In the study period 2010 to 2019, total imports of frog’s legs into the EU numbered 40.7 million kg. This total weight can be converted, when 1 kg equals 20–50 individual frogs (Veith et al. 2000), to at least 814 million and up to roughly 2 billion frogs.

The EU’s role of responsibility should also imply sustainable harvest and trade of species in supplier countries. The following example makes it clear what abuses accompany this trade; (1) according to Indonesia’s annual harvest/export quotas for F. cancrivora (there is no information on how Indonesia derives its annual quotas, so there is no data basis for sustainable trade), in the period 2016–2020, ca. 295.3 million kg were exported (1 kg equates to 15–22 individual specimens [cf. with Veith et al. (2000) above]; Indonesian Ministry of Environment and Forestry 2016–2020) resulting in ca. 5.4 million individual F. cancrivora; (2) to date there is no information on mortality prior to export. As early as 1986, an estimated pre-export mortality rate of 10–20% was reported, but mortality during the export process may be highly variable (Niekisch 1986). Herein, we assume that every export also includes an estimated number of dead animals for which the importer is also responsible; here we refer to the EU.

Regarding the export of live animals, wholesalers have been found to have mortality rates of around 45% for amphibians, meaning live trade levels may need to be in higher volumes to satisfy demand when many frogs die in transit, with many coming from the wild (Ashley et al. 2014).

In the study period 2010–19 (EUROSTAT 2020), Belgium leads EU countries in imported quantities of frogs’ legs, with a total of 28,430 tonnes (69.8%), ahead of France with 6790 tonnes (16.6%), followed by the Netherlands (2620 tonnes; 6.4%), Italy (1790 tonnes; 4.3%) and Spain (923.4 tonnes; 2.2%) (Table 1). Smaller quantities were imported by the United Kingdom (68.8 tonnes), Croatia (28.5 tonnes), the Czech Republic (27.8 tonnes), Poland (12.5 tonnes), Romania (2.8 tonnes) and Germany (1.8 tonnes). Within the EU, Belgium re-exports a large part of its imports to other EU countries. For example, Belgium re-exported 20,920 tonnes to France (> 73% of all its imports in the study period) and 1410 tonnes to the Netherlands (ca. 5% of all its imports in the study period) and, accordingly, Belgium consumed 21% of its total imports.

Due to the introduction of advanced technologies of freezing methods in the 1970s, storage constraints were reduced and transport routes of frogs’ legs became possible. This transformed the traditional frogs’ leg trade in France, bringing some local frog populations to the brink of extinction (Ohler and Nicolas 2017 and references therein). Since at least the 1980s, France has historically been considered the main consumer of frogs’ legs. According to Le Serrec (1988), France imported a total of 4522 tonnes of frogs’ legs in 1983. Based on this fact, France initiated studies to gain clarity on species composition as well as potential ecological damage from intense commercialised trade (MNHN 2012; Ohler and Nicolas 2017).

From 2010–19, France imported 30,015 tonnes of fresh, refrigerated or frozen frogs’ legs (ca. 600–1500 million frogs; Veith et al. (2000)), according to French customs statistics (https://leKiosque.finances.gouv.fr/). France’s main suppliers are Indonesia (24,102 tonnes or 80.3%), Vietnam (3941 tonnes or 13.1%), Turkey (1017 tonnes or 3.4%), Belgium (226 tonnes or 0.8%) and Albania (219.6 tonnes, 0.7%). For the same period, the quantities imported from Belgium to France differ widely depending on whether the data source is Eurostat or French customs due to two different statistical concepts. France separately lists the country of direct export origin and country of original export when the country of origin is not an EU country. Original origin prevails in the French statistical data. As a result, some frogs’ legs are considered by the French methodology as imported from Indonesia and not from Belgium, even if they have transited through Belgium. Annual imports did not fluctuate significantly between 2017 and 2020, with an average of 2669 tonnes/year. A drop to 1826 tonnes is prominent in 2021, still a relatively high figure despite the paralysis of international trade due to COVID-19. Similarly, France also is a hub for re-exportation of frogs’ legs. From 2017–20, France shipped 385 tonnes of frogs’ legs, mainly destined for markets in Belgium (292 tonnes; 75.8% of total tonnage shipped), Luxembourg (24.4 tonnes; 6.4%) and Germany (16.6 tonnes; 4.3%). In 2021, it is notable that France also re-exported 13.9 tonnes (3.6%) to Vietnam.

Results of the online market survey in December 2021 indicate 20 frogs’ legs food products readily available. Of these 20 products, 11 originated from Indonesia, three from Vietnam, one from France and one from the “EEC (Turkey, Albania etc.)”. This last indication is confusing because the European Economic Community (EEC) was dissolved in 1993 excluding Turkey and Albania and both are not EU member States. With regard to the indication of France as a source country, these products are pre-cooked frogs’ legs that do not originate from France and the species indicated is “wild Limnonectes [Rana] macrodon” endemic to western Indonesia (cf. Table 2). Four sources do not provide information on the country of origin within the product description or packaging. Regarding species name, six sources indicate Rana macrodon, three Fejervarya cancrivora, another three Hoplobatrachus rugulosus, one “Rana macrodon or Fejervarya cancrivora” (here we assume the sourcing from different suppliers, resulting in insufficient traceability for species identification) and one Rana esculenta.

For six sources, both product description and packaging do not indicate a species name. With regard to EU legislation, lack of information (species or country of origin) is a violation of EU rules [Commission Regulation (EC) No 2065/2001 of 22 October 2001 detailing rules for the application of Council Regulation (EC) No 104/2000 as regards informing consumers about fishery and aquaculture products; https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32001R2065&from=FR). In eight sources, origin is highlighted as “wild”, three refer to “fishing” (e.g. fresh water, rice fields) and, in one, “collected” is indicated as the source. Not a single product, however, indicates a captive bred or farmed source. Besides raw or cooked frogs’ legs, “frairine” is also offered for sale, a mixture of pork and frogs’ legs seasoned with white wine. For this mixed product, there is no information on the origin or species involved.

An additional market survey through Google Alert for more than 10 weeks (see Methods) identified 38 commercial offers for frogs’ legs (20 from Belgium and 18 from France). Regarding the offers from France, trends from the December 2021 study are largely confirmed, with only one offer indicating an origin “Vietnam and/or Indonesia captive bred”.

In addition to imports, the French market is also supplied with wild-caught native species. Short marketing circuits, such as local restaurants, are supplied with Rana temporaria, a nationally protected species in France (https://www.legifrance.gouv.fr/loda/id/JORFTEXT000017876248/, accessed April 2022, see Suppl. material 3). Despite the legal framework for harvest, numerous exemptions are granted. For example, > 2 million R. temporaria are legally caught each year in the Franche-Comté region (https://www.bourgogne-franche-comte.developpement-durable.gouv.fr/ranaculture-bourgogne-franche-comte-dossiers-de-a6583.html, accessed June 2022, see Suppl. material 3). An exemption may exist if an offtake of < 1500 frogs is requested, as this is considered “familial”. Poaching offences are also recorded and a distinction is made between captures without a permit, those exceeding quotas or if the captures are outside authorised time periods. In October 2018, a couple were fined €2500 for the capture of 4000 R. temporaria, even though they possessed a permit for the capture of 1000 specimens (https://robindesbois.org/en/a-la-trace-n23-le-bulletin-de-la-defaunation/RobindesBois, “On the Trail” No. 23, 2019). In the same year, during eight inspections and three searches conducted under a judicial warrant, a total of 171 traps were seized, enabling the release of 17,950 grass frogs (R. temporaria) and 10 m³ of eggs into the natural environment (Office national de la chasse et de la faune sauvage; ONCFS, 9 May 2018).

There is no doubt that the trade in frogs’ legs for consumption is a global issue, with most countries involved in the trade as exporter, importer or some combination (Gratwicke et al. 2010; Suppl. material 1: figs 2, 3). In recent decades, there have been four major source regions exporting edible frogs or body parts (wild and/or farmed) into the EU: (1) East Asia, i.e. China and Taiwan (Warkentin et al. 2009; Altherr et al. 2011; Shreshta 2019), (2) Southeast Asia, i.e. Indonesia and Vietnam (Niekisch 1986; Kusrini and Alford 2006; Warkentin et al. 2009; Gratwicke et al. 2010; Ohler and Nicolas 2017; Shreshta 2019), (3) South Asia, i.e. India and Bangladesh (Niekisch 1986; Le Serrec 1988; Warkentin et al. 2009) and (4) eastern Europe i.e. Turkey and Albania (Warkentin et al. 2009; Şereflişan and Alkaya 2016; Çiçek et al. 2021). The United States, another major importing country for frogs and their body parts, is supplied from Asia and South America (Warkentin et al. 2009; US USFWS-LEMIS Database 2023). Based on LEMIS data, main suppliers for the US market for L. catesbeianus were Mexico (labelled as wc, “wild capture”), Ecuador (farmed) and China (farmed). Hoplobatrachus rugulosus was imported from Thailand (farmed) and Vietnam (wc) and L. forreri only from Mexico.

Figure 2.

EU’s frogs’ legs imports (tonnes) during the period 2000-2019. Source: EUROSTAT (2020).

Figure 3.

Anuran species imported for the purpose of consumption into the US in the period 2015-20, in which weight (left) is compared to the number of individuals (right) to illustrate how unequally these variables are aligned with each other. Source: USFWS-LEMIS Database (2023).

For most recent trade routes from source countries to importers and consumers into the EU, see Fig. 1.

Within the study period 2010–19, Indonesia clearly represents the leading supplier for the European Union’s frogs’ legs with 30,019.4 tonnes (74%), followed by Vietnam (8439.4 tonnes; 21%), Turkey (1593.7 tonnes; 4%) and Albania (586.5 tonnes; 1%) (Table 1, Fig. 1).

Comparatively smaller amounts were supplied by China (37.7 tonnes), India (15 tonnes), Thailand (9.2 tonnes), Malaysia (7.6 tonnes) and South Korea (0.3 tonnes), resulting in less than 1% of the EU’s total imports (EUROSTAT 2020).

Import data for the period 2010–19 were compared with data of the previous decade (see Altherr et al. (2011)) and three trends stand out: (1) a decrease of roughly 12.3% in EU imports of frogs’ legs (now 40,700 tonnes instead of 46,400 tonnes) with marked fluctuations underscoring this decline (Fig. 2), (2) the role of Belgium as the highest importing country with 70% of imports in the period under review (in contrast, France’s import volumes decreased from 23% to 17% and those of the Netherlands’ from 17% to 7%) and (3) the significant increase in the role of Vietnam in exporting frogs, from 8% to 21% of total imports, with China simultaneously dropping from 3% to less than 1%.

Forensic studies have shown that the species composition and labelling in Indonesia’s trade has changed over recent decades (Ohler and Nicolas 2017). Fejervarya limnocharis and Limnonectes macrodon were amongst the most common documented species exported (Kusrini 2005), but F. cancrivora represents the major species in trade.

While this study focuses on the EU, the current role of the United States is briefly highlighted, as the US also represents a major consumer of frogs’ legs (cf. Warkentin et al. (2009); Gratwicke et al. (2010); Altherr et al. (2011)). In the period 2015–2020, at least four anuran species were imported by the US for consumption, Lithobates catesbeianus, L. forreri, L. grylio and Hoplobatrachus rugulosus (USFWS-LEMIS Database 2023). Lithobates catesbeianus (either alive, dead or legs only) represented the major species by a large margin, predominantly supplied by Mexico (mainly wild), Ecuador, and China (farmed) (Fig. 3). This species, the American Bullfrog, Lithobates catesbeianus, has also been widely introduced into Latin America and Europe for commercial breeding purposes (Carraro 2008). In 2018, imports of H. rugulosus emerged and were declared as exports from Thailand either as captive-bred or ranched, while exports from Vietnam also included wild individuals. Mexico exclusively supplied the United States with wild sourced L. forreri, shipped as meat or legs. In 2015–16, the US imported more than 90 tonnes of meat of L. grylio, all noted as captive bred (USFWS-LEMIS Database 2023), but this species is native to the United States (Fig. 3). It is noteworthy that the large quantities of frogs’ legs of species harvested in Indonesia and eastern Europe have no sales in the USA.

As can be seen in the individual IUCN Red List assessments on exploited amphibian species (Suppl. materials 1, 4, Fig. 2), many species are harvested at local/national levels for consumption, medicinal and/or spiritual purposes (e.g. Nepal 1990). Although this issue is not the focus of this paper, some light can be shed on aspects of local use of frogs for consumption from a conservation perspective. International trade activities can only claim to be sustainable if offtakes for national needs are also managed sustainably. This implies that monitoring of harvest levels for both local/national and international consumption need to be in place (Leader-Williams 2002). There are numerous published examples that describe the domestic trade of amphibians and the impact it may have on local frog populations. Species harvested for consumption within national borders and across range States, are reported for Greece (Hatziioannou et al. 2022), West and Central Africa (Mohneke et al. 2009, 2010; Akinyemi and Ogaga 2015; Efenakpo et al. 2015), Burundi of eastern Africa (Verbanis et al. 1993), India (Pandian and Marian 1986; Ahmed 2012; Talukdar and Sengupta 2020), Nepal (Shresta and Gurung 2019), PDR China (Zhang et al. 2008; Chan et al. 2014; Turvey et al. 2021), Malaysia (Hardouin 1997), Vietnam (Nguyen 2000), Mexico (Barragán-Ramírez et al. 2021) and the USA (Ugarte 2004; Ugarte et al. 2005), as exemplars of some countries/regions. The proportion of national vs. international trade is of particular interest when some countries document high annual exports for the international frogs’ legs industry on a regular basis, while ignoring the fact that some species have been consumed locally for decades/centuries (Angulo 2008; Onadeko et al. 2011; Ahmed 2012). It would not be problematic if species are traditionally consumed at the local/national level and this use was deemed sustainable. However, harvest for international exports (above local/traditional harvest) often means overexploitation of local populations (Oza 1990 and cf. species compiled in Suppl. material 4). In addition, for Indonesia, it has been estimated that offtakes of edible frogs on a national level are up to 142 million frogs or seven times as many as that of annual international exports (see Kusrini (2005)), with no documentation of the impact on wild populations and highlighting the need for better monitoring of base populations and trade.

The conservation of species in trade only makes sense if the species or species complexes are known. The trade in animals with unclear taxonomic status ignores a fundamental condition, namely the lack of any data basis for taxa to, for example, conduct a non-detriment finding for the species to evaluate threat (see below). In order to obtain an overview of the species involved in the food trade (whether at local, national or international level) and their respective origins, the IUCN Red List was filtered (Fig. 4; Suppl. material 4). Regions where most species are harvested for consumption are Southeast and East Asia and it is also these regions that supply the EU market with most of their frogs’ legs. Furthermore, many species are consumed in Central America and (northern) South America, all of which are traded either locally, nationally or exported to the USA (predominantly L. catesbeianus from breeding farms; https://www.fao.org/fishery/en/culturedspecies/rana_catesbeiana/en, accessed March 2022, see Suppl. material 3). Interestingly, the EU is not a consumer of species from these regions. Likewise, all species consumed in Africa, with the West African region forming a species focus, are consumed in Africa and the EU is not a consumer of African species (Suppl. material 1, Fig. 4).

Figure 4.

Number of species per country in trade for consumption, see Suppl. material 1: figs S2, S3 for more detailed range data and for species in international trade. Notably African species are largely consumed domestically rather than exported (Suppl. material 1: figs 2, 4). Source: IUCN (2022).

At least 187 species of anurans and salamanders/newts are collected locally/nationally for food and for the international frogs’ legs industry (Suppl. material 4). According to information of Red List assessments, the local/national use of 13 species (filtered by the search criteria given above) was not explicitly stated, was more generally indicated (i.e. “species in the genus are also commonly used for food”) or the use has been not necessarily considered a threat (e.g. Leptobrachium hainanense, IUCN SSC Amphibian Specialist Group (2020a); Suppl. material 4). Of the remaining 174 species, all but two are consumed on a local/national scale. For Lithobates pipiens, only national trade for research purposes is indicated (IUCN SSC Amphibian Specialist Group 2022i); however, according to Herrel and van der Meijden (2014), L. pipiens is also involved in the international trade, without details on the purpose of exports. Of all species of amphibians for which we found data, at least 20 species are potentially involved in international food trade activities. In some species (for example, Limnonectes shompenorum, IUCN SSC Amphibian Specialist Group (2022g)), cross-border trade was assumed, but not substantiated. In other species, the Red List assessment notes the presence of trade, i.e. Rana amurensis (IUCN SSC Amphibian Specialist Group 2020f). Therefore, in Red List assessed species that indicate international trade or questionable cross-border trade, uncertainty is involved in individual assessments (Table 3, Suppl. material 2).

Amongst the 30 species compiled in Table 3 and Suppl. material 2 that are consumed and traded locally, nationally and/or internationally (relevant for the European frogs’ legs trade), uncertainties persist in several species regarding the level of exploitation.

When we submitted the first version of this work in August 2022, several Red List assessments for these species were outdated with 16 species last assessed ≥ 15 years ago, i.e. 11 species assessed in 2004 and five species assessed in 2008, while in 2022/23, several of those species had been re-assessed, leaving six species with outdated assessments (see Table 3, Suppl. material 2).

Current Red List assessments of aforementioned 30 amphibian species, now refer to 24 that have been evaluated “Least Concern”, three “Near Threatened (NT)”, one “Vulnerable (VU)”, one “Endangered (EN)” and one “Critically Endangered (CR) (Table 3, Suppl. material 2).

Population trends of the 30 species indicate 20 species “decreasing”, five species “stable”, two species “increasing” and three species with an “unknown” population trend; it is worth mentioning here in addition that 14 species assessed “Least Concern” have a decreasing population trend (Table 3, Suppl. material 2).

Notable is the fact, that the two large-legged species, i.e. Limnonectes blythii and L. malesianus had been assessed “Near Threatened” in 2004 with decreasing populations in both species (van Dijk and Iskandar 2004b; van Dijk et al. 2004c); however, both were re-evaluated as “Least Concern” in 2021, despite decreasing population trends (IUCN SSC Amphibian Specialist Group 2022d, f).

Outdated assessments are further exacerbated by the fact that the species are regionally overharvested for consumption as well as being involved in the international trade at uncertain levels. However, of all 30 species known to be consumed, 16 species have special mention of harvest that might influence their conservation status. Of these, 13 species (Leptodactylus fallax, Limnonectes blythii, L. kuhlii, L. leporinus, L. macrodon, L. malesianus, Lithobates pipiens, Pelophylax caralitanus, P. kurtmuelleri, P. ridibundus, P. shqipericus, Rana amurensis and R. chensinensis), have either “regional overexploitation-collection” or “harvest leading to declines” explicitly stated in their IUCN Red List assessments. Another three species (Fejervarya cancrivora, Hoplobatrachus rugulosus, Limnonectes microtympanum), have these same parameters as ‘presumed’ within their Red List assessments (Table 3, Suppl. material 2). A detrimental harvest impact is indicated for Rana dybowskii for the medicinal trade (Kuzmin et al. 2004) and in Limnonectes grunniens and Pelophylax bedriagae, harvest for the food trade is considered a significant threat. In Limnonectes ibanorum and L. ingeri, harvest is considered detrimental due to the species’ unfavourable life history traits (Table 3, Suppl. material 2).

Of the 187 species filtered from the IUCN Red List that are collected for either local, national or international consumption (Suppl. material 4), assessments of population trends since 2004 to 2021 reflect an increase in population declines, while the number of species in threat categories also seems to increase over the course of the Red List assessments; however, the species’ to be evaluated as well as the geographical distribution play decisive roles; thus, a threat to species evaluated as “Data Deficient” cannot be ruled out and geographically widespread species assessed as “Least Concern” may be threatened at the national level (cf. Figs 5, 6).

Figure 5.

Red List status of 187 amphibian species globally utilised for consumption that have been assessed in eleven assessment periods between 2004 and 2021. Source: IUCN (2022); cf. Suppl. material 4).

Figure 6.

Population trends assessed in 187 amphibian species, consumed for food, in eleven assessment periods between 2004-2021 (cf. Suppl. material 4). Source: IUCN (2022).

It is noteworthy to mention that 57 species of the respective 187 species, have a decreasing population trend, but categorised as “Least Concern” (see Suppl. material 4).

Assessments of 28 species were re-evaluated in 2022/23, in seven species, the Red List status was changed (see Suppl. material 4). Uncertainties outlined in this review remain unevaluated and a resolution of these for individual species assessments would likely influence the categorisation of the threat status and population trends.

The Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) currently lists 220 amphibian species in their appendices, equating to ca. 2.6% of all amphibian species (8,386 spp.; Frost (2021)) recognised by science. The CITES trade database (https://trade.cites.org/, accessed January 2022, see Suppl. material 3) merely lists seven anuran species that are traded for the purpose of consumption (Table 3). Nonetheless, the majority of species involved in the frogs’ legs trade are not listed in the appendices of CITES (cf. Table 3, Suppl. materials 2, 4): thus, trade in these non-CITES listed species is not documented in the CITES trade database, making information on species volumes traded per annum or analysis of specific trends not possible.

All seven CITES listed anuran species are utilised on a local/national scale and three (i.e. Pelophylax shqipericus, Limnonectes macrodon and Hoplobatrachus tigerinus), are involved in the international frogs’ legs trade (cf. Table 4) as well. All seven species have been evaluated in the IUCN Red List and three other species, (i.e. Conraua goliath, Laotriton laoensis and P. shqipericus) are not listed in the appendices of CITES, but appear in the annexes of the European Wildlife Trade Regulations (EU WTR). All but two of these species have a decreasing population trend and two species were last assessed in 2004 and 2008.

Four species are listed in CITES App. II and one in CITES App. III that are consumed either locally/nationally and/or internationally traded for consumption, while another four species are only listed in the annexes of the EU-WTR (Table 4, Suppl. material 4).

1. Calyptocephalella gayi . – Since 2011, the species is listed on CITES App. III in Chile. In 2012–2016, reported exports of 114 live individuals were recorded at the same time that 550 live individuals were imported. In 2012, 14 live individuals were seized in Japan and the 550 animals were sourced from captivity in Chile and imported by the US and Japan. International trade for the purpose of consumption is not explicitly documented, despite the fact that the species is nationally and internationally involved in the food trade (IUCN SSC Amphibian Specialist Group 2019a).
2. Conraua goliath . – This species is not listed in the appendices of CITES, but in Annex B of the EU-WTR. However, eight transactions 1998–2019 of wild-sourced individuals were documented in the CITES trade database. All exports were from Cameroon, with 19 live individuals commercially exported by Cameroon and 65 individuals claimed as commercial imports by EU importing countries. In 2004, Cameroon exported 199 specimens to the United States for scientific purposes. International trade for the purpose of consumption is not documented despite the species being locally/nationally consumed (IUCN SSC Amphibian Specialist Group 2019b).
3. Euphlyctis hexadactylus . – International trade has been documented since 1985 (date of CITES listing), with India as the major supplying country until 1986, documenting the export of roughly 1215 tonnes of meat, while importing countries documented the import of at least 871.5 tonnes meat (https://trade.cites.org, see Suppl. material 3). In 1986, the United States imported another ~ 80 tonnes meat indicating India as the country of origin.
4. Hoplobatrachus tigerinus . – Exports are documented since 1985 (date of CITES listing) and transactions have been reported until 2019. However, the largest quantities were shipped in 2007. Analysis of trade data of this species is particularly challenging because quantities are misleadingly indicated and non-range States of the species export large quantities, including meat of wild-sourced individuals (for example, from Vietnam and Madagascar, documented in the CITES trade database).
5. Limnonectes macrodon . – This species is not listed in the appendices of CITES, but in Annex D of the EU-WTR. However, a single transaction was documented in the CITES trade database. In 2016, Germany reported the import of two live individuals from Indonesia, sourced from the wild. The species is intensively involved in the local, national and international food trade (IUCN SSC Amphibian Specialist Group 2018). It is remarkable that the Annex D records do not reflect an intense EU import of frogs’ legs officially labelled as “ Limnonectes macrodon”, as noted by Dittrich et al. (2017), since this is almost a certainty.
6. Pelophylax shqipericus . – This species is not listed in the appendices of CITES, but is in Annex D of the EU-WTR since 2009 because there was concern regarding the numbers imported into the EU, with monitoring of this trade warranted and a distinct lack of a rigorous non-detriment finding (https://www.speciesplus.net/species#/taxon_concepts/5193/legal, see Suppl. material 3).
7. Telmatobius culeus . – Commercial trade of the species was suspended in 2017, with listing in CITES Appendix I and EU-WTR Annex A. In the period 2010–2022, the CITES trade database indicates only two transactions: the import of 20 live individuals to Canada and 150 live animals to the UK. In both cases, the animals were destined for zoos and sourced as “farmed” from the USA. According to the IUCN SSC Amphibian Specialist Group (2020g), it is estimated that > 15,000 animals/year are used to prepare frogs’ legs.

The farming and regional/international trade activities involving amphibian species for consumption purposes is associated with numerous risks. Here, we outline these more specifically.

The use and trade of species in their country of origin and whose taxonomic status is uncertain affects at least four species involved in the international frogs’ legs industry as well. Amongst these, three are designated as species complexes (i.e. more than one species under one current scientific name) and species with unresolved taxonomy in IUCN Red List assessments. They are: Fejervarya cancrivora, Hoplobatrachus tigerinus and Limnonectes blythii. There are many other species complexes, wherein the taxonomy is extremely complex and uncertainties are even more fraught with problems. Fejervarya moodiei was described from “Manila”, Luzon Island Philippines and, hence, taxonomic studies should initially be conducted on that population. In another two species (Limnonectes grunniens and L. kuhlii), where the impact of international trade for frogs’ legs has not been explicitly ascertained within their assessments (but is very high), taxonomy remains unresolved. In these species of Limnonectes, both their geographic range and number of cryptic species ‘hiding’ under one scientific name are still unclear (IUCN SSC Amphibian Specialist Group 2020b; IUCN SSC Amphibian Specialist Group 2022e). To what extent populations assigned to L. kuhlii are involved in the international frogs’ legs industry is not indicated in the species’ Red List assessment. Since all but two assessments are from 2004, H. tigerinus in 2008 and L. grunniens in 2019 (Table 3, Suppl. material 2), recent research findings sometimes provide more clarity regarding the unsettled taxonomy of the aforementioned species/taxa.

Of the three “species” that clearly represent complexes of many different species, we highlight what is known here, but reiterate that the dearth of data is staggering, considering that these are the most economically valuable species in terms of the known trade in commercial frogs’ legs.

Fejervarya cancrivora. – An initial molecular analysis, six years after F. cancrivora was evaluated in the IUCN Red List (IUCN SSC Amphibian Specialist Group 2022a), revealed three geographically distinct clades/subclades: one confined to Bangladesh, Thailand and the Philippines; another representing Malaysia and Indonesia (Greater Sundas); and the remaining one from Sulawesi (incl. one population in southern West Java, as a result of human introduction) (Kurniawan et al. 2010). A second study by Kurniawan et al. (2011) examined the species’ morphological traits and crossing experiments through artificial insemination that resulted in three distinct taxa: 1) populations of West Java, peninsular Malaysia and Bangladesh assigned to F. cancrivora, 2) populations from the Philippines and China previously referred to as F. moodiei and 3) a new species endemic to Sulawesi. However, findings of a more recent study delimit F. cancrivora to Thailand, peninsular Malaysia and Indonesia (Sumatra, Kalimantan, western and central Java, Bali), with introduced populations occurring in Papua New Guinea and Guam (Yodthong et al. 2019; and refs therein). According to Dubois and Ohler (2000), F. moodiei was considered a valid species and almost 20 years later, the species was confirmed from mainly coastal areas of South Asia (eastern India, Andaman and Nicobar Isl.,), East Asia (southern China) and Southeast Asia (Vietnam, Thailand, Myanmar, Malaysia and the Philippines [Luzon Isl.]) (Yodthong et al. 2019; and references therein).

Clear taxonomy is the foundation of efficient and sustainable species conservation and so is the naming of the species or parts thereof that are to be traded. Examination of 209 frozen frogs’ legs sold in supermarkets in France listed exclusively as Limnonectes [Rana] macrodon (based on product labelling), revealed that almost all (206 of the 209 or 98.6%) were in fact legs of F. cancrivora and only 2 (0.96%) could be attributed to L. macrodon, while one sample was revealed to be F. moodiei (Ohler and Nicolas 2017). Such forensic studies clearly highlight the importance of competent species identification, especially when it comes to evaluating current use in terms of sustainability, as the lack of such information precludes accurate monitoring of trade as a consequence of misidentification. Many more members of both the Dicroglossidae and Ranidae families are commercially involved in the frogs’ legs industry and their taxonomic status remains blurry at best.

Hoplobatrachus tigerinus. – In their Red List assessment, Padhye et al. (2008) indicate H. tigerinus reflects a species complex including an unknown number of morphologically very similar (cryptic) species. This was confirmed by Hasan et al. (2012). Most recent research identified populations of H. tigerinus from Pakistan and Bangladesh as genetically identical to those from Nepal (Khatiwada et al. 2017), but genetically different from Indian populations (Akram et al. 2021). Clearly, this is a complex issue with much more clarity needed before the trade becomes sustainable.

Limnonectes kuhlii. – The taxonomic status of L. kuhlii associated with the species’ currently known distribution range has been described as particularly uncertain within the Red List assessment (IUCN SSC Amphibian Specialist Group 2022e) and many more new taxa have been assumed with some revealing range-restricted distributions. Following genetic research, this complex now includes a minimum of 22 “distinct evolutionary lineages” (McCleod 2010). Again, the real biological entities that are involved in the commercial frogs’ legs trade clearly are not well understood, much less studied to the degree to which we can provide realistic plans or guidelines for sustainable trade.

Sixteen of the 30 anuran species listed in Table 3 and Suppl. material 2, (i.e. Fejervarya cancrivora, Limnonectes blythii, L. grunniens, L. kuhlii, L. leporinus, L. macrodon, L. malesianus, L. microtympanum, Lithobates pipiens, Pelophylax bedriagae, P. caralitanus, P. kurtmuelleri, P. ridibundus, P. shqipericus, Rana amurensis and R. chensinensis) have commercial (regional) overharvest/overexploitation as a significant/main threat (both for food) indicated as assumed or known threat in their respective Red List Assessments. Species that were previously intensively exploited were not included (i.e. Hoplobatrachus tigerinus and Leptodactylus fallax), as former legal trade was banned in the mid-1990s (Padhye et al. 2008) and other utilisation has been banned since the 2000s (IUCN SSC Amphibian Specialist Group 2017). It is important to note that the conservation status of most species involved in the food trade (Table 3; Suppl. material 2) is not up to date (53% or 16 species last assessed 2004–08) and re-assessments of some species might indicate overexploitation, adding more species where commercial exploitation for international consumption is considered unsustainable.

Prior to export for international trade, a considerable number of live animals die on arrival to the processing facilities. For Indian exports, this loss has been estimated at 10–20%, in Indonesia it is 40–50% because quality is not sufficient for export and some frogs are killed prior to being exported (Niekisch 1986 and references therein). Information on pre-export mortality rates in countries of origin were not easy to obtain within the scope of our study. These figures are also relevant when it comes to evaluating the ecological impact of harvest and more clear understanding of how these losses could be lowered would benefit both the people involved in the trade and the frog populations.

Initial reports on the sustainability of this trade were published more than 20 years ago; however, large-scale ecological studies to assess offtake rates and their sustainability appear severely lacking. Here, we highlight studies that indicate amphibian declines associated with harvest for the food trade both regionally and internationally. Historically, overharvest was detected in Californian populations of Rana aurora draytonii (Jennings and Hayes 1985). In Florida, harvest regimes of Lithobates [Rana] grylio affect population structure and survival rates (Ugarte 2004). The increasingly intense regional harvest of frogs in West Africa, particularly in Nigeria where trade has moved across borders (e.g. Benin), clearly demonstrates overexploited species and populations (Mohneke 2011). The harvest of populations of Quasipaa spinosa in Hong Kong is also detrimental to populations in the long-term (Chan et al. 2014). Below, we highlight case studies that report on overexploitation of species/populations from Indonesia and Turkey involved in the international commercial trade.

The high uncertainty of the assumed number of individual frogs within total imports throughout the study period impressively illustrates the opacity of the trade. Actual harvest numbers imported into the EU for annual consumption remain unknown and very difficult to quantify. This is undoubtedly due to the fact that they are non-CITES species and, thus, international trade data (species/volumes) remain undocumented. Listing species in the appendices of CITES is justified when international trade poses a severe threat to the conservation status of a species. The scientific authority of a CITES member state must review the harvest/export for Appendix II in terms of compatible offtake numbers/quotas in order to maintain the species’ ecological function in its native habitat (https://cites.org/eng/disc/text.php#IV, accessed May 2022, see Suppl. material 3). Complete transparency of annual quotas and the quantification of numbers of individual frogs per kilo must be ensured if a kilo value is to represent the number of affected individuals. It remains unclear for what reasons the calculations of the number of individuals per kilo have been reduced by seven animals as of 2018 (Table 2) and we remain sceptical of these numbers.

Anurans involved in the international frogs’ legs trade are all r-strategists, which means that they have large numbers of offspring, a rapid developmental rate and a high reproductive output. This also makes these species more amenable to regular (monitored) harvest while remaining viable. However, r-strategists also define themselves in having highly variable population sizes over time and mortalities may be density-independent or even catastrophic (Pianka 1970). Despite relatively high individual densities of some species in agroecosystems, regular removal of thousands of individuals still raises questions about the extent that the ecosystems can compensate for this intervention. For example, negative ecological shifts may have already occurred (for example, can ecologically more flexible species outcompete more specialised species and how have populations of insect pests been affected by fluctuations in frog populations?). There is also no doubt that trophic interactions in certain agro-ecosystems such as rice fields are very complex and we still do not have a grasp of the main drivers of the complexities. For example, type of cultivation and human impact can have severe implications on biodiversity. Abrupt regular removal of rice plants in a wet paddy, for instance, results in a considerable sudden loss of energy for the entire biotic community (cf. Bambaradeniya et al. (2004)). A decline in pond frogs (Pelophylax nigormaculatus) in rice field-dominated landscapes in Japan has been noted as a result of the modernisation of drainage systems which also led to the decline of the grey-faced buzzard (Butastur indicus) (Fujita et al. 2015). It is clear that human impacts on nutrient supply and food web structure have strong and interdependent effects on biodiversity and ecosystem functioning and it is, therefore, essential to monitor/these both (see Worm et al. (2002)).

These considerations may, however, be too complex to be actively explored within the framework of the EU. We highlighted that there are many internationally traded species/species groups with sales in the EU where unsustainable trade has been detected (cf. Symes et al. (2018)), that could be regulated more easily. Governmental priorities within transnational cooperation projects should develop common methodological approaches that include genetics (species identification and origin) and biosecurity measures to prevent the spread of disease.

However, in the context of amphibians that are, for example, imported live into the EU for the exotic pet trade industry, amongst which many are traded that are also known to be infected with Bd/Bsal, (see Wombwell et al. (2016); Nguyen et al. (2017); Fitzpatrick et al. (2018)), even here, biosecurity measures prior to the import into the EU (incl. non-EU- European States) have not been implemented to prevent cross-infections, despite the fact that Bd was listed as a notifiable disease by the World Organisation for Animal Health (OIE) in 2008 (Schloegel et al. 2010) and Bsal in 2017 (https://www.oie.int/app/uploads/2021/03/a-bsal-disease-card.pdf, see Suppl. material 3).

Required data for the IUCN Red List are crucial for assessing the conservation status of species. In Red List assessments, trade in a species can either: (1) be mentioned at the national/ international level, (2) go unmentioned (despite the fact that trade occurs) or (3) if mentioned, in some cases be designated as an acute threat to a species/population. In such cases, it is particularly problematic when Red List assessments are up to 19 years old (Table 3, Suppl. materials 2, 4) and for species utilised domestically or traded internationally where overexploitation was already identified in 2004, but impact on the local populations has not been well assessed (e.g. Limnonectes blythii, L. kuhlii or L. malesianus; see Table 3). Our query in August 2022 retrieved 187 species (see Methods); a query in January 2023 resulted in 219 species; however, the selected query parameters do not cover all relevant species, for example, Fejervarya moodiei (cf. Table 3).

Several Pelophylax, Limnonectes and Fejervarya spp. are morphologically very difficult to distinguish and many taxa are taxonomically treated as cryptic species complexes (see Bickford et al. (2007)) within their genera (Kurniawan et al. 2011; Dufresnes et al. 2018; Yodthong et al. 2019). Therefore, challenges of quantifying actual harvest of each species are substantial if these taxa are harvested in the hundreds of thousands to millions of individuals per year. Accurate identification of species is the foundation for any management plan and trade and conservation need to go hand in hand. Disregard of this basic knowledge and trading activity can cause fundamental damage to the species and, in the worst case, to respective ecosystems (Estes et al. 2011). Unfortunately, it is precisely this taxonomic uncertainty that is exploited by companies, for example, as done in Turkey, labelling frogs as the hybrid Pelophylax esculentus which does not occur in Turkey but does in other parts of Europe (Çiçek et al. 2021). Evidence provided by genetic methods could reveal incorrect labelling in Indonesian exports of frozen frog’s legs destined for European markets with packages indicating Limnonectes [Rana] macrodon rather than as Fejervarya cancrivora, but rigorous assessments of accuracy of species identification have not been conducted (Dittrich et al. 2017; Ohler and Nicolas 2017). In 2001, Veith and colleagues could separate F. iskandari as a valid species from the F. limnocharis-complex through allozyme data. Another clear example is F. iskandari (restricted to the island of Java) which was previously traded undetected within the F. limnocharis-complex (Kusrini 2005) and could be negatively impacted by overharvest. Apart from these examples of harvested taxa included in species-complexes with uncertainty in their taxonomic status (e.g. Holsbeek et al. (2008); McLeod (2010); McLeod et al. (2011); Dehling and Dehling (2017); Yodthong et al. (2019); Stuart et al. (2020)), introduction of exotic species that interbreed with closely related species or crossbreeding incidences of farm escapees into other ecosystems (Yu et al. 2015), may lead to a replacement of formerly native species (cf. Leuenberger et al. (2014)). In addition to these concerns is the potential for an invasive species (e.g. Lithobates catesbeianus) to become a driver of ecological trophic cascades in naive ecosystems (e.g. Gobel et al. 2019). Such issues are well known from other taxa, yet the lack of monitoring and the number of cryptic species underscores the under-appreciated risks associated with hybridisation of these as yet unrecognised frog species. Species identification of skinned or frozen frogs’ legs is impossible without genetic techniques, thus mislabelling may not have been strategic, but an indication that processors and exporters in Indonesia are not trained in frog species identification. This knowledge was not considered a prerequisite for the export of frogs’ legs and, as there are no strict checks, the trade of potentially misidentified species has been allowed to continue. Of more concern, it may also be that maintaining consistent supplies would not be possible if adequate scrutiny of what is in the trade, where it is from and how availability fluctuates, are taken into account. In fact, it must be clearly emphasised that the prerequisite, “we only eat/trade what we know”, has not yet been met and relevant stakeholders (including government agencies) have not made an adequate effort to address this issue.

Sustainable international trade can only be ensured if the use and movement of species within national borders is managed in such a way that species or populations maintain their viability and do not show shifts in physical traits due to bias in selection of key traits (cf. Leader-Williams 2002). In fact, differences in body size in intensely harvested populations of Lithobates [Rana] grylio are probably due to selective harvesting pressure on larger size classes (Ugarte 2004). Kusrini (2005) found that body sizes of captured adults are smaller than those of the same species in other un-harvested regions and capturing larger adults may lead to lower recruitment rates. Similarly, the pronounced sexual dimorphism in species attractive to hunters (e.g. F. cancrivora and L. macrodon), leads to reduction in the number of those larger individuals. As females are usually larger in anurans (Duellman and Trueb 1994), the collection of breeding females, in particular, significantly reduces the potential reproductive effort and, thus, will detrimentally impact populations. For the maintenance of viable populations, sexual dimorphism traits should, therefore, be considered in harvest regimes to sustain populations. According to Kusrini (2005), one important criterion for monitoring is the recording of body size. These worrying, but prescient data from 17 years ago do not seem to have been properly considered until now and viability of harvested frog populations has largely been overlooked.

In this context, governments are called upon to use resources in an adaptive and sustainable manner. Furthermore, EU commitments to Environmental Impact Assessments (EIAs) of imported wildlife mean that the EU is obligated to monitor what is in trade as well as the impact it is likely to have on source populations. As soon as the species triggers international demand and sales, importing countries are equally held accountable to take responsibility, whereby relevant stakeholders must ensure that their consumption of exotic species does not lead to population declines. Clearly, this will entail other anthropogenically induced threats affecting these species/populations (e.g. Chen et al. (2019)). It is worrying to note that there are very few studies reviewing current trade in terms of sustainability and the little information that is published, implies very strongly that current harvest/trade is unsustainable. For example, populations of Pelophylax caralitanus are still locally widespread in Turkey, but the species is considered endangered (Ǒz et al. 2009), not only because of habitat loss, but also because of local overexploitation for trade with the EU (Erismis 2018; Çiçek et al. 2021). Further, overharvest of Pelophylax shqipericus has been noted in the species’ Red List assessment (IUCN SSC Amphibian Specialist Group 2020e) and the unsustainable trade of this species has been highlighted (Gratwicke et al. 2010). However, populations of P. shqipericus in Albania (core distribution of the species) have not yet been considered within a conservation management plan (Eco Albania 2019).

Numerous examples of overexploited species assessed in the IUCN Red List assessments are detailed (see Table 3, Suppl. materials 2, 4) and examples of unsustainable trade at the regional level (for example, in western Africa and that of species and species complexes in Southeast Asia) have also been presented. However, there is a severe shortage of established field studies (cf. Auliya et al. (2016); Morton et al. (2021)) over longer periods of time to provide not only snapshots of single localities, populations and their harvest status, but also long-term studies (for example, use of pesticides and potential residues on populations in trade, impact of local population declines, if populations can maintain their role as pest control etc.).

According to Raghavendra et al. (2008), comprehensive ecological field studies in India investigating the function of anuran communities and their control of pests such as mosquitoes are still in their infancy. Local knowledge in West Java (Indonesia) reveals that at least Fejervarya limnocharis is perceived in functioning as pest control (Partasasmita et al. 2016).

A two-year field study in the Philippines compared prey items of the native Luzon wartfrog (Fejervarya vittigera) with that of the introduced cane toad (Rhinella marina) to determine the proportion of rice pests in their diets and which of the two species was more efficient feeding on rice pests. It turned out that the proportion of pests eaten by F. vittigera was significantly larger than that of R. marina, which mainly preyed on beneficial arthropods in the rice-ecosystems. The authors conclude that adult F. vittigera may provide effective pest control services and suggest protecting and promoting F. vittigera populations (as opposed to reducing R. marina populations) to minimise the use of insecticides (Shuman-Goodier et al. 2019).

Due to problems of sustainability caused by the removal of species from their ecosystems (see Table 3), various authors suggest a focus on commercial frog farming (e.g. Nguyen (2014); Şereflişan and Alkaya 2016; Ribeiro et al. (2019)). Indeed, commercialisation of frog farming appeared to be the way forward for a promising industry in many countries (first attempts at breeding Lithobates catesbeianus in the US and Canada are dated before 1900), but continuing efforts to implement these plans have proved less successful (Helfrich et al. 2009; Dodd and Jennings 2021). Such ventures have been discouraged since the 1930s and many problems (for example, live food and water quality availability, risk of spreading disease, slow mass increase or growth and economic start-up constraints) were known to the early proponents of such ventures. However, because investments are relatively low and profits can be many times higher, this branch of business creation continues.

Globally, Lithobates catesbeianus is the most widespread species involved in farming operations and has been introduced for the purpose of commercial farming into more than 40 countries (FAO 2021).

In other parts of the world, initiatives to commercialise frog farming are also being publicised as a result of increased demand. For example, under EU funding, the CaPFish Capture and Aquaculture programmes were launched to promote aquaculture in 10 provinces of Cambodia, primarily to promote food security in line with national government plans for fisheries development. Specifically, the Minister of Agriculture, Forestry and Fisheries, “Veng Sakhon”, encouraged farmers to raise frogs due to an increased market demand (https://en.khmerpostasia.com/2020/10/16/frog-farming-encouraged-as-market-demand-rising/, accessed, June 2022, see Suppl. material 3). However, this programme is explicitly designed for national needs, not international export.

Likewise, in Thailand, establishment of commercial frog breeding facilities has been described and limited for national consumption (Pariyanonth and Daorerk 1994).

A major problem underlying establishment of commercial frog farming facilities is that there are no international standards or hygiene guidelines (see Dittrich et al. (2017)). In some of EU’s major supplying countries, i.e. Vietnam, frog farms remain being insufficiently controlled (Nguyen 2014; Nguyen and Tran 2021) implying that no health controls are imposed on farms and processing into frogs’ legs, as well as testing for disease. As a result, the risk of international trade spreading diseases, such as ranavirus and Bd into naive amphibian populations, is ever-present (cf. Gratwicke et al. (2010); Gilbert et al. (2013)). However, unfavourable conditions are present, for example, the lack of appropriate management measures, resulting in the (unintentional) release of disease-infected L. catesbeianus into the environment of supplier countries (cf. Ribeiro et al. (2019)). Species escaping from breeding farms may also hybridise with congeners and here the problem of genetic pollution needs to be addressed.

An additional complicating factor for international control is that species harvested for frogs’ legs are exclusively non-CITES species, implying that there is no documentation across international borders.

Modern innovative scientific methods are required to ensure a fully transparent, legal, traceable and sustainable trade. We will need to implement scientific methodologies to distinguish farmed vs. wild individuals (cf. Dittrich et al. (2017)) and to obtain sufficient data on all source populations to ensure that harvest levels fall below annual population replacement levels. Basically, taxonomic uncertainties need to be clarified and the formation of specific research groups (for example, taxonomists, field ecologists, experts of current legal frameworks etc.) is highly recommended.

To prevent the spread of infectious diseases, biosecurity measures need to be established at distinct points along the trade chain. Interestingly, such measures were already proposed at the 37^th Standing Committee of the Convention on the Conservation of European Wildlife and Natural Habitats, in December 2017 (https://rm.coe.int/recommendation-on-biosafety-measures-for-the-prevention-of-the-spread-/168075a4b0, accessed May 2022, see Suppl. material 3), but never implemented. Therein, recommendation No. 197 refers to “biosafety measures for the prevention of the spread of amphibian and reptile species diseases”. This document lists 10 recommendations for contracting parties, none of which includes information on species traded either alive or processed for the frogs’ legs trade. The majority of recommendations encourage support for increased research. However, recommendation 5, “Using the most appropriate legal framework and, at the earliest opportunity, implement immediate restrictions on the amphibian and reptile species trade when an emerging pathogen spread with significant impact on wild populations has been identified until necessary preventative and management measures are designed, based on evidence, throughout the entire commercial chain”, does not reflect an expansion of the regulatory framework, but describes a direct suspension of trade in an infected species. With regard to the prevention and spread of known diseases identified by OIE (such as Bd), we reference a document from 2015 by the Standing Committee to the Convention on the Conservation of European Wildlife and Natural Habitats on the Recommendation on the Prevention and Control of the Bsal fungus (https://rm.coe.int/1680746acf, accessed May 2022, see Suppl. material 3). The implementation of these recommendations, however, cannot be verified. The need for supervision of hygiene and veterinary inspections for edible frogs (also those farmed and are non-native) in the Asian region has been indicated (Grano 2020; Borzée et al. 2021), given the tight links observed between market locations and detection of Bd in wild amphibian populations.

Hardouin (1997) stated that authorities in countries that import frogs’ legs should be encouraged to regulate international trade more closely by banning products that cannot be sourced from farms where they are subject to official controls. He further notes that Europe cannot ignore risk of wild harvests that may lead to declines in local frog populations as a result of overexploitation. We also recommend the listing of some if not all species in trade on CITES App. II. International trade should be regulated for those species that are already documented in an IUCN Red List threat category and those for which there is published evidence that trade has depleted local or regional populations. Taxa in species complexes whose morphological differentiation is not readily possible or are processed only as frogs’ legs are particularly vulnerable, so standardised use of molecular approaches to verify and monitor trade would be particularly useful.

Results outlined in this review provide strong clear recommendations for both source and consuming countries. Promptly counteracting abuses in the international trade of frogs’ legs by adapting existing legislation and applying the precautionary principle to prevent irreversible damage to populations or species will help to promote the sustainability of the trade in the long-term. Recommendations for source and consuming countries are listed separately below.

• conduct field surveys at comparative study areas to estimate size and trends of wild frog populations and of the impact of harvest for both national consumption and international trade.
• validate species identity through centralised authorities to check and certify trade exports through the use of genetic tools.
• include analyses of trade data and standardise documentation of volumes (number of individuals must be considered, not an estimate of the number of individuals by means of weight).
• establish long-term field studies in selected areas (where regular harvest takes place) to assess biotic communities in relation to the application of pesticides.
• make non-detriment findings (NDFs) a result of CITES listings at regular time intervals.
• examine the domestic/national use of frogs’ legs versus exports to decipher the complexity of this resource use and improve equity and fairness within each source country.
• study mortality rates of frogs in transport and processing prior to export. When identifiable loopholes exist, source countries should make every effort to minimise mortality and economic loss.
• accurately and regularly verify harvest rates, including both local as well as harvest for international trade. As highlighted earlier, it has been estimated that offtakes of edible frogs on a national level can be seven times as much as that of annual exports (Kusrini 2005).
• establish conservative, but reasonable harvest and export quotas, based on high quality data for targeted species/populations and taking into account other threats that affect species/populations.
• ban harvest during the mating season. Specific management measures have been highlighted for the harvest of Pelophylax spp. in Turkey and claim, “that further harvest restrictions are essential for the sustainability of Anatolian water frog populations” (Çiçek et al. 2021).
• evaluate and implement adaptive management measures for all harvested species, i.e. the ban of certain size classes for a given period/season as a default to help ensure sustainability.
• define and implement stricter regulations for farming operations to ensure closed systems, prevent re-stocking from the wild, release of farmed animals back into the environment, as well as avoiding farming of non-native species when possible.
• register and monitor all export companies and their suppliers and require that exporters identify processed frog products by DNA analysis.

Consumer countries have the obligation to take appropriate responsibility for the consumption of a resource. Accordingly, it would be obligatory to transparently inform relevant societies on which information basis trade is permitted.

• implement a centralised database to document all imports of all wildlife and list species and quantities in the Annexes of the EU Wildlife Trade Regulation, using the USFWS-LEMIS Database (2023) as a model.
• list all species in trade in CITES to regulate international trade and enforce restrictions.
• implement NDFs for the import of species from the wild, regardless of CITES status.
• provide captive breeding guarantees for species claimed to be of captive origin.
• push for improved standards (based on revised guidelines), such as import bans on wild harvested species that have been evaluated in one of the IUCN Red List threat categories.
• impose trade suspensions if trade data are not provided in full transparency.
• check all imports for pesticides and other pollutant residues.
• assist range states in conducting surveys of wild frog populations and to create a biobank with references samples from species/populations of major harvest regions to cross-check genetic identities of shipments imported.
• conduct random DNA analysis of frogs’ legs shipments to determine if shipment labelling is correct and ban imports for persistent mislabelling.
• allow only positively identified, skinned, processed and frozen frogs’ legs to be imported to avoid the introduction and spreading of diseases and invasive species.
• rigorously catalogue all imported species with standards parallel to those implemented under the USFWS-LEMIS Database (2023).
• improve regional monitoring schemes with joint-efforts between stakeholders and governments to bolster the sustainability of the trade along multiple facets.