Terrestrial Slugs in Neotropical Agroecosystems

Slugs can be important agricultural pests in tropical regions. They are also intermediate hosts of parasitic nematodes, such as Angiostrongylus costaricensis and A. cantonensis, which can cause abdominal and cerebral angiostrongyliasis in humans. Management of slugs in conventional agriculture has relied heavily in the use of pellets containing metaldehyde. In this article, we review cases of slug problems and their management in neotropical agroecosystems.


INTRODUCTION
Terrestrial gastropods (land snails, semi-slugs and slugs) are a very diverse group, with ∼24,000 described species and 11,000-40,000 undescribed species worldwide (Lydeard et al., 2004). Several species of terrestrial gastropods are considered pests, however this fraction is small compared to the diversity of terrestrial Gastropoda (Neubert et al., 2019). Terrestrial gastropods contribute to litter decomposition (Meyer et al., 2013), and may even have roles in plant disease control in agroecosystems (Hajian-Forooshani et al., 2020).
Among the terrestrial gastropods, slugs have been identified as one of the most successful pest groups (South, 2012), especially in crops that require high physical disturbance (Port and Ester, 2002). For example, the Iberian slug Arion vulgaris (Moquin-Tandon, 1855) is a serious pest of vegetables and arable crops in Europe (Kozlowski et al., 2018;CABI, 2020), and is considered one of the top invasive species by the European Network on Invasive Alien Species (Slotsbo, 2014). In the US Midwest, slugs are one of the main pest problems for no-till growers (Douglas and Tooker, 2012).
Terrestrial slugs are not a monophyletic group, but a case of convergent evolution in which the slug form evolved from different lineages of land snails that gradually lost their shell, through a process called limacization (Simone, 2018). The slug body form is present in the Stylommatophora (land snails and slugs) and Systellommatophora (aquatic and terrestrial slugs) clades of the Eupulmonata (Schrodl, 2014). Thomé and Gomes (2011) cited 13 families of slugs in Brazil, of which only two of them are native to the Americas: Veronicellidae (common in tropical areas around the word) and Philomycidae (native from Asia and Northern America). Thomé (1993) reviewed the native Veronicellidae in the Americas, mentioning 144 species names classified in 18 genera.
Limacization resulted in adaptive radiation in land snail lineages, as slugs became adapted to diverse moist and protected spaces, such as crevices in rocks and wood debris (Hausdorf, 2001). The loss of the shell also allowed for more movement and less calcium dependence, making slugs more successful as pests (South, 2012). For example, in Venezuela, farmers do not consider snails pests but consider that slugs are a significant pest problem (Perichi, 2014). Slugs in neotropical agroecosystems have been poorly documented in the literature. An exception to this, was the study of the bean slug, Sarasinula plebeia (Fischer, 1868), infestations in Central America in the 1980's. It was estimated that more than 400,000 farmers suffered economic losses due to this slug (Andrews, 1987a). Veronicellid slugs are also considered the main intermediate hosts of Angiostrongylus costaricensis (Morera and Céspedes, 1971), and can be naturally infected with A. cantonensis (Chen, 1935), which can cause abdominal and cerebral angiostrongyliasis in humans, respectively (Valente et al., 2020).
In this mini-review, we explore slug problems and management in the Neotropics. We review the literature: (1) to describe examples of slug infestations in the Neotropics and their potential causes; and (2) to explore preventive management strategies that have been used for slugs in this region. We expand on the review by Rueda et al. (2002), by including slug experiences from South America and the Caribbean.

METHODS
We conducted a literature review using the following search terms in the topics category: slugs AND pest AND management. We used Web of Science All Databases, which include: WOS, BCI, CABI, KJD, MEDLINE, RSCI, SCIELO, ZOOREC. We search for all available years (1864-February 2020). A total of 1,320 records were obtained in 89 research categories. We refined research results to limit results to "Agriculture" research area, resulting in 866 records. Most records are from Europe (526 records) and North America (114 records). South America (16 records) and Africa (13 records) are the continents with the least number of records. In order to find more articles from the Neotropics, we conducted additional searches using the terms "babosas" (71 records), "lesmas" (13 records), and "limaces" (99 records). In addition, we used the term "babosas" and "lesmas" in Google search engine to find information in local management guides, thesis, or technical reports.

SLUGS AS PESTS IN THE NEOTROPICS
The first reports of crop damage by slugs in the Neotropics are from Jamaica, where slugs were observed attacking coffee (Cockerell, 1893a) and strawberries (Cockerell, 1893b). In Trinidad, slugs caused significant damage to sweet potato plantations early in the 20th century (Callan, 1941). Pereira and Gonçalves (1949) later described slugs that affected bananas and vegetables in the coast of São Paulo state, Brazil. After a hiatus in publications, more articles about slugs in the Neotropics were published in the 1980's. Since the late 1960's, the bean slug S. plebeia started to cause severe damage to beans, a staple food, in Central America (Rueda et al., 2002). The Integrated Pest Management Project of Zamorano University (with financial support from USAID) conducted many studies to find suitable management practices to control S. plebeia. Sarasinula plebeia is believed to be native to the Americas despite its widespread distribution throughout tropical regions (Daglio et al., 2020). Andrews and Dundee (1987) studied the historic reports of the species and concluded that it was accidentally introduced to El Salvador, from where it spread throughout Central America, causing havoc to Central American farmers. In South America, S. plebeia is also known to cause damage to bean crops (Sannazzaro et al., 2000), but it is not considered a major pest (Cardona, 1994). Currently, S. plebeia is present in Central America, but farmers do not consider it a pest as important as whiteflies or leafhoppers in beans (R. Trabanino personal communication).
In the 1990's, the concern for slug damage intensified in Colombia, when higher than average rainfall was recorded, and significant coffee losses were experienced (Posada-Flórez et al., 2001). Slugs are considered pests in vegetable (Santacruz et al., 2011), flower and coffee production of Colombia. Martinez et al. (1994) first mentioned that slugs were problematic for flower production in Colombia, as they fed on the new shoots of plants and reduced flower production. Slugs are also problematic in flower production and many slugs are considered quarantine pests in importing countries (Robinson, 1999). Hausdorf (2002)   , a newly described species, was frequently intercepted by US authorities in fresh flowers coming from Colombia and Ecuador . C. confusus also feeds and can be problematic on coffee plantations in Colombia and Peru (Constantino et al., 2010;Gomes et al., 2013). After 2000, researcher groups were formed to study the systematics, biology and management of terrestrial mollusks in Colombia (Moreno Suárez et al., 2008). Currently, slugs continue to be a problem in Colombian floriculture (mainly for Alstroemeria, Dianthus and Hydrangea), but are not perceived as major pests (A. Paez personal communication).
In recent years, slug species have been reported to cause damage in several crops in the Neotropics, including vanilla (Velazquez-Montes de Oca et al., 2014), passion fruit (de Oliveira and Frizzas, 2014), strawberry (Landal et al., 2019;Castellanos Gonzalez et al., 2020), and grapes (Baronio et al., 2014;Rodriguez et al., 2019). Latipes erinaceus (Colosi, 1921) and Sarasinula linguaefomis (Semper, 1885) are other species that are becoming problematic in the Neotropics. They are currently, one of the main pests of large-scale soy production in Brazil. Observations of L. erinaceus in the lab, have shown that it has a higher reproductive capacity than other species of the Veronicellidae family, spawning every month (S. R. Gomes unpublished).
From a food safety standpoint, the presence of slugs in leafy greens and fresh fruit production in the Neotropics is of particular concern. Slugs eat and leave secretions in leaves and fruits. Terrestrial mollusks, including slugs, are vectors of the parasite A. costaricensis and A. cantonensis. These parasites can survive in produce that is consumed raw (Kramer et al., 1998). In Santa Catarina, Brazil, A. costaricensis infection of 86% was observed in S. linguaeformis (Laitano et al., 2001). S. linguaefomis can be a pest of corn, beans, soybeans and leafy greens (Grisotti and Ávila-Pires, 2011;Moura et al., 2018). This slug is considered native to Brazil and it is widespread throughout the neotropical region (Valente et al., 2020). It usually occurs in low numbers in Rio de Janeiro State (Oliveira et al., 2015). However, in 2015, we studied an outbreak of this species in small scale vegetable farms of the Cachoeiras de Macacu Municipality (unpublished). In 2020, this species was also found in Puerto Rico, affecting small-scale agroecological vegetable production. Recently, the exotic Chinese slug Meghimatium pictum (Stoliczka, 1873) was associated to a case of abdominal angiostrongyliasis in a region where it is a grape pest (Rodriguez et al., 2019). This brings attention to the possible risk of human infection associated with accidental ingestion of contaminated fruit or vegetables containing larvae of the parasitic nematode A. costaricensis, the etiological agent of this parasitosis.

SLUG MANAGEMENT IN THE NEOTROPICS
After examining abstracts and articles, we found 389 records that covered the topic of slug management. In general, seventyfive records (19% of records) covered metaldehyde use, and 55 covered other organic (i.e., with carbon) pesticides such as methiocarb. Records about metaldehyde use start in 1940 (with Gimingham, 1940), while records about organic pesticides start in the 1960's (with Henderson, 1968). Twenty-six records were about the use of inorganic compounds for slug management, which was the oldest slug management practice described in the literature (Baltet, 1889). Two studies in the Neotropics mentioned the use of lime for semi-slug and slug management (Bastos Garcia et al., 2012;Capinera, 2018). Literature about biological control of slugs is dominated by studies about carabid beetles (25 records) and nematodes (45 records). These records begin in the 1980's for carabids (Symondson, 1989), and early 1990's for nematodes (Wilson et al., 1994). Other slug management practices include the use of botanical extracts, traps, solutions of conspecifics, Bacillus thuringiensis, and barriers. Also, there were records that examined the effect of different farm management practices on slugs, such as tilling, use of cover crops, irrigation regimes and crop rotation. Most slug management records are from temperate areas and were recently reviewed by Le Gall and Tooker (2017).
The management of slugs (and snails) in conventional agricultural areas has relied heavily on the use of metaldehyde pellets. The use of metaldehyde (C 8 H 16 O 4 ) as a molluscicide began in the 1930's, after the discovery of its molluscicide properties by women gardeners in South Africa (Gimingham, 1940). In the Caribbean, it started to be used in 1937 (Callan, 1941). In South America, it has been known to be used since the 1940's (Pereira and Gonçalves, 1949). Slugs died 1-2 days after consuming metaldehyde pellets, but the adequate consumption of pellets does not always occur, making the method inefficient (Bailey, 2002). Metaldehyde is soluble in water, highly mobile in soils and generally stable to abiotic degradation (EPA, 2006). In Europe, it has been identified as a water pollutant, being frequently detected in surface waters above the EU Drinking Water Directive (Kay and Grayson, 2014). Metaldehyde was banned in 2018 in England because of its pollutant potential, however the ban was later overturned. Metaldehyde is commonly used by farmers in the Neotropics. Five (out of 11) management articles in the Neotropics examined the used of metaldehyde to manage slugs. Metaldehyde residue limits exists for the export of some products, such as legume foliage (EPA, 2015).
Pellets containing iron phosphate have been used since the 1990's as an alternative to metaldehyde pellets in some countries (EPA, 1998). Iron phosphate pellets can be an effective curative slug control method (Speiser and Kistler, 2002). Also, iron phosphate is non-harmful to humans or the environment (EPA, 1998) and can be used in organic agriculture (USDA-NOP). Recently, pellets containing ferric sodium EDTA have also become commercially available in some countries as another curative slug control method (Capinera and Rodrigues, 2015). Ferric sodium EDTA also has very low toxicity to humans and the environment (EPA, 2008). In the Neotropics, iron phosphate is commercially available but can be less accessible than metaldehyde molluscicides (slightly higher cost per gram and lower distribution to rural areas). Laboratory experiments in Brazil and Florida, showed that iron phosphate pellets can be more effective or slightly less effective than metaldehyde pellets, respectively (Baronio et al., 2014;Capinera and Rodrigues, 2015).
Several management practices were followed by farmers to manage S. plebeia in Central America. Beans are a staple food of family farmers in Central America, and are usually planted in a relay system with maize. Some of the management practices first implemented by Honduran farmers to deal with the new slug problem were: planting in slopes, burning maize residues before bean planting, deep tilling, and empirical pesticide use (Andrews, 1987a). From 1975 to 1987, the Honduran government subsidized the purchase of metaldehyde to distribute among farmers. Andrews (1985) argued that this subsidy resulted in less effective chemical control than that of Mexico or El Salvador, where private chemical companies lead control efforts. Several slug management techniques were evaluated at Zamorano, Honduras with farmer participation (Rueda et al., 2002). Some Honduran farmers preferred to combine different management techniques, including weed management, weed traps, night killing of slugs with a stake, and use of homemade baits (with Jatropha curcas seeds or metaldehyde) when there are more than 5 slugs per ten plants (FAO, 2005). Farmers using traditional "frijol tapado" systems in Costa Rica preferably use east-facing slopes and eliminate some of the plant species that favor slugs (Meléndez, 2004).
Agroecosystems in the Neotropics can range from smallscale highly diverse home gardens to large-scale conventional monocultures of crops such as sugar cane, soy, banana or palm oil. Family farming and agroforestry systems are also important in the region (Peters et al., 2016;Schneider, 2016), and can contribute to the conservation of biodiversity in these landscapes (e.g., Rooduijn et al., 2018). A comparison of slug abundance and plant damage in two agroecological and conventional farms in Colombia, showed that although slugs were more abundant in the agroecological farm, only one variety of lettuce had more damage in this system (Cordoba Vargas and Leon Sicard, 2010).
Agroforestry systems and landscapes with natural vegetation can benefit from higher predation of herbivores that feed on crops (Maas et al., 2020). Natural enemies of slugs include birds, reptiles, mammals, planarians, nematodes, insects, and mites, among others (Baker, 2004). In other regions, there has been an emphasis on the study of carabids and nematodes for slug biological control. In the Neotropics, natural enemies of slugs have not been studied in detail, but there are some documented and anecdotal examples. Firefly (Coleoptera: Lampyridae) larvae are known to feed on slugs (Viviani, 2001), but the rate of predation has been low under laboratory conditions (1 slug every 5 days; Rueda et al., 2002). Native planarians preferentially feed on introduced slugs and snails (Boll and Leal-Zanchet, 2015). Toads were used successfully in gardens of Colombia to reduce slug populations (Posada-Flórez et al., 2001). Since the Neotropics are very diverse, many natural enemies of slugs may remain to be discovered. For example, five new species of snakes that feed on terrestrial mollusks were recently discovered in Ecuador and Peru (Arteaga et al., 2018).

OTHER CONSIDERATIONS Global Change and Slugs in Agricultural Areas
The global distribution of terrestrial gastropods is changing. Humans have facilitated the dispersal of gastropod species across the globe, breaking geographic barriers and homogenizing global terrestrial gastropod diversity (Capinha et al., 2015). In this process, many species have become endangered, sometimes because of other mollusk pest biological control efforts (Lydeard et al., 2004). Other terrestrial gastropod species have benefited from human-facilitated dispersal, and have become invasive pests (Cowie et al., 2008). This was observed early on by Binnei (1871), when he described "All the species mentioned below are of foreign origin. They were imported from England. They are found only in close proximity to man. They have also been imported into other colonies of England, and probably are destined to become the most cosmopolitan of mollusks." There are at least 13 alien species of slugs in South America (Rumi et al., 2010;Gregoric et al., 2013). In Colombia, at least seven European slug species have been introduced in the highlands, some more than a century ago (Hausdorf, 2002).
Introduced non-native slug species can have detrimental effects in agricultural areas. This is the case of slugs such as Deroceras reticulatum in Australia (Nash et al., 2007), and the Cuban slug Veronicella cubensis (Pfeiffer, 1840) in the island of Rota (Robinson and Fields, 2010). We discussed examples in this article showing that this can also be the case in the Neotropics. Early on, Cockerell (1893b) described the introduction of European slugs in strawberries in Jamaica, stating that "They and their eggs come in the earth about the roots, and, in many cases, it must be practically impossible to detect them on arrival." In the case of the bean slug, the species was accidentally introduced from South America and caused damage to both conventional and diverse small-scale farmers in Central America (Andrews and Dundee, 1987). In the island of Puerto Rico, the non-natives V. cubensis and S. linguaefomis are causing problems to smallscale vegetable farmers. Introduced slug species are expected to thrive more in disturbed habitats, such as agricultural lands, than native slug species (Ryser et al., 2011). However, they can colonize natural areas and affect native plant communities (Shiels et al., 2014).
The range of invasive slugs may increase with climate change, since some species are favored by warmer conditions (Sommer and Cowie, 2020). Although terrestrial gastropods are susceptible to desiccation, they can also have costly behavioral and physiological mechanisms that help them cope with high temperature and drought periods (Nicolai and Ansart, 2017). Temperature increases, changes in rainfall patterns and increase of extreme weather events are expected for the Neotropics in the next 100 years. Traditionally, farmers associate periods of rains with higher slug abundance. Extreme rain events, such as hurricanes, decreased the abundance of the native semi-slug Gaeotis nigrolineata (Shuttleworth, 1854) in Puerto Rico (Willig and Camilo, 1991).

Slug Identification and Monitoring
Slug diversity in agroecosystems in the Neotropics should be studied and considered more. This could be a first step to detect potential new slug invaders, and to better understand the drivers of sudden slug infestations in farms. A caveat to this is that slug identification can be difficult. A combination of external characters and internal anatomy is usually required to correctly identify species. Robinson (1999) presents the example of the process from the USDA Plant Protection and Quarantine Program, in which: (1) port inspectors find mollusks in arriving shipments; (2) port identifiers (PI) try to identify the species; (3) if the PI cannot identify the species, the PI dissects the individual and takes digital images of it; (4) the images are rapidly sent to the National Malacology Laboratory for identification; (5) if needed, the individual is sent for molecular analysis.
Recently, a new network of malacologists was created to track the presence of non-native mollusks in South America (Darrigran et al., 2020). The group is an important first step to increase awareness about this topic in our region. More attention should also be given to train extension and other agriculture professionals in slug identification, and to improve the identification (e.g., molecular) tools of slugs in the Neotropics. Participatory strategies may also help in the detection of new slug invaders or infestations. Knowledge dialogues or exchanges with and between farmers are an important agroecological practice in Latin America (Mier y Terán Giménez Cacho et al., 2018). In the case of the bean slug in Honduras, farmers would prefer using "trash" traps (i.e., pilled plants residues from weeding and pruning) to monitor slug presence than other traps that required materials from outside the farm (Andrews, 1987b). Relying on damage observation was not recommended, as slug populations were difficult to manage at that point. In other regions, citizen science has been used to detect the occurrence of invasive slugs in private and public lands (Dorler et al., 2018;Morii et al., 2018). The participation of malacologists in social media groups, such as Facebook Groups "Moluscos del Ecuador" or "Biodiversidad de Puerto Rico, " can make slug identification more accessible to the general public.
We found few initiatives that were testing unmanned vehicles for slug detection in farms. In Colombia, drones were experimentally used to determine the presence of slugs in flower cultivation (Caceres Florez et al., 2015). The drones detected leave damage, not slug individuals. Because slugs are frequently hidden in the soil, land unmanned vehicles have also been considered in other regions (Godeke et al., 2019).

CONCLUSIONS
Slugs can be important pests in agricultural areas worldwide. In the Neotropics, reported examples of slug pests are fewer than in other regions such as Western Europe. Many factors could be hypothesized to explain this, starting with less abundance of malacologists studying terrestrial gastropods in the region. Also, the ecosystem service of natural pest regulation could be preventing slug populations to reach economic injury levels. Another possibility is that slugs are not considered pests by farmers, as observed for insect herbivores in traditional milpa systems of Guatemala (Morales and Perfecto, 2000). Alternatively, management of slugs is not studied extensively in the Neotropics because metaldehyde has been used to kill outbreaks of terrestrial mollusks for nearly 80 years. The examples presented in this article show that there are instances when slugs have been problematic in the Neotropics. It is important to understand these and other examples in order to prevent future slug pest problems in farms of our region.
Slugs and their management have been understudied in the Neotropics. More research is needed about their diversity and biology in our region. Weather extremes and international trade are expected with global change, and invasive slugs could benefit from these changes. Farmers already use a variety of techniques including barriers, traps and resistant crops, and socialize this information among themselves. However, these preventive management practices have not been sufficiently documented or validated in the scientific literature. The case of the bean slug provided important information about slug management in our region. However, these studies were conducted almost 40 years ago. New social and ecological technologies may exist to advance slug management in the Neotropics.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
OR-R and MR conducted the systematic literature review. SG contributed additional literature, slug infestation cases, and expertise on slug taxonomy and biology. All authors contributed to the writing of the manuscript, with MR and SG contributing the most.

ACKNOWLEDGMENTS
We thank R. Trabanino and A. Paez for their communications. Two reviewers provided-important feedback to improve the article. As part of this article was written during the COVID-19 pandemic, we would like to thank our families and editors for their support in the writing process.