Economic Importance

Agromyzidae certainly attract the most (negative) attention when they are competitors of human beings. Yet a few species are beneficial in damaging weeds, especially in some parasitic plants. Some of them are effective control agents. In this section both aspects of agromyzid impact to human civilization are discussed.

DAMAGE CAUSED BY AGROMYZIDS
Agromyzids occur on various plants, including trees and cereals, which may be used by human beings. However, the main problems with mining flies on cultivated plants are concentrated on a relatively small selection of vegetable crops, ornamentals and tropical legumes. The main pest species are a relatively small number of Liriomyza- and Chromatomyia-leaf miners and Ophiomyia-stem miners.
Leaf-mining agromyzid larvae reduce the vigor of their host plant by both consuming photosynthetically active leaf tissue and inducing plant response. Heavy infestation not only reduces the yield but in some cases also causes high plant mortality. That is especially true for some stem miners, which can affect the transport and allocation of resources by damaging vascular bundles of the stem and the root. Yet for aesthetical reasons, also a much lower level of infestation can be of commercial significance; a few feeding traces of miners can render some products almost unmarketable. Thus, the purpose of the crop can be important for considering control measures (see Heinz and Parrella, 1992).
Sometimes even the feeding punctures caused by females can become a problem when lots of them damage the leaf surface (Parrella et al., 1985, Hendrickson and Barth, 1978).

In the course of environmental and climatic changes the pest status of many species may change concurrently. A species may cause some heavy outbreaks and afterwards is hardly found on cultivated plants again. In future some so far unknown species may become a serious threat for certain crops.
Previous historic developments are only partly understood: Some species became an economic problem since they managed to extend their geographical distribution. This is due to the increased international trade that enabled some major pest species to establish themselves in new areas (for example Agromyza frontella, Liriomyza trifolii, Liriomyza huidobrensis).
In some cases, the application of some insecticides may actually result in a rise of miner fly populations since naturally occurring parasitoids can be stronger affected than the pest species itself. The problems with some species can even be characterized as actually created by chemical pest control (e.g. Spencer, 1973, Lange et al., 1980, Saito et al., 1993).
Pesticides might have initially been applied to get rid of other insect pests than Agromyzidae. Instead, by wiping out agromyzid enemies, a new pest problem may arise. This sort of interference between different pests was observed for example in greenhouses in the Netherlands (Minkenberg and van Lenteren, 1986).

Monocultures may reduce the specificity and effectiveness of host selection because of the strongly reduced risk to find an unsuitable host plant. When an alternative crop is planted adjacent to the traditional host plant of the agromyzid, the reduced specificity may enable some individuals to establish themselves on the new plant. Indeed, it seems reasonable to suspect that the ability of some Liriomyza species to infest a large array of vegetable plants is the result of human-induced evolutionary processes occurring on cultivated areas.

CONTROL MEASURES
In spite of frequently observed resistance against insecticides, especially in some Liriomyza species (Parrella and Lindquist, 1983, Parrella et al., 1981), chemical treatments are in many cases indispensable. Especially abamectin and the insect growth regulator cyromazine seem to be effective against many leaf miners, at least at the time of the preparation of this text.
In greenhouses, biological control by cumulative release and support of parasitoids is widely applied. Parasitoids active against several agromyzid pest species in greenhouses (mainly Liriomyza) are commercially available. Most of the parasitoids are quite unspecific and can be applied for several agromyzid pests, even in different genera (Minkenberg and van Lenteren, 1986 and references therein, van der Linden, 1992, Benuzzi and Raboni, 1992).

A number of parasitoids were actively introduced into areas where alien agromyzids have established themselves. (Harcourt et al., 1988, Waterhouse and Norris, 1987). However, because of various reasons these introductions are often not successful. Murphy and LaSalle, 1999 argued, instead of introducing foreign species, native parasitoids with the potential to adopt the newly invaded agromyzids should be supported.
However, the efficacy of parasitoids can be strongly diminished by the application of insecticides against agromyzids or other pest insects. Some further informations on this problem and other natural enemies of Agromyzidae is given in the section on Bionomics.

The second important attempt to biological control of Agromyzidae is breeding resistant or pest-tolerant plants. This route was extensively pursued to protect tropical beans and peas (Shanower et al., 1999, Talekar, 1990). However, resistant accessions often failed to be economically successful because the taste and quality of most of them are often inferior to susceptible ones (Waterhouse, 1998).
Not much research is done on the parameters that repel agromyzids to develop on a certain plant. The females of several species are deterred from oviposition by dense trichomes on the leaf surface (Fagoonee and Toory, 1983, Chiang and Norris, 1983 a).
Other parameters influencing the well-being of the larvae can be stem diameter, leaf area, water content and chemical characteristics (see Chiang and Norris, 1983 a, Chiang and Norris, 1983 b, Chiang and Norris, 1983 c, Talekar et al., 1988, Wei et al., 2000).

The knowledge of the bionomics of the flies and their relation to the host plants might give some clues for the protection of cultural plants. For example, alternative host plants could be removed from the vicinity of the plantations; if the larvae are known to pupate in the host plants, the leaf litter remaining after harvest should be strictly removed from the field. Such measures can often lower the agromyzid abundance of the subsequent generation.
Young bean plants infested by Ophiomyia phaseoli are particularly vulnerable, for the larvae often destroy vascular tissue near the root. Only those plants, which manage to develop adventitious roots above the destroyed tissue have a chance to survive. Plant survival can considerably be improved by irrigation and planting in furrows filled with soil, thus facilitating the growth of adventitious roots (Talekar, 1990).

BENEFICIAL AGROMYZIDAE
Theoretically, all agromyzid flies feeding on unwanted plants can be considered as beneficial from the human point of view. Most of them, however, do not receive much attention since they are often one among several other phytophagous insects and have a limited impact to their host plants under natural conditions. One of the reasons for that is the presence of parasitoid and other natural enemies that keep the natural agromyzid populations under control. The situation is different if a neophyte weed has to be controlled. Agromyzids might be possible candidates because they normally are highly host specific. Their efficacy might be higher in new areas if they were introduced without their parasitoids. However, both the specificity and the efficiency has to be tested rigorously prior to an expensive and risky introduction. The effect of introduced control agents generally should not be overestimated because it is likely that sooner or later native predators and parasitoids become adapted to the new agromyzid species and diminish its impact.
To illustrate the merits and problems of Agromyzidae as weed eater and to save the reputation of the Agromyzidae, some cases are treated on this CD-ROM: Calycomyza lantanae, Ophiomyia lantanae are both specialized to the weed and ornamental plants of the genus Lantana. Ophiomyia strigalis. infests the seeds of the parasitic plant Striga in Africa. The European Napomyza lateralis is discussed as control agent for scentless chamomile (Matricaria) in Canada (Hinz and McClay, 2000).
Some other potential weed control agents were not mentioned on this CD-ROM: Phytomyza vitalbae Kaltenbach, 1872, leaf miner on Clematis vitalba L. was introduced to New Zealand to control where the host plant, viz. Clematis, grow rampant (Hill et al., 2001). Melanagromyza cuscutae Hering, 1958 is a seed head and stem miner of the parasitic weed Cuscuta spp. (Cuscutaceae) (Spencer, 1973). Further examples were summarized in (Spencer, 1973); a survey about some European weed control agents were given by (SpasiƧ and Smiljanic, 1998).
Yet the perhaps best-known control agent among Agromyzidae is Phytomyza orobanchia. This species was successfully used as control agent in its natural habitats. Their host plants, some dangerous parasitic plants of the genus Orobanche, has been controlled by inundate release of adult specimens (Klein and Kroschel, 2002).