Leaf miners are a special insect group whose larvae feed and live on leaf tissues lying between the upper and lower epidermis, with the epidermis or at least the outer wall remaining undamaged (Hering, 1951; Kang, 1996; Dai et al., 2011b ). Most leaf-mining insects are monophagous or oligophagous. Their distinct species-specific feeding marks (that is, leaf mines) usually persist for a long time. Therefore, an ecologist or a paleontologist may easily reconstruct the life histories of leaf miners, identify their taxonomic group, evaluate the influences of leaf miners on host plants, and clarify the relationship between leaf miners and plants (Hespenheide, 1991; Labandeira et al., 1994; Lopez-Vaamonde et al., 2006; Dai et al., 2011b ). Thus, leaf-mining insects can be used as model organisms to study plant–insect–environment interactions (Hirowatari, 2009; Kang et al., 2009). Systematic research on leaf-mining insects has been performed in Europe, America and Australia (Needham et al., 1928; Hering, 1951, 1957; Spencer, 1990; Hespenheide, 1991). For example, the UK, the Netherlands and Australia have constructed websites on leaf miners and their host plants.

Compared with ectophagous insects, endophagous leaf miners are small; each individual consumes only a small quantity of leaves during the whole life history, generally not causing a large loss on plant biomass. They are more difficult to control due to their hiding habits; more importantly, leaf-mining might not damage the integrity of leaf surface so the impact of leaf-mining on plant physiology should be lower than expected (Pincebourde et al., 2006). However, a few mining species are considered as economic pests of agricultural crops and garden plants, e.g. coffee leaf miner (Leucoptera coffeella), apple blotch leaf miner (Phyllonorycter crataegella), Citrus leaf miner (Phyllocnistis citrella) and serpentine leaf miners (Liriomyza spp.). Their serious outbreaks may cause considerable damage. Leaf-mining insects have several direct effects on host plants, such as leaf morphology, leaf chemistry, plant physiology, plant growth and production. Plant responses to leaf-mining may vary depending on the leaf miner species and the plant species. Changing plant characteristics would have either negative or even positive impacts on the behavior, growth, or survival of leaf-mining herbivores. In comparison with ectophagous insects, there are fewer studies on the influence of leaf miners on host plants, especially in China. Therefore, it is necessary to summarize the effects of leaf-mining insects on their host plants, which may provide a basis for further studies on leaf miner-plant interactions and sustainable ecological control of leaf-mining pests.

There are about 10,000 described species of leaf miners in more than 50 families and four orders: moths (Lepidoptera), flies (Diptera), beetles (Coleoptera) and sawflies (Hymenoptera) (Csóka, 2003). However, many more leaf-mining species remain undiscovered (Csóka, 2003; Hirowatari, 2009). For example, most nepticulid moths are leaf miners, with 804 described species but the estimated species number may be about 2500 (E. J. van Nieukerken, pers. comm.). After a long-period of coevolution and interspecific competition, there are diverse leaf-mining patterns (Fig. 1), even on the same plant species (Dai et al., 2013, 2014). Leaf-mine diversity shows niche differentiation of leaf-mining insects on host plant leaves, and the trade-off between seeking favorable microhabitats and nutrients and avoiding leaf defensive structures and competition.

The diversity of leaf-mining patterns may refer to different leaf mine shapes and different leaf-mining depths (Table 1; Hering, 1951; Dai et al., 2013). The types of leaf mine shapes could be classified as linear mines, blotch mines, and linear-blotch mines (Fig. 1; Hering, 1951). Linear mines occur when the leaf-mining larvae keep feeding forward and have subtypes such as serpentine, spiral and branched (Kato, 1984). Blotch mines made by leaf miners have several feeding directions, with subtypes of circular, oval, rectangular, tentiform, ameba shape, etc. Among linear and blotch leaf mines, there are a series of transitional types. For example, linear mines could suddenly become much wider, or blotch mines may extend a narrow mine. Both are called “linear-blotch mines” (Hering, 1951; Kang, 1996; Csóka, 2003), which may be due to a change on feeding habitats in different larval stages (Hering, 1951). The vertical feeding sites of leaf miners may be epidermal cells, palisade tissues, spongy tissues or whole leaf tissues, which are called epidermal mines, upper surface mines, lower surface mines, and full depth mines, respectively (Hering, 1951; Kang, 1996). One special case is interparenchymal mines when leaf miners forage among the lower layer of palisade tissues and the upper layer of spongy tissues (Hering, 1951). Depending on the different leaf-mining pattern, we can divide leaf-mining insects into different functional groups, and study the ecological effects of different functional groups on host plants.

Table 1.

The feeding or oviposition of leaf miners often contributes to deformities and cracks in plant leaves (Hering, 1951; Kang, 1996). Due to the presence of insect larvae inside leaves, the leaves become asymmetrical, holes appear or parts are missing from the mined leaves (Hering, 1951). The tunnel of some leaf-mining weevils ends as a circular blotch from which an exit is cut leaving a round hole. Some infested leaves are crumpled because the cuticular layer is damaged (Raimondo et al., 2013). As for leaf asymmetry, the injured side of a birch leaf damaged by Phylloporia bistrigella is often smaller than the other (Hering, 1951); the width of each half and between side-ribs is variable in the leaves of elm (Ulmus glabra Huds.) mined by a leaf-mining weevil (Rhynchaenus rufus) (Møller, 1995). The abundance of leaf mines is associated with increasing leaf asymmetry in stone oak (Quercus rotundifolia Lam.) (Møller & Lope, 1998). In unpolluted forests, the mined leaves are smaller and more asymmetric than the intact ones (Kozlov, 2005). There are a couple of factors responsible for the above growth abnormalities. If a new leaf has been oviposited in, young leaf cells are damaged and striking deformities occur during later growth (Hering, 1951). A disproportionate distribution in nutrients between damaged and normal areas of a leaf may lead to leaf asymmetry. The level of leaf asymmetry was higher when elm was provided with fertilizer. Besides, there is a positive relationship between leaf asymmetry and leaf size. That is, large-sized leaves infested by a leaf miner result in greater asymmetry than infested small ones (Møller, 1995). Oak trees with higher levels of asymmetry are more likely to die than those with lower levels (Møller & Lope, 1998). However, some researchers have argued that leaf asymmetry is induced by other environmental factors, which also result in higher leaf miner numbers (Møller, 1995). Not all miners have impacts on leaf external morphology. For example, Liriomyza strigata mines the main vein of leaves, and has minor effects on leaf distortion (Kang, 1996). In the other way, leaf anatomy such as vein patterns can also dictate the shape of leaf mines (Scheirs et al., 1997).

Leaf-mining may alter host plant chemical composition especially for secondary substances. Trees with leaf miners have less root carbohydrate and twig starch than healthy ones (Percival et al., 2011). The midrib damaged by leaf-mining moths (Lepidoptera: Eriocranniidae) has slighty higher nitrogen concentration in tunnels than those in which the midrib was intact (Johnson et al., 2002). The injured midrib would restrict water supply to the xylem, speed up protein degradation, and increase available nitrogen compounds (White, 1984). On the contrary, the alfalfa blotch leaf miner decreased the yield of protein and digestible dry matter in the field (Byers & Valley, 1981). Leaf nutritional quality changed by leaf miners may indirectly affect other insects (Johnson et al., 2002). Plant secondary chemicals induced by leaf-mining larvae includes alkaloids, phenolic, benzaldehydes, methyl salicylates, benzyl alcohols, jasmonic acid, salicylic acid and so on (Mopper et al., 2004; Johne et al., 2006; Magalhães et al., 2010). The production of secondary metabolites may have either positive or negative impacts on the oviposition and feeding of conspecific or heterospecific herbivores (Johne et al., 2006; Magalhães et al., 2010). Mined leaves can induce chemical, physical and even visual defensive signals to herbivores (Yamazaki, 2010). Leaf-mining impacts are not the same for different leaf-mining behaviors. For example, apple leaves injured by the tissue-mining stages of the spotted tentiform leaf miner (Phyllonorycter blancardella) contain higher levels of ethylene and 1-aminocyclopropane-1-carboxylic acid (ACC) than intact leaves or leaves injured by the sap-feeding stages of the same leaf miner (Kappel et al., 1987).

Another reaction by the plant to leaf-mining is to form callus tissue in the damaged mine tunnels. A callus is a loose layer of parenchymal tissue formed by not only meristems but also in fully developed tissues (Hering, 1951). The function of callus tissue remains unclear. One reasonable guess is that callus might help to repair the leaf mines and compensate for leaf damage in photosynthesis (Hering, 1951). The main mortality factor of Phytomyza ilicis is the increasing quantities of callus which may squeeze the larvae to death (Ellis, 2000). On the other hand, Liriomyza strigata sometimes feeds on callus tissues (Kang, 1996).

Cytokinins are phytohormones with significant roles in plant growth and development (Naseem et al., 2014), such as maintenance of chlorophyll (Engelbrecht et al., 1969) and inhibition of leaf senescence (Pritchard & James, 1984b ). Leaf miners could make “green islands” around leaf mines in autumn leaves, and leaf-mining larvae manage to complete their life history inside green islands. This phenomenon is attributed to high concentrations of cytokinins in the mined leaf area to keep living green tissues functional for photosynthetic activity (Giron et al., 2007). It is amazing that insect bacterial symbionts are found to manipulate cytokinin levels to induce green islands (Kaiser et al., 2010).

The activities of leaf miners such as oviposition and feeding have negative effects on plant photosynthesis, which varies with different leaf-mining depths. Generally, the impacts on photosynthesis from large to small should be: full depth mining > lower surface mining > interparenchymal mining > upper surface mining > epidermal mining (Table 1; Johnson et al., 1983; Parrella et al., 1985; Trumble et al., 1985; Kappel, 1986; Martens & Trumble, 1987; Raimondo et al., 2003; Bueno et al., 2007; Wagner et al., 2008; Liao et al., 2014). The possible explanation should be that different leaf tissues play different roles in photosynthesis. For example, stomata are mainly distributed on the lower epidermis and mining in the spongy tissue and the lower epidermis may lead to the close of stomata (Parrella et al., 1985; Wagner et al., 2008). Different leaf-mining depths thus have different impacts on leaf hydraulics, gas exchange, water use and chlorophyll fluorescence (Johnson et al., 1983; Parrella et al., 1985; Trumble et al., 1985; Kappel, 1986; Raimondo et al., 2003, 2013; Pincebourde et al., 2006; Bueno et al., 2007; Wagner et al., 2008; Lombardini et al., 2013). However, the whole-plant photosynthetic loss from most leaf miners is slight because insect populations are small in natural vegetation (Hileman & Lieto, 1981; Nardini et al., 2004) and the photosynthetic loss does not extend beyond the mined leaf area (Lombardini et al., 2013; Raimondo et al., 2013). While in some economic plants, photosynthesis is negatively correlated with leaf miner density and injury leaf area (Fujiie, 1982; Johnson et al., 1983; Lombardini et al., 2013).

Plant growth may be significantly subjected to leaf miner activities such as oviposition and feeding (Wagner et al., 2008). One interesting phenomenon is that mined leaves fall earlier than unmined leaves (Owen, 1978; Fujiie, 1982; Maier, 1983, 1989; Nagasaki, 2004; Nardini et al., 2004; Bueno et al., 2007; Wagner et al., 2008). The effects of leaf-mining on leaf life span are stronger than other herbivore guilds (Kozlov & Zvereva, 2014). Some researchers argued that premature leaf abscission is simply a response to herbivore damage (Pritchard & James, 1984b ; Stiling & Simberloff, 1989). While others suggested that early leaf fall is an evolutionary adaptation, which is beneficial to host plants themselves (Owen, 1978; Pritchard & James, 1984b ). First, premature leaf fall stops the further development of young leaf miner larvae lessening leaf miner numbers in the next generation. The falling of mined leaves earlier than unmined leaves may attribute high mortality of some leaf miners (Faeth et al., 1981; Simberloff & Stiling, 1987; Stiling et al., 1987, 1991; Auerbach & Simberloff, 1989; Preszler & Price, 1993; Connor et al., 1994; Waddell et al., 2001), since the development of most leaf miner larvae is restricted to a single leaf (Zvereva & Kozlov, 2014). Second, leaf miners could make leaves fall year-round and return the nutrients back to the tree in a more even way (Owen, 1978). Premature leaf abscission is positively correlated with the densities of the mines (Shinozaki et al., 2012). Infested cranberry leaves by Coptodisca negligens, those with two mines drop earlier than those with only one mine (Maier, 1989). Leaf-mining (and other herbivorous feeding) damage to young leaves and to plants with a higher specific leaf area reduce more in leaf life span than damage to old leaves and to plants with a lower specific leaf area (Zvereva & Kozlov, 2014). Premature abscission due to leaf-mining is lower for deciduous trees than for evergreen ones (Bultman & Faeth, 1986). Impacts of mining on plant leaves differ from season to season or from insect generation to generation (Pritchard & James, 1984b ).

However, early leaf drop is not a characteristic of all plants infested with leaf miners and appears to be a relatively minor ecological factor. For example, <40% of mined leaves drop early (Kahn & Cornell, 1983); less than 3% of leaf miner larvae in dropped leaves suffer mortality (Pritchard & James, 1984a ; Oishi & Sato, 2007; Gripenberg & Roslin, 2008; Shinozaki et al., 2012); and photosynthesis is not reduced enough to severely damage the host plant (Parrella et al., 1985; Martens & Trumble, 1987; Raimondo et al., 2003; Bueno et al., 2007). Moreover, early leaf abscission has no influence on holly leaf miner survivorship (Kahn & Cornell, 1989). Possible reasons are that fallen leaves onto the forest floor would escape searching by parasitoids and the maggots are already in the pupal stage (Kahn & Cornell, 1989). Coptotriche japoniella can prevent premature leaf abscission to make sure the sucessful emergence of the imago before leaf falls (Oishi & Sato, 2007). Some plants retain and change the visibility of leaf mines to deter other herbivores and to attract natural enemies (Yamazaki, 2010).

Leaf-mining may have a detrimental effect on whole plant primary productivity (Nardini et al., 2004). Leaf miner outbreaks have strong negative influences on plant development and aboveground biomass production: more basal sprout dieback, lower plant height, lower girth, smaller leave size, fewer leaf number, fewer petiole number, and fewer shoot number (Trumble et al., 1985; Norris, 1997; Kozlov, 2005; Wagner & Doak, 2013). Leaf-mining injury could cause severe loss of the total leaf area (Hileman & Lieto, 1981; Peña et al., 2000; Nardini et al., 2004). Leaf miners markedly increased the mortality of floral leaves, but not of vegetative leaves (Hileman & Lieto, 1981). Leaf miners could significantly reduce whole-tree energy, seed weight, seed germination, seedling vigor and stem extension (Percival et al., 2011; Straw & Williams, 2013). Serious leaf-mining could also lead to reduction in fruit set, fruit size, fruit weight and fruit quality, and even premature ripening, early fruit fall, and total yield loss (Fujiie, 1982; Reissig et al., 1982; Peña et al., 2000; Nardini et al., 2004; Straw & Williams, 2013). However, leaf-mining does not reduce plant productivity in three Citrus L. species (Raimondo et al., 2013). Leaf-mining could also mediate the competition between different plant species (Norris, 1997).

Leaf miners can change resource availability to other organisms. First, other arthropods such as springtails can use empty leaf mines as feeding and reproductive sites (Kagata & Ohgushi, 2004). We also noticed that many insects and mites overwinter inside abandoned leaf mines in the field. Second, leaf-mining damage to leaf midribs may reduce the survival rate of an aphid (Johnson et al., 2002). Third, the feeding galleries of leaf miners can be an entrance for pathogens. For example, the Asian citrus leaf miner (Phyllocnistis citrella) has increased the incidence of Asiatic citrus canker (Xanthomonas axonopodis pv. citri) and changed the spatial distribution type of canker-infested trees (Chagas et al., 2001; Christiano et al., 2007). Moreover, leaf miners may transmit plant virus as a vector. A leaf-mining beetle (Dactylispa lenta) unexpectedly transmits rice yellow mottle virus in Africa (Banwo et al., 2001). Another leaf-mining beetle (Phylloplatypus pandani) can also transmit diseases among plants (Sugiura & Masuya, 2010).

Leaf miner taxonomy

Researchers have described some major leaf miner groups in China, although some publications have not emphasized much on leaf-mining habit. Well-studied leaf-mining moths (Lepidoptera) included Gracillariidae (Liu & Zeng, 1985, 1989; Yuan, 1986, 1992; Yuan & Robinson, 1993; Bai & Li, 2008, 2009a , 2009b , 2011a , 2011b , 2012; Xu et al., 2009; Bai, 2011; Kobayashi et al., 2011, 2013; Cai, 2013), Lyonetiidae (Wu et al., 2006), Nepticulidae (Yang, 1989; Nieukerken & Liu, 2000; Liu & Nieukerken, 2001), and Tischeriidae (Huang & Tan, 2009). Leaf-mining flies (Diptera) included Agromyzidae (Lin, 1983; Feng & Jer, 1989; Gu et al., 1991; Shiao et al., 1991; Shiao & Wu, 1995, 1996, 1999, 2005; Chen & Wang, 2001, 2003a , 2003b , 2003c , 2003d , 2008; Shiao, 2004; Sasakawa, 2006, 2008; Chen et al., 2007). Leaf-mining beetles (Coleoptera) included Alticinae (Wang, 1990; Zhang & Yang, 2004), Cassidinae (Gressitt, 1950, 1953; Chen et al., 1961, 1962, 1964, 1986; Gressitt & Kimoto, 1963, 1965; Chen & Tan, 1964, 1985; Takizawa, 1978; Yu, 1985; Lee, 2009), and Rhynchaenae (Yang et al., 1991, 1996). Leaf-mining sawflies included Tenthredinidae (Wei, 1994, 1997; Nie & Wei, 1998; Wei & Nie, 1998, 1999; Wei et al., 2003; Wu et al., 2010; Hou et al., 2012). However, the reported number of Chinese leaf miner species is much lower than estimated. For example, China hosts much more species in Fagaceae family than North America (7 genera and 294 species vs. 5 genera and 97 species) (Nixon, 1997; Huang et al., 1998), but species number of Tischeriidae are 4 vs. 40 (J.-S. Xu et al., unpubl. data). There are over 400 leaf-mining species feeding on 600–1000 Fagaceae plant species in the whole world, with only 10% leaf miners with Chinese names (X.-H. Dai et al., unpubl. data). Our Leaf Miner Group began to collect and rear leaf miners in 2007, and at present we have built a database on Chinese leaf miners and their host plants collected in 25 provinces. We have got nearly 300 leaf-mining species in over 30 families and four orders, with about 140 described species. We have also obtained 3000 parasitoid specimens (X.-H. Dai et al., unpubl. data).

Host plant diversity

Not much systematic work on host plant diversity has been completed in China (Dai et al., 2011b ), except for some economically important leaf-mining pests such as Liriomyza spp. (Kang, 1996; Dai et al., 2000, 2001d ; Zhang et al., 2000; He et al., 2001) and Phyllocnistis spp. (Liu & Zeng, 1980). Host plant records of non-economic leaf miners are still poor in China. At Mt. Jiulianshan in South China, from 2008 to 2011 we have collected 69 families, 155 genera, and 293 species of host plants of leaf miners (Dai et al., 2011a ), while in 2014 the numbers have increased to 92 families, 232 genera, and 471 species, occupying about 1/5 of total plant species number in the area (X.-H. Dai et al., unpubl. data).

Insect-plant relationship

Chinese scholars have done many studies on the interactions between leaf miners and host plants, mainly on Liriomyza spp. Topics are as follows: (1) host selection and location by leaf-mining insects (e.g. Wei et al., 2000; Dai et al., 2001c , 2003; He et al., 2001; Zhao & Kang, 2001, 2002a , 2002b , 2003; Zeng et al., 2003; Pang et al., 2004, 2006; Zhang et al., 2004; Ko, 2006; Yan et al., 2008; Yu et al., 2008; Zheng et al., 2010); (2) plant resistance to leaf miners (e.g. Cheng et al., 2006; Zhang et al., 2008); (3) influences of different host plants on leaf miners (Zhang et al., 1998; Li et al., 2002; Wei et al., 2004; Pang et al., 2005); (4) impacts of leaf miners on plants (Sun et al., 2012); (5) effects of environmental factors on insect-plant interactions (e.g. Dai et al., 2001a , 2001b , 2002, 2011a , 2013, 2014; Cen et al., 2003); (6) coevolution and codiversification (e.g. Du et al., 2008; Liao et al., 2015).

Different leaf miners have variable degrees of impact on the morphology, chemistry, physiology and growth of their host plants, while different plant species show inconsistent resistance or tolerance to leaf-mining (Trumble et al., 1985; Williams, 1989). The influence of leaf-mining on plants is also affected by other ecological factors such as light, water, soil, temperature and biotic relationship. Such factors not only directly act on plant growth and development, but also on the performance of leaf-mining insects. In turn, the induced changes of plant physical and chemical features may have either positive or negative impacts on oviposition and feeding of leaf-mining insects. Studies on plant responses to leaf-mining and the defensive mechanisms of plants are helpful in understanding the coevolution between leaf miners and their food plants. Plant resistance to leaf-mining may be altered by both the environment and genetics, therefore we could adopt suitable management strategies and genetic breeding technology to control pest leaf miners. Future research trends may include but not be limited to: (1) cophylogeny and cophylogeography between leaf miners and host plants; (2) coexistence and niche differentiation of different leaf-mining species in the same plant; (3) how plant community diversity affect leaf miner abundance and richness, and vice versa.