a hybrid DNA of the two species, the products of 7 out of 15 H type specimens were cloned into pGEM-T Easy vector (Promega, Madison, USA) according to the manufacturer's instructions. A total of 7 type H PCR products were cloned (according to table 3) and 5 clones were sequenced for each product. Following transformations and colony picking, individual clones were subjected to sequencing by Macrogen Inc., Seoul, Korea). For direct sequencing of PCR products, the products were run on 1% agarose gel and the expected band (size ∼800 bp) was excised from the gel and cleaned with the Qiaquick Gel Extraction Kit (Qiagen, Hilden, Germany). The cleaned DNA samples were sequenced by Macrogen, and the sequences were compared with the non-redundant (nr) nucleotide database in GenBank to verify the identity of the sequenced DNA.
Pearson contingency analysis was conducted for comparison of pherotype distribution between and within populations, by using JMP software version 7 (SAS Institute Inc., Cary, NC, USA). One-way MANOVA was conducted with the same software, for comparison among females with differing ITS2 sequence identities that were sampled from different populations.
Pherotype distribution among mealybug populations from different habitats
Male pherotypes and their distributions among the sampled populations from diverse locations and various host plants, arranged according to mealybug species are displayed in Fig. 2. All examined P. citri populations included male pherotypes that were attracted to either P. citri or P. ficus pheromones, or both. Males attracted solely to the sex pheromone of P. ficus were detected in 20 out of the 23 tested populations of P. citri. In all tested populations of each species considerable numbers of males were indifferent to all three tested pheromones, i.e., they were N pherotype. Among P. citri populations, an average of 29.8% of the males (range, 11–54%) were C pherotype, i.e., attracted to P. citri pheromone only; 39.4% of the males (range, 14–66%) were FC, i.e., attracted to both P. citri and P. ficus pheromones; 9.3% of the males (range, 0–13%) were F pherotype, i.e., attracted only to P. ficus pheromones; and the remaining 21.4% (range, 5–56%) were N pherotype, i.e., showing no attraction to either of the tested pheromones. All 11 examined P. ficus populations consisted of male pherotypes that were attracted to P. ficus pheromones (LS, LI or both). On average, 70.4% (range, 52–83%) of males of the P. ficus populations were F pherotype and the remaining 29.6% (range, 17–48%) were N pherotype. Pearson contingency analysis showed that pherotype distributions differed significantly between the P. citri populations (P<0.0001; χ2 = 310.7 between hosts; P<0.0001; χ2 = 291.8 between origins) and within P. ficus populations (P<0.001; χ2 = 10.8 between hosts; P<0.0001; χ2 = 49.7 between origins), but a significantly larger distance was found between species (P<0.0001; χ2 = 1793) than within species. Pearson contingency analysis further showed a larger variance or heterogeneity within the P. citri group than within the P. ficus group, probably because of higher pherotype variability among males of the former group.
The males were divided into four groups: (1) F- males – attracted to one or both P. ficus pheromones; (2) C- males – attracted only to P. citri pheromone; (3) FC- males – attracted to P. citri pheromone and also to one or both of P. ficus pheromones; (4) N- males – not attracted to any of the tested pheromones. Unless otherwise specified, the sampled hosts originated from Israel. The numbers preceding the host names are the serial numbers shown in Table 2; the numbers in parentheses are the numbers of tested males per tested population. The species of the populations were determined by COI sequencing and comparison with gene bank references (P. ficus: JN120845, EU250573, DQ238220; P. citri: AB439517, AF483204).
Among male P. citri populations from the East Mediterranean region, pherotypes that were attracted to P. citri pheromone and to LS (i.e., C, S and CS pherotypes) comprised 67 and 71% of those from Israel and Turkey, respectively; those that were attracted to LI pheromone, alone or in combination with P. citri pheromone, and to P. ficus pheromone (i.e., I, SI, CI and CSI) comprised 25 and 19% of those from Israel and Turkey, respectively; and those that were characterized as not attracted to any pheromone (i.e., N) comprised 8 and 10% of the tested males, from Israel and Turkey, respectively. Planococcus ficus males of the East Mediterranean populations consisted of pherotypes that were attracted to LS pheromone (S) – 33 and 19% from Israel and Turkey respectively; those that were attracted to LI, alone or in combination with LS (I, SI) – 41 and 33% from Israel and Turkey, respectively. The N males consisted of 26 and 48% of the sampled populations from Israel and Turkey respectively.
Planococcus citri males from the American, west- and mid-Mediterranean populations consisted of pherotypes that were attracted to P. citri pheromone and others that were attracted to LS pheromones (C, S, CS); they formed 92, 84, 77 and 67% of the male populations from California, Portugal, Spain and Sicily, respectively; and unnattracted males (N) males formed 8, 16, 23 and 33%, respectively, of the populations from these areas. Among Planococcus ficus males of the Californian, mid- and west-Mediterranean populations, pherotypes that were attracted to LS pheromone (S) formed 79, 70, 73 and 79% of the populations from California, Portugal, Spain and Sicily, respectively; and N males formed 21, 30, 27 and 21%, respectively, of these populations. No male attraction to LI pheromone was detected among Californian, west- and mid-Mediterranean populations of either P. citri or P. ficus, whereas males that were attracted to LI were found among east-Mediterranean populations of both P. citri and P. ficus.
For statistical analysis all males attracted to LI, alone or in combination with I, SI, CSI or CI, were considered as one pherotype group, designated as the I-group; males that were not attracted to LI (i.e., groups S, C, CS, N) were considered as a second pherotype group, designated the Non-I group. Pearson contingency analysis showed significant differences between P. citri and P. ficus (P<0.0001; χ2 = 57.7), in the distributions between the I-group and the Non-I group. A significant and larger distance between I pherotypes and Non-I pherotypes was observed among east-Mediterranean populations than among west- and mid-Mediterranean populations (P<0.0001; χ2 = 337.3). No significant differences were found within the group with LI attraction, in east-Mediterranean populations (P = 0.15; χ2 = 2.1). No attraction to LI was found among American, west- and mid-Mediterranean populations. The distribution of male attraction to LI was influenced by the species but more strongly by the area of origin.
Genetic identity of the tested mealybug populations
Out of 13 tested P. citri populations 11 contained females that showed less than 92% identity to P. citri and P. ficus GenBank references (type H in Fig. 1) but such females were not detected among P. ficus populations (Table 3). In addition, the nucleotide chromatograms of all H females showed double-peak signals lengthwise in each sequence (as indicated by arrows in Fig. 1), indicating that the sequencing reactions suggested the presence of two PCR products. Following cloning and sequencing of the pair of ITS2 sequences of some of the type H females, comparison with GenBank references , ,  revealed that one member of the ITS2 pair displayed ≥98% similarity to either P. citri or P. ficus, therefore the ITS2 sequences of H females are likely to represent hybridization between P. citri and P. ficus.
One-way MANOVA revealed a significant difference between P. citri and P. ficus populations in their ITS2 identities (types A, B, H as described in Fig. 1) (P<0.0001; F1 = 70.3439, P<0.0001; F1 = 90.2481, P<0.0001; F1 = 23.1465; for types A, B, and H respectively). These results reveal a clear relationship, in the studied mealybug species, between the type of species and the ITS2 identities of females. On the other hand, the one-way MANOVA suggested no significant differences of ITS2 identity between the studied populations from different origins (P = 0.7; F4 = 0.6, P = 0.08; F4 = 2.5, P = 0.3; F4 = 1.4; for types A, B, and H, respectively) or different host plants (P = 0.3; F14 = 1.4, P = 0.2; F14 = 1.8; for types A and H respectively). These findings imply that neither host plant nor sampling area factors were related to ITS2 identity of females in the tested populations.
The two mealybug species addressed in the present study – P. ficus and P. citri – belong to the ‘citri’ species group of the genus Planococcus  and they are closely genetically related , , . The sex pheromones of three among the 12 members of the ‘citri’ group were previously identified as those of the species addressed in the present study – ficus and citri – and that of Planococcus minor . The sex pheromone compounds of these three species are structurally different , , , , therefore, the cross attraction between P. ficus and P. citri was unexpected. Insect sex-pheromone signals are highly species-specific, as recorded in aphids , moths ,  and in scale insects . However, cross-attraction between congeners is a common, but not always clear, phenomenon; it may result from very similar or shared identical sex pheromone compounds , , ,  which does not apply to the presently studied mealybug species. Cross-attraction between the closely related ermine moth species Yponomeuta spp. was reported by Hendrikse 1988 ; and later Löfstedt et al. 1991  suggested that cross-attraction between allochronic species of ermine moths in the laboratory, and the formation of hybrids in the laboratory between species that are reproductively isolated by pheromone differences, is evidence for the role of pheromones as reproductive isolation mechanisms among these moth species. Little is known about the genetic mechanisms that underlie the evolution of new sex pheromones between related species and still less with regard to specific olfactory pheromone receptors. However, it is to be expected that the male preference for a sex pheromone should co-evolve in parallel with the female sex pheromones . Pre-mating barriers, such as lack of pheromone attraction, may be the most important factors determining reproductive isolation in many taxa , but occurrence of hybrids may be a reason for cross-attraction. In the European corn borer moth, Ostrinia nubilalis (Crambidae) there are two pheromone-related strains – Z and E – which produce and respond to different E/Z-11-tetradecenyl acetate ratios . In the laboratory E and Z borer strains can interbreed  but, in spite of the use of the same sex pheromone blend, cross-attraction between these moths was rare. However, hybrid males produced in the laboratory responded over a broad range of Z and E pheromone blends . Rotundo and Tremblay 1982  were the first to show that laboratory-produced hybrid males of P. ficus and P. citri displayed an attraction to their maternal sex pheromone. However, the small number of hybrid males produced in that study probably prevented the occurrence of males attracted to their paternal-line sex pheromone.
In the present study we found that certain proportions (23–82%) of male pherotypes produced by any of the tested P. citri populations were attracted to the P. ficus pheromone. However, no male pherotypes attracted to P. citri pheromone were detected in any of the tested populations of P. ficus. This phenotypic picture is meaningful because the tested populations were taken from different areas and different host plants. The genetic aspect of this phenomenon is indirectly supported by cloning and sequencing of ITS2 DNA fragments, which revealed the existence of hybrid females in most of the sampled P. citri populations. It is suggested that in three out of 13 tested P. citri populations, these hybrids were not detected because of the small sample size and a minority of them in those cases (males in those populations were attracted to P. ficus pheromones); similarly, hybrid females were not detected among the tested P. ficus populations. The ITS2 gene sequence is highly preserved within species, because of its significant function among all organisms, as a ribosomal gene . Therefore, the occurrence of ITS2 sequence polymorphism among populations. Because of its significant function among all organisms, the supported by cloning and sequencing, is an indication of gene flow, probably through hybridization between species , , .
One explanation for the occurrence of pherotypes of P. citri males that are attracted to the pheromone of P. ficus may be the occurrence of gene flow from P. ficus to P. citri. These species share the same habitate landscape in many areas. For example, in various Mediterranean-climatic areas, the widespread host plants of the presently studied mealybug species are citrus groves that support large population of P. citri, and that grow adjacent to vineyards, fig (Ficus carica) and pomegranate trees that harbor the vine mealybug. The observation of populations of P. citri and P. ficus hybrids on banana and plantain in Uganda, as suggested by Watson and Kubiriba 2005 , was based on female morphology. However, intermediate female morphology could be seen in populations growing under extreme temperatures . Thus, so far no conclusive evidence was found in the field for the occurrence of natural hybrids of P. citri and P. ficus, or of any other scale insect species, which could show contemporary gene flow. The possible existence of inter-species hybrids is also often ruled out because of low or absence of fitness, as expressed in sterility, host incompatibility, and other factors , . Production of hybrids in an artificial environment may be possible because the usual ecological isolation can be breached under laboratory conditions by interbreeding , .
Laboratory crosses between Pseudococcus spp. and Phenococcus spp. and P. citri produced no viable adult offspring . However, P. citri and P. ficus are easily hybridized in the laboratory, and the hybrids are fertile to some extent , therefore, the possible occurrence of natural populations including hybrids should be taken into account. It is interesting to note that laboratory-generated hybrids that resulted from crossing P. ficus females with P. citri males resulted in high mortality, as compared with lower mortality of the reciprocal crossing (Kol-Maimon et al., unpublished data). In general, interspecific hybrids are less fit than thoroughbreds because of the large number of changes in their genomes . Although there is no information on the occurrence of hybrid scale insects in the wild, we may assume that survival of hybrids on wild plants would be even less probable than their survival on potato sprouts in the laboratory. Offspring produced as a result of cross-mating between P. ficus females and P. citri males (unlike the reciprocal cross-mating) hardly ever survive in the laboratory, therefore, P. ficus males displaying characteristics of P. citri males are unlikely to complete their development on plants in the field. This directional gene flow, from P. ficus to P. citri, but not vice versa, is a meaningful topic for future research on reproduction and genetic systems in mealybugs.
The presence of similar traits and gene sequences among two different species is not necessarily evidence for contemporary gene flow between species , , . The question is whether the occurrence of a ‘hybrid P. citri’ population, as shown in our present study, is a modern event brought about by anthropogenic activity, or whether, in fact, P. ficus resulted from sympatric speciation. Both mealybug species almost certainly share a common ancestor . The phylogenic relationships among the 12 species of the ‘citri’ group  are not clear. In light of information on the ‘citri’ members accumulated by Cox 1989 , it seems that the origin of the group is in Africa. According to Rung et al. , P. minor is genetically closer to P. citri than is P. ficus. The distribution of P. minor was discussed by Rung et al. , who emphasized the fact that this species is absent from the African mainland; they suggested that the origin of P. minor is in the eastern Palaearctic, which suggests that the speciation of P. citri and P. minor may have happened in Asia. The area of origin of P. citri is speculative, and ideas about it were derived from the areas from which the four principle parasitoids of the mealybug were first collected. Leptomastix dactylopii was believed to be from South America . However, it was suggested that L. dactylopii spread to South America during the era of the slave trade, because it belongs to the fauna of the African Leptomastix spp. . The origin of another major parasitoid, Anagyrus pseudococci (a complex of two species or sub-species)  is the Mediterranean Basin . The latter area is also the suggested origin of Leptomastidea abnormis . Coccidoxenoides perminutus is believed to originate from southern China  or Australia . All these four parasitoid species are known from populations of P. ficus in the Mediterranean area; A. pseudococci is attracted to the female sex pheromone of P. ficus but not to that of P. citri , . This information suggests that P. citri, unlike P. ficus, was not originally associated with L. dactylopii or A. pseudococci. The area of origin of P. ficus is not discussed in the literature, although many authors claim the Mediterranean basin to be its area of origin. Miller et al. 2005  suggested that it belonged to the Palearctic region. The natural distribution of P. ficus was accepted to be the Mediterranean Basin, but it easily could be the larger area that includes East Africa. If the natural area of P. citri, similarly to that of P. minor, is the East Palearctic, then it is more likely that the occurrence of P. ficus characteristics among P. citri populations is a result of a modern process related to transfer of these mealybugs to these same areas on cultivated plant species.
However, the question of whether the separation between P. citri and P. ficus is a result of sympatric or allopatric speciation is still open; shared traits among sympatric and allopatric populations of the same species are evidence for ancient rather than contemporary gene flow . Other studies suggest that introgression could be a significant source of genetic variation in hybridizing species groups . Qualitative manifestations of introgression may appear to be most impressive when hybridizing species meet in zones of secondary contact, following a period of divergence and lineage sorting in allopatry , and this may apply to the two presently studied mealybug species. The widespread occurrence of male pherotypes attracted to the P. ficus pheromone lavandulyl senecioate, i.e., LS males, in all P. citri populations (Table 4) might suggest that development of this pherotypic male structure could be a modern event. The occurrence of P. citri males attracted to lavandulyl isovalerate, i.e., LI males, among the East Mediterranean population of P. citri (Table 4) suggests a recent gene flow from P. ficus to P. citri in this particular area. Lavandulyl isovalerate is a pheromone compound produced by P. ficus populations, and P. ficus males attracted to it were found only in East Mediterranean populations, as in the course of the present study as well as in previous research , , .
Phylogenetic study of the populations of P. ficus and P. citri may shed more light on the relationship between these populations (Fig. S1). The close genetic relationship between P. citri and P. minor led to the speculation that the existence of male pherotypes of P. citri that respond to the P. minor female sex pheromone cannot be ruled out. In both P. citri and P. ficus significant percentages of males did not respond to the pheromones produced by either of these species. The occurrence of such males was already addressed by Kol-Maimon et al. 2010  with regard to P. ficus, and was demonstrated in the present study for both species in all tested populations. The question of whether these male pherotypes might respond to the pheromones of other close congeners, such as P. minor or P. halli, is a legitimate. Finally, we suggest that the pherotype issue might serve as an important tool for revealing genetic associations and past interaction(s) between scale insect congeners.
Phylogenetic tree based on representative ITS2 sequences of the populations from table 3 in the manuscript. Host, country of origin (serial number from table 2 - GenBank accession number) Clone = After cloning sequence according to table 3, PC or PF = Planococcus citri or Planococcus ficus based on GenBank alignment. Branches colors: Red- PC, Blue- PF, Green- Clone.
The authors wish to thank many colleagues for their support by supplying tested mealybug populations or technical help, (in alphabetic order): Mary Lu Arpaia, Jose Eduardo Belda, Ilan Ben- Dov, Yair Ben- Dov, Kent Daane, Miktat Doğanlar, Rebecca Duncan, Miriam Eliyahu, Rutty Harpaz, Poonam Jasrotia, Anat Levi-Zada, Douglass R. Miller, Jocelyn G. Millar, Joseph G. Morse, Alex Protasov, Laura Ross, Pompeo Suma, Rakefet Sharon, Selma Ulgenturk, Serguei Triapitsyn.
Conceived and designed the experiments: HKM MG JCF ZM. Performed the experiments: HKM. Analyzed the data: HKM MG ZM. Contributed reagents/materials/analysis tools: HKM MG ZM. Wrote the paper: HKM MG JCF ZM.
- 1. Coyne JA, Orr HA (2004) Speciation. Sunderland, Massachusetts U.S.A.
- 2. Feder JL, Chilcote CA, Bush GL (1989) Are the apple maggot, Rhagoletis Pomonella, and blueberry maggot, Rhagoletis Mendax, distinct species- Implication for sympatric speciation. Entomol Exp Appl 51: 113–123.
- 3. Feder JL, Hunt TA, Bush L (1993) The effect of climate, host- plant phenology and host fidelity on the genetics apple and hawthorn infesting races of Rhagoletis Pomonella. Entomol Exp Appl 69: 117–139.
- 4. Dres M, Mallet J (2002) Host races in plant-feeding insects and their importance in sympatric speciation. Phil Trans R Soc Lond B 357: 471–492.
- 5. Presgraves DC (2002) Patterns of postzygotic isolation in Lepidoptera. Evolution 56: 1168–1183.
- 6. Wiernasz DC, Kingsolver JG (1992) Wing melanin pattern mediates species recognition in Pieris occidentakis. Anim Behav 43: 89–94.
- 7. Coyne JA, Crittenden AP, Mah K (1994) Genetics of a pheromonal difference contributing to reproductive isolation in Drosophila. Science 265: 1461–1464.
- 8. Roelofs WL, Liu WT, Hao GX, Jiao HM, Rooney AP, et al. (2002) Evolution of moth sex pheromones via ancestral genes. Proc Natl Acad Sci U S A 99: 13621–13626.
- 9. Mallet J (2007) Hybrid speciation. Nature Reviews 446: 279–283.
- 10. Mallet J, Barton NH (1989) Strong natural selection in a warning-color hybrid zone. Evolution 43: 421–431.
- 11. Jiggins CD, Mallet J (2000) Bimodal hybrid zones and speciation. Tr Ecol Evol 15: 250–255.
- 12. Nielsen EE, Hansen MM, Ruzzante DE, Meldrup D, Grønkjær P (2003) Evidence of a hybrid-zone in Atlantic cod (Gadus morhua) in the Baltic and the Danish Belt Sea revealed by individual admixture analysis. Mol Ecol 12: 1497–1508.
- 13. Seehausen O (2004) Hybridization and adaptive radiation. Tr Ecol Evol 19: 198–207.
- 14. Peccound J, Ollivier A, Plantegenest M, Simon JC (2009) A continuum of genetic divergence from sympatric host races to species in the pea aphid complex. PANS 106: 7495–7500.
- 15. Mcmillan OW, Jiggens CD, Mallet J (1997) What initiates speciation in passion-vine butterflies? Evolution 94: 8628–8633.
- 16. Sota T (2002) Radiation and reticulation: extensive introgressive hybridization in the carabid beetles Ohomopterus inferred from mitochondrial gene genealogy. Pop Ecol 44: 145–156.
- 17. Turner TL, Hahn MW, Nuzhdin SV (2005) Genomic Islands of Speciation in Anopheles gambiae. Plos Biol 3: 1572–1578.
- 18. Stevens JR, Wall R, Wells JD (2002) Paraphyly in Hawaiian hybrid blowfly populations and the evolutionary history of anthropophilic species. Ins Mol Biol 11: 141–148.
- 19. Nur U, Chandra SH (1963) Interspecific hybridization and gynogenesis in mealybugs. Am Natur 895: 197–202.
- 20. Rotundo G, Tremblay E (1982) Hybridization and sex pheromone response in two closely related mealybug species (Homoptera: Pseudococcidae). Syst Entomol 7: 475–478.
- 21. Tranfaglia A, Tremblay E (1982) A morphological comparison between Planococcus citri (Risso), Planococcus ficus (Signoret) and their F1 hybrids. Entomotaxonomia 4: 1–5.
- 22. Charles JG, Froud KJ, Henderson RC (2000) Morphological variation and mating compatibility within the mealybugs Pseudococcus calceolariae and P. similans (Hemiptera: Pseudococcidae), and a new synonymy. Syst Entomol 25: 285–294.
- 23. Millar JG, Daane KM, McElfresh JS, Moreira JA, Malakar-Kuenen R, et al. (2002) Development and optimization of methods for using sex pheromone for monitoring the mealybug Planococcus ficus (Homoptera: Pseudococcidae) in California vineyards. J Econ Entomol 95: 706–714.
- 24. Walton VM, Pringle KL (2004) Vine mealybug, Planococcus ficus (Signoret) (Hemiptera: Pseudococcidae), a key pest in South African vineyards. SA J Enol Viticult 25: 54–62.
- 25. Bier-Leonhardt BA, Moreno DS, Schwarz M, Fargerlund J, Plimmer JR (1981) Isolation, identification and synthesis of the sex-pheromone of the citrus mealybug, Planococcus citri (RISSO). Tetrah Lett 22: 389–392.
- 26. Franco CJ, Suma P, Zada A, Mendel Z (2004) Comparative biology of the citrus mealybug and the vineyard mealybug (Hemiptera: Pseudococcidae). X International Symposium on Scale Insect Studies – ISSIS. Adana, Turkey. 19–23 April. (in English).
- 27. Cox JM (1981) Identification of Planococcus citri (Homoptera: Pseudococcidae) and the description of new species. Syst Entomol 6: 47–53.
- 28. Cox JM (1983) An experimental study of morphological variation in mealybugs (Homoptera: Coccoidea: Pseudococcidae). Syst Entomol 4: 361–382.
- 29. De Lotto G (1975) Notes on the vine mealybug (Homoptera: Coccoidea: Pseudococcidae). J Ent Soc SA 38: 125–130.
- 30. Hinkens DM, McElfresh SJ, Millar JG (2001) Identification and synthesis of the sex pheromone of the vine mealybug Planococcus ficus. Tetrah Lett 42: 1619–1621.
- 31. Zada A, Dunkelblum E, Assael F, Harel M, Cojocaru M, et al. (2003) Sex pheromone of the vine mealybug, Planococcus ficus in Israel: Occurrence of a second component in mass-reared population. J Chem Ecol 29: 977–988.
- 32. Kol-Maimon H, Zada A, Franco JC, Dunkelblum E, Protasov A, et al. (2010) Male behaviors reveal multiple pherotypes within vine mealybug Planococcus ficus (Signoret) (Hemiptera; Pseudococcidae) populations. Naturwissenschaften 97: 1047–1057.
- 33. Gray HE (1954) The development of the citrus mealybug. J Econ Entomol 47: 174–176.
- 34. Gullan PJ, Kosztarab M (1997) Adaptations in scale insects. Ann Rev Entomol 42: 23–50.
- 35. Rung A, Scheffer SJ, Evans E, Miller D (2008) Molecular identification of two closely related species of mealybugs of the genus Planococcus (Homoptera: Pseudococcidae). Annu Entomol Soc Am 101: 525–532.
- 36. Demontis MA, Ortu S, Cocco A, Lentini A, Migheli Q (2007) Diagnostic markers for Planococcus ficus (Signoret) and Planococcus citri (Risso) by random amplification of polymorphic DNA-polymerase chain reaction and species-specific mitochondrial DNA primers. J App Entomol 131: 59–64.
- 37. Malausa T, Fenis A, Warot S, Germain JF, Ris N, et al. (2010) DNA markers to disentangle complexes of cryptic taxa in mealybugs (Hemiptera: Pseudococcidae). J App Entomol 135: 142–155.
- 38. Wakabayashi K, Komatsu M, Murakami M, Hori I, Takegami T (2008) Morphology and gene analysis of hybrids between two congeneric sea stars with different modes of development. Biol Bull 215: 89–97.
- 39. La Rosa G, Marucci G, Zarlenga DS, Casulli A, Zarnke RL, et al. (2003) Molecular identification of natural hybrids between Trichinella native and Trichinella T6 provides evidence of gene flow and ongoing genetic divergence. Int J Parasitol 33: 209–216.
- 40. Moody ML, Les D (2002) Evidence of hybridity in invasive watermilfoil (Myriophyllum) populations. Proc Natl Acad Sci U S A 99: 14867–14871.
- 41. Sota T (2002) Radiation and reticulation: extensive introgressive hybridization in the carabid beetles Ohomopterus inferred from mitochondrial gene genealogy. Popul Ecol 44: 145–156.
- 42. Abe TA, Spence JR, Sperling FAH (2005) Mitochondrial introgression is restricted relative to nuclear markers in a water strider (Hemiptera: Gerridae) hybrid zone. Can J Zool 83: 432–444.
- 43. Ahmed F, Yasuyuki K, Strüssmann CA, Yamasaki I, Yokota M, et al. (2008) Genetic characterization and gonad development of artificially produced interspecific hybrids of the abalones, Haliotis discus discus Reeve, Haliotis gigantea Gmelin and Haliotis madaka Habe. Agri Res 39: 532–541.
- 44. Rees DJ, Dioli M, Kirkendall LR (2003) Molecules and morphology: evidence for cryptic hybridization in African Hyalomma (Acari: Ixodidae). Mol Phylo Evol 27: 131–142.
- 45. Lovette IJ, Bermingham E, Rohwer S, Wood C (1999) Mitochondrial restriction fragment length polymorphism (RFLP) and sequence variation among closely related avian species and the genetic characterization of hybrid Dendroica warblers. Mol Ecol 8: 1431–1441.
- 46. Turner TL, Hahn MW, Nuzhdin SV (2005) Genomic islands of speciation in Anopheles gambiae. PLoS Biol 3: 1572–1578.
- 47. Stevens JR, Wall R, Wells JD (2002) Paraphyly in Hawaiian hybrid blowfly populations and the evolutionary history of anthropophilic species. Ins Mol Biol 11: 141–148.
- 48. Nur U, Chandra HS (1963) Interspecific hybridization and gynogenesis in mealybugs. Am Natur 97: 197–202.
- 49. Barkam TJ, Simpson BB (2002) Hybrid origin and parentage of Dendrochilum acuiferum (Orchidaceae) inferred in a phylogenetic context using nuclear and plastid DNA sequence data. Syst Bot 27: 209–220.
- 50. Abd-Rabou S, Shalaby H, Germain JF, Ris N, Kreiter P, et al. (2011) Identification of mealybug pest species (Hemiptera: Pseudococcidae) in Egypt and France using a DNA barcoding approach. Bull Entomol Res 102: 515–223.
- 51. Beltrà A, Soto A, Malausa T (2012) Molecular and morphological characterization of Pseudococcidae surveyed on crops and ornamental plants in Spain. Bull Entomol Res 102: 165–172.
- 52. Baumann L, Baumann P (2005) Cospeciation between the primary endosymbionts of mealybugs and their hosts. Curr Microbiol 50: 84–87.
- 53. Ho HY, Hung CC, Chuang TH, Wang WL (2007) Identification and synthesis of the sex pheromone of the passionvine mealybug, Planococcus minor (Maskell). J Chem Ecol 33: 1986–1996.
- 54. Thieme T, Dixon AFG (1996) Mate recognition in the Aphis fabae complex: daily rhythm of release and specificity of sex pheromones. Entomol Exp App 79: 85–89.
- 55. Tumlinson JH, Yonce CE, Doolittle RE, Heath RR, Gentry CR, et al. (1974) Sex pheromones and reproductive isolation of the lesser peachtree borer and the peachtree borer. Science 185: 614–616.
- 56. Phelan PL, Baker TC (1986) Cross-attraction of five species of stored-product Phycitinae (Lepidoptera: Pyralidae) in a wind tunnel. Env Entomol 15: 369–372.
- 57. Zilkowski BW, Bartelt RJ (1999) Cross-attraction of Carpophilus humeralis to pheromone components of other Carpophilus species. J Chem Ecol 25: 1759–1770.
- 58. Caceres A, Segura D, Vera MT, Wornoayporn V, Cladera JL, et al. (2009) Incipient speciation revealed in Anastrepha fraterculus (Diptera; Tephritidae) by studies on mating compatibility, sex pheromones, hybridization, and cytology. Biol J Linn Soc 97: 152–165.
- 59. Hendrikse A (1988) Hybridization and sex-pheromone responses among members of the Yponomeuta padellus - complex. Ent Exp App 48: 213–223.
- 60. Löfstedt C, Herrebout WM, Menken SBJ (1991) Sex pheromones and their potential role in the evolution of reproductive isolation in small ermine moths (Yponomeutidae). Chemoecology 2: 20–28.
- 61. Lassance JM, Löfstedt C (2009) Concerted evolution of male and female display traits in the European corn borer, Ostrinia nubilalis. BMC Biol 7: 7–10.
- 62. Coyne JA, Orr HA (1989) Patterns of speciation in Drosophila nubilalis. Ann Entomol Soc Am 68: 305–309. Evolution 43: 362–381.
- 63. Kochansky J, Carde RT, Liebherr J, Roelofs WL (1975) Sex pheromones of the European corn borer in New York. J Chem Ecol 1: 225–231.
- 64. Linn C, Poole K, Zhang A, Roelofs W (1999) Pheromone-blend discrimination by European corn borer moths with inter-race and inter-sex antennal transplants. J Comp Physio A 184: 273–278.
- 65. Glover TJ, Knodel JJ, Robbins PS, Eckenrode CJ, Roelofs WL (1991) Gene flow among three races of European corn borers (Lepidoptera: Pyralidae) in New York state. J Econ Entomol 20: 1356–1362.
- 66. Coleman AW (2003) ITS2 is a double-edged tool for eukaryote evolutionary comparisons. Tr Gen 19: 370–375.
- 67. Casteleyn G, Adams NG, Vanormelingen P, Debeer AE, Sabbe K, et al. (2009) Natural hybrids in the marine giatom Pseudo-nitzschia pungens (Bacillariophyceae). Genetic and morphological evidence. Protist 160: 343–354.
- 68. Takeshi A, Yohei A, Moritoshi I, Yumiko H, Umi C, et al. (2000) Molecular evidence of natural hybridization between Fasciola hepatica and F. gigantica. Parasitol Int 49: 231–238.
- 69. Watson GW, Kubiriba J (2005) Identification of mealybugs (Hemiptera: Pseudococcidae) on banana and plantain in Africa. Afr Entomol 13: 35–47.
- 70. Mallet J (2005) Hybridization as an invasion of the genome. Tr Ecol 20: 229–237.
- 71. Wu C, Palopoli MF (1994) Genetics of postmating reproductive isolation in animals. Ann Rev Genet 27: 283–308.
- 72. Machado CA, Kliman RM, Markert JA, Hey J (2002) Inferring the history of speciation using multilocus sequence data: the case of Drosophila pseudoobscura and its close relatives. Mol Biol Evol 19: 472–488.
- 73. Baker AM, Hurwood DA, Krogh M, Hughes JM (2004) Mitochondrial DNA signatures of restricted gene flow within divergent lineages of an atyid shrimp (Paratya australiensis). Heredity 93: 196–207.
- 74. Karl SA, Bowem BW, Avise JC (1992) Global population genetic structure and male-mediated gene flow in the green turtle (Chelonia mydas): RFLP analyses of anonymous nuclear loci. Gen Soc Am 131: 163–173.
- 75. Cox J (1989) The mealybug genus Planococcus (Homoptera: Pseudococcidae). Journal Bulletin of the British Museum (Natural History), Entomology 58: 1–78.
- 76. Compere H (1939) Mealybugs and their insect enemies in South America. Univ Calif Publs Entomol 7: 57–73.
- 77. Anga JM, Noyes JS (1999) A revision of the African and Malagasy species of the genus Leptomastix (Hymenoptera, Encyrtidae), parasitoids of mealybugs (Homoptera: Pseudococcidae). Bull Nat Hist Mus Lond (Entomol) 68: 93–128.
- 78. Triapitsyn SV, González D, Danel B, Vickerman DB, Noyes JS, et al. (2007) Morphological, biological, and molecular comparisons among the different geographical populations of Anagyrus pseudococci (Hymenoptera: Encyrtidae), parasitoids of Planococcus spp. (Hemiptera: Pseudococcidae), with notes on Anagyrus dactylopii. Biol Cont 41: 14–24.
- 79. Rivnay E (1960) Notes on parasites of Planococcus citri Risso in Israel. Ktavin 10: 223–224.
- 80. Viereck HL (1915) Notes on the life history of a species of wasp-like parasites of the genus Leptomastix, parasitic on the mealybug. Mon Bull Calif Hort Comm 4: 208–211.
- 81. Flanders SE (1951) Citrus mealybug. Four new parasites studied in biological control experiments. Calif Agric 5: 11.
- 82. Girault AA (1915) Australian Hymenoptera Chalcidoidea-VII. The family Encyrtidae with descriptions of new genera and species. Mem Queensland Mus 4: 1–184.
- 83. Franco JC, Borges da Silva E, Fortuna T, Cortegano E, Branco M, et al. (2011) Vine mealybug sex pheromone increases citrus mealybug parasitism by Anagyrus sp. near pseudococci (Girault). Biol Cont 58: 230–238.
- 84. Miller DR, Miller GL, Hodges GS, Davidson JA (2005) Introduced scale insects (Hemiptera: Coccoidea) of the United States and their impact on US agriculture. Proc Entomol Soc Wash 107: 123–158.
- 85. Kronforst MR, Young LG, Blume LM, Gilbert LE (2006) Multilocus analyses of admixture and introgression among hybridizing Heliconius butterflies. Evolution 60: 1254–1268.
- 86. Harper FM, Addison JA, Hart MW (2007) Introgression versus immigration in hybridizing high-dispersal echinoderms. Evolution 61: 2410–2418.
- 87. Mendel Z, Jasrptia P, Protasov A, Kol-Maimon H, Levi Zada A, et al. (2012) Responses of second-instar male nymphs of four mealybug species (Hemiptera: Pseudococcidae) to conspecific and heterospecific female sex pheromones. J Ins Behav 25: 504–513.
- 88. Ben-Dov Y (1994) A systematic catalogue of the mealybugs of the world (Insecta: Homptera: Coccoidea: Pseudoccocidae and Putoidae) with data on their geographical distribution, host plants, biology and economic importance. Andover: Intercept
- 89. Williams DJ (1985) Australian mealybugs. London: British Museum (Natural History) 431p.
- 90. Bodenheimer FS (1951) Citrus entomology in the Middle East. Groningenthe Netherlands: Hoitsema Brothers. 400 p.
- 91. Cavalieri V, Mazzeo G, Garzia GT, Buonocore E, Russo A (2008) Identification of Planococcus ficus and Planococcus citri (Hemiptera: Pseudococcidae) by PCR-RFLP of COI gene. Zootaxa 68: 65–68.
- 92. Cabaleiro C, Segura A (1997) Some characteristics of the transmission of grapevine leafroll associated virus 3 by Planococcus citri Risso. Eur J Pl Pathol 103: 373–378.
- 93. Meyer JB, Kasdorf GGP, Nel LH, Pietersen G (2008) Transmission of activated-episomal Banana streak OL (badna) virus (BSOLV) to cv. Williams Banana (Musa sp.) by three mealybug species. Pl Dis 92: 1158–1163.
- 94. Tsai CW, Chau J, Fernandez L, Bosco D, Daane KM, et al. (2008) Transmission of Grapevine leafroll-associated virus 3 by the vine mealybug (Planococcus ficus). Phytopathology 98: 1093–1098.
- 95. De Silva DPP, Jones P, Shaw MW (2002) Identification and transmission of Piper yellow mottle virus and Cucumber mosaic virus infecting black pepper (Piper nigrum) in Sri Lanka. Pl Path 51: 537–545.
- 96. Franco JC, Zada A, Mendel Z (2009) Novel approaches for the management of mealybug pests. Ishaaya I, Horowitz AR (eds.) Biorational control of arthropod pests: application and resistance managements. Netherlands: Springer. pp. 233–278.