Summary

1. Around 1.2 to 1.9 trillion fish fry have been produced by artificial propagation (breeding) annually in China, and it seems those farmed for direct human consumption and pre-harvest mortality can only account for at most around 460 billion of them (more).
2. I estimate that 170 billion to 700 billion animals (my 80% credible interval) — probably almost all artificially propagated fish — are fed live to mandarin fish (or Chinese perch), Siniperca chuatsi, in China annually, with 9 to 55 billion of them (my 80% credible interval) alive at any time (more, Guesstimate model).
3. By contrast, the number of farmed fish produced globally for direct human consumption is around 111 billion per year, and 103 billion alive at a time, with another 35 to 150 billion raised and stocked per year (Šimčikas, 2020).
4. It’s unclear how bad their deaths are as live feed to mandarin fish, but I’d guess they die by suffocation, digestion (stomach acid, enzymes), or mechanical injury, e.g. crushing, after being swallowed live, and probably a common way for aquatic animals to die by predation by fish in the wild (more).
5. It’s unclear if there’s much we can do to help these feed fish. There’s been some progress in substituting artificial diets (including dead animal protein) for live feed for mandarin fish, but this has been a topic of research for over 20 years. Human diet interventions would need to be fairly targeted to be effective. I give a shallow overview of some possible interventions and encourage further investigation (more).

Acknowledgements

Thanks to Vasco Grilo, Saulius Šimčikas and Max Carpendale for feedback. All errors are my own.

Fish fry production in China

One of the early developmental stages of fish is the fry stage (Juvenile fish - Wikipedia). Šimčikas (2019, EA Forum), in his appendix section, raised the question of why hundreds of billions of fish fry were produced artificially (via artificial breeding, i.e. artificial propagation) in China in each of multiple years, yet only “28-92 billion” farmed fish were produced in China in 2015, “according to an estimate from Fishcount”. He found that if the apparent discrepancy were due to pre-slaughter mortality, then this would indicate unusually low survival rates. He left open the reason for the apparent discrepancy and recommended further investigation.

Before going into potential explanations for the discrepancy, I share some more recent numbers for the artificial propagation of fish: 1.9143 trillion fish fry in China in 2013 (Li & Xia, 2018) and 1.252 trillion freshwater fry and 167 million marine fish fry in China in 2019 (Hu et al., 2021). The 2019 numbers seem substantially lower than in 2013, so the trend may have reversed, one of these numbers is inaccurate, there’s high variance across years or one of the years was unusual.

Li and Xia (2018) also plot the trend over time up to 2013, along with total freshwater aquaculture:

28 to 92 billion farmed fish produced in China in 2015 (Fishcount) from 1 to 2 trillion artificially propagated fish fry, would suggest a pre-slaughter/pre-harvest survival rate of 1.4% to 9.2% (from the fry stage on). Survival rates are typically at least 20% for the most commonly farmed species, including carps and tilapias (Animal Charity Evaluators, 2020, table 4), for which China accounts for most production. And Šimčikas (2019, EA Forum) notes:

Since hatchery-produced juveniles are already past the first stage of their lives in which pre-slaughter mortality is the highest, mortality during the grow-out period shouldn’t be that high.

Under fairly generous assumptions for an explanation based on pre-harvest mortality, using the upper bound of 92 billion farmed fish produced from Fishcount and a relatively low 20% survival rate to harvest would only account for 460 billion fish fry. On the other hand, it seems over a trillion have been produced per year since 2011. Farmed fish produced for direct human consumption would then seem to fail to account for the rest, over 500 billion fish fry produced in China per year. Using more central estimates of 60 billion farmed fish produced in China per year and an average pre-harvest/pre-slaughter survival rate of 40%, this would only account for 150 billion fish fry in China per year. The apparent discrepancy would then be over a trillion fish fry per year.

There are multiple possibilities to account for the apparently missing >500 billion fish fry per year, and it could turn out to be a combination of them:

1. The survival rates in China are particularly low.
2. Some of the numbers or estimates may be inaccurate (raised by Šimčikas (2019, EA Forum).
3. They are used for restocking wild fisheries (Animal Charity Evaluators, 2020). Šimčikas (2019, EA Forum) cited around 20–40 billion fish stocked per year by China, according to plans by the Chinese government. This could account for 50–100 billion fry, assuming 40% survive to stocking.
4. They are used as bait to catch wild fish (Animal Charity Evaluators, 2020).
5. They are released into the wild to feed wild fish that will eventually be caught.
6. They are used as feed for other farmed animals, especially other farmed aquatic animals.
7. They are used to improve or maintain water quality for higher-value farmed species in polyculture/co-culture (Wang et al., 2016, He et al., 2018, Yao & Li, 2018).

Fish raised to become feed for farmed animals

In particular, within the use as feed, a substantial share of artificially propagated fish fry seems to be accounted for as live feed for farmed mandarin fish (or Chinese perch), Siniperca chuatsi (not the other fishes also called mandarin fish). In this Guesstimate model, I estimate the number of live animals fed to mandarin fish annually in China to be around 170 billion to 700 billion, as my 80% credible interval (the Bayesian version of a credence interval). As discussed in the next section, the feed is generally reported to be live fish and in particular fish species — mostly carps — that are commonly farmed. Li and Xia (2018) also claim “Almost all prey for mandarin fish is provided through artificial propagation”, and FAO indicates independent culture (cultured separately from the mandarin fish) and concurrent culture (cultured with the mandarin fish) as the most common approaches to live food production. And the number of live fed prey also seems too large to be reflected in wild capture statistics, estimated at 1.1 to 2.2 trillion globally per year and 130 to 220 billion annually by China (Mood and Brooke, 2024).

Feed for mandarin fish may not count for the rest of the apparent discrepancy, but it seems likely to account for a decent share of it.

The mandarin fish is only one species of farmed fish fed live animals and artificially propagated fish in particular. There may be others, whether or not fed exclusively live feed. Like for mandarin fish according to my Guesstimate model, based on decreasing daily consumption as a share of body weight (Yao & Li, 2018, table 3.7.2), a large share of the number of live feed animals may be eaten in early life stages of the animals eating them. Mandarin fish only account for 0.5% of all animal aquaculture in China,^[1] and I estimate that mandarin fish eat over half of the animals (by number) mandarin fish will eat live in the first 20–25 days of mandarin fish’s lives (accounting for mortality), so if fish fry are fed to other farmed animals during the first few weeks of life, this could amount to many individuals.

Indeed, the use of live feed seems common in the diets of shrimp larvae and fish larvae, fry and fingerlings, but (globally) live animal feeds seem to generally be small zooplankton like brine shrimp (Artemia) and rotifers, not fish (Conceição et al., 2010, Sales, 2011, Das et al., 2014, Santosh et al., 2023, Xavier et al., 2023, Kandathil Radhakrishnan et al., 2020, Barad et al., 2017, Chakraborty & Priyanka Halder Mallick, 2023, Nielsen et al., 2017). Around 500 trillion brine shrimp nauplii seem to be used as live feed per year, based on estimates by Aaron Boddy from Shrimp Welfare Project (St. Jules, 2024).

However, I couldn’t find much online indicating substantial fish use as live feed or artificial propagation of fish for feed for any other species. For China, Newton et al. (2021) only mention mandarin fish and swamp eels as being fed fish, and for swamp eels, the use of minced silver carps, not live animals.^[2] Multiple large language models I checked said various other species are fed live fish, but I couldn’t verify any of them.

Wang et al. (2016) describe common freshwater aquaculture practices in China, including the stocking of silver carp (Hypophthalmichthys molitrix Valenciennes 1844) and/or big head carp (Aristichthys nobilis Richardson 1845) to maintain water quality for the culture of higher value species, like the Chinese mitten crab (Eriocheir sinensis), turtles and mandarin fish. They also wrote that in the surveyed Chinese mitten crab farms, when harvesting, the remaining carps were normally “frozen and used as trash fish for the next rearing cycle”.^[3] He et al. (2018) also indicate polyculture (also called co-culture) of soft-shelled turtles with carps, other fish including the mandarin carp, shrimps and/or the Chinese mitten crab is common practice, although they didn’t say what happens to the fish upon harvest of the higher value species.^[4] In particular, “silver carp and bighead carp clean water through filtering plankton” (He et al., 2018). Yao and Li (2018) also indicate co-culture of mandarin fish with other fish and the Chinese mitten crab is common, where “mandarin fish is not the main culture species, but plays a major role in controlling abundance of low‐valued small fish which may compete for feed with the main farming species.”

It’s possible then that many fish fry in China are produced primarily to maintain water quality for high-value species other than mandarin fish. Some may be used with or without mandarin fish. I don’t attempt to estimate their numbers here.

Some others may be raised with mandarin fish, but would be used for water quality even if the co-cultured mandarin fish were not also farmed and sold. Still, separating intensive pond culture of mandarin fish from the co-culture of mandarin fish with other species, Yao and Li (2018), wrote “Intensive pond culture is the dominant practice of mandarin fish culture in China due to the availability of ponds, a short culture period, low infrastructural investment, and high economic benefits compared to cage and lake practices.” So, it seems more likely than not that most fish eaten by mandarin fish are used primarily just for the aquaculture of mandarin fish, not for water quality.

Mandarin fish have had prices “4-10 times higher than the traditional carp” and “ten times greater than that of silver carp”, with 40-45% of the cost of production from live feed (Kuanhong/FAO, 2009). In September 2023 across select locations across China, market prices were (FAO, 2023):

Such differences in prices seem like they could justify the use of artificially propagated carps and tilapia just as feed for mandarin fish, ignoring the benefits of water quality control for other higher-value species. However, carp used as feed are not generally fed to reach market size for direct human consumption. The price of carp fingerlings (older than fry, but fry, fingerlings and older fish will also be fed to mandarin fish) averaged 6.76 to 13.5 USD/1000 fingerlings in China (Kumar et al., 2008, table 3), and fingerlings weigh around 0.25 grams up to around 120 grams each (FAO, table 3; Peteri et al., 1992, table 6), although I’d guess the 6.76 to 13.5 USD/1000 figure reflects fish around 0.25 to 10 grams each. This could be around 2 to 10 USD/kg of fingerlings.^[5]

On the other hand, fishmeal has been around 1.3–1.9 USD/kg from April 2014 to February 2024 (IndexMundi, OECD/FAO, 2023, figure 8.1), but fishmeal is dried (and defatted) fish, and the value of the fresh fish from which fishmeal is produced would be around 3–5 times less. It therefore wouldn’t make sense to farm carps or tilapia to typical market sizes just to render and use or sell as fishmeal, because better prices would be obtainable for live or fresh market size fish.

Mandarin fish farming

In this section, I give an overview of what mandarin fish eat, and the scale of mandarin fish farming and feed for mandarin fish.

Mandarin fish diets

Mandarin fish (or Chinese perch), Siniperca chuatsi, is a piscivorous fish farmed (almost) exclusively in China (Tower, 2010, FAO FishStat). Multiple sources claim they feed exclusively on live animals, usually fish when farmed. About their diets, Yao and Li (2018) wrote:

Mandarin fish have unusual feeding habits. The fish only eat live fish and shrimps, and do not consume dead prey or artificial diets during all lifecycle stages (Chiang 1959; Li et al. 2014a; Yao and Liang 2015).

Mo et al. (2019) say essentially the same. Li and Xia (2018) wrote “Almost all prey for mandarin fish is provided through artificial propagation”, and single out mud carp as the favourite feed fish, although others are reported elsewhere, e.g. FAO:

Common live foods for mandarin fish include mud carp (Cirrhinus molitorella), Wuchang fish (also called Chinese bream, Megalobrama amblycephala), silver carp (Hypophthalmichthys molitrix), bighead carp (H. nobilis), grass carp (Ctenopharyngodon idellus), crucian carp (Carassius carassius), common carp (Cyprinus carpio), stone moroko (Pseudorasbora parva) and other wild and trash fish. Wuchang fish fry is preferred at the start of food intake, then feeding bighead and silver carp follows. When body length reaches 25 cm, common and crucian carps are fed.

And “silver carp, bighead, grass carp, Wuchang fish or tilapia fry” (Kuanhong/FAO, 2009).

A FAO species profile for mandarin fish (a and b) also says that if a mandarin fish consumes non-live food, the food is regurgitated. However, the claim is unsourced, and I could not verify it.

More recently, Xu et al. (2024) wrote:

Consequently, farming Chinese perch using an artificial diet has become a viable solution to the challenges associated with live fish feeding and has been implemented in several aquaculture farms in Chinese provinces, including Jiangxi, Hubei, and Guangdong [3,5]. However, a critical issue with this farming method is that the technology for the domestication of Chinese perch to eat artificial diet is not yet mature. If the domestication process fails, the Chinese perch will refuse to eat the artificial diet.

It seems unlikely that artificial diets currently account for much of what mandarin fish are fed in China.

Scale of mandarin fish farming and feed

In 2021, 373,954 tonnes of mandarin fish were produced in China (FAO, FishStat), with a generally increasing trend over decades:

Li and Xia (2018) plot mandarin fish production and the number of artificially propagated fry together up to 2013:

In this Guesstimate model, assuming (almost) all feed is live feed, I estimate the number of live animals fed to mandarin fish annually in China to be around 170 billion to 700 billion, as my 80% credible interval (the Bayesian version of a confidence interval). This may not necessarily just be artificially propagated fish; some may be wild-caught and others may be invertebrates, but it’s probably mostly artificially propagated fish. However, mandarin fish are also often cultured to control the populations of low-value fish that end up in polyculture ponds and would otherwise eat the food intended for primary higher-value cultured species (Yao & Li, 2018, Wang et al., 2016).

I further estimate in the same Guesstimate model the number of animals alive at a time that will be fed (or made available as feed) to mandarin fish to be 9 to 55 billion as my 80% credible interval, based on an estimated average age of death of 10.5 to 50.8 days (my 80% credible interval).

For these numbers, I’ve attempted to account for the age-dependent survival rates of mandarin fish. Only mandarin fish who survive all the way to harvest are included in the annual reported number of tonnes of mandarin fish produced per year. Mandarin fish who survive to harvest eat a certain number of animals, but just multiplying the average number by the number of mandarin fish harvested will not count the animals fed to mandarin fish who didn’t survive to slaughter. I assume those who die just before harvest would eat the same amount, and the earlier a mandarin fish dies, the fewer they would have eaten. It’s plausible that those who die early eat less on average per day at any given age than those who survive to harvest or longer, but I assume the number per day is the same.

On the other hand, I don’t account for the mortality of feed fish before being available for mandarin fish to eat (e.g. before being added to the same pond). Survival can differ by age before being available to mandarin fish. To account for the statistics on artificial propagation of fish fry in China, this would be mortality after they are counted towards those statistics and before being available for mandarin fish.

Animals used as live feed during their earlier life stages — small enough for the earlier life stages of what eats them — are less likely to be conscious, may have lower welfare ranges or otherwise matter less morally. Accounting for this, the welfare of feed fish during their lives, excluding when eaten, may be important, but not dominant. Their deaths may be among the worst from farmed animals by scale. See the following table for comparison of animals exploited for food:

The table is available here as a sheet with some notes. See also St. Jules, 2024d for further discussion.

How bad is being live fed to mandarin fish?

I don’t investigate the welfare of the feed fish while being raised, but that may be worth doing, if there appear to be any promising interventions. When fed to mandarin fish, it seems more likely than not that they are swallowed live and whole, and so die by digestion (stomach acid), suffocation or crushing. This could be painful and protracted, but it’s unclear.

Predatory/carnivorous fish typically swallow their prey whole (Lindsay, 1984, Meekan et al., 2018, Luiz, 2019, Yasuda, 1960, Amundsen & Sánchez‐Hernández, 2019, St John, 1999, Gill, 2003, Lundstedt et al., 2004), and so without tearing or chewing. This is despite having teeth, and mandarin fish in particular have teeth (Cao et al., 2021). The cause of death seems most likely to be suffocation/asphyxiation, i.e. too little oxygen in their blood, due to too little dissolved oxygen in the predator fish’s stomach or due to damage to the prey’s gills from digestive juices (Waterfield, 2021, Reddit AskScience thread, Poe GPT-4o, Poe Web-Search). Other possibilities include digestive processes (stomach acid, enzymes) and mechanical injury, e.g. crushing (own guesses, Poe GPT-4o, Poe Web-Search).

Fish can survive minutes outside of water or without oxygen in water, and another fish’s stomach may have some swallowed oxygenated water. The fisheries scientist Gerald Waterfield (2021) wrote:

My best estimate of the time that the consumed fish stays alive is from about 15 to 25 minutes, after which the fish dies from lack of oxygen. This process starts as soon as the fish enters the predator’s throat. It happens a little slower at lower temperatures. Even if the prey fish were regurgitated a few minutes fewer than this time, it probably would still expire due to brain damage from the restricted oxygen intake and it would be blinded by its eyes having been greatly damaged from stomach acid.

On the other hand, Poe GPT-4o responded that asphyxiation “can cause death within a minute or two” and death “typically occurs within a matter of seconds to a few minutes, primarily due to asphyxiation and physical trauma”, but could not provide direct sources for its claims when prompted. My best guess is that it takes at least minutes, in line with the survival time for fish out of water, and because they probably won’t have been substantially injured until reaching the stomach.

During this time, besides potential suffering from suffocation and fear, they probably suffer from chemical burns and tissue damage from digestive juices. They might lose consciousness and so stop suffering some time before they die, but I don’t know how long before.

The prey of mandarin fish probably die in similar ways, too.^[6] Furthermore, as discussed in the previous section, mandarin fish are typically fed live fish (fry), and mandarin fish apparently regurgitate already dead food (FAO, a, FAO, b, although the clam is unsourced and I couldn’t verify it). To regurgitate dead food, these mandarin fish — or their bodies generally — must be able to tell the difference, which could be by the movement of the animal they’re eating, in the mouth, throat or stomach. If this is how, then the live feed animals must survive long enough to move and prevent triggering regurgitation, and so may be swallowed live (only swallowed food can be regurgitated). FAO (c) write “live foods are digested and excreted as faeces in 6.5–7.5 h”, although presumably they die much earlier, so this only provides a hard upper bound.

I don’t consider the above referenced sources highly reliable.

Intervention ideas

Given the scale of artificial propagation for live feed for mandarin fish, an apparently narrow use, well-targeted interventions could be extremely high leverage. However, as discussed above, these fish may not just be used as mandarin fish feed, but also to clean the water in ponds in polyculture ponds.

Feed substitution

If mandarin fish can only eat live animals, this makes substitutes for live feed for mandarin fish not very promising; it seems they’d have to be other (moving!) live animals. Attempts to feed them artificial diets have generally been unsuccessful (FAO) and no formulated feeds appear available commercially (Sankian et al., 2019), but research is still ongoing (Liang et al., 2001, Sun et al., 2015, Liu, 2017, Mo et al., 2019, Chen et al., 2021a, Chen et al., 2021b, Shen et al., 2021, Shen et al., 2023). Further R&D could be worth supporting.

A close relative species, Siniperca scherzeri, called the leopard mandarin fish, golden mandarin fish and spotted mandarin fish, although still typically fed live feed, seems to do well on artificial diets (Liu et al., 201, Sankian et al., 2019a, Sankian et al., 2019b, Kim et al., 2022). So, we could support their farming or hybrids with them (Liu et al., 2017) over the mandarin fish Siniperca chuatsi. The mandarin fish Siniperca chuatsi grows faster than the leopard mandarin fish, so artificially selecting the leopard mandarin fish for growth or hybridizing them may be necessary to make them economically competitive with the mandarin fish Siniperca chuatsi (Liu et al., 2017). (EDIT August 22, 2025): Or mandarin fish could be artificially selected to accept artificial diets.

More recently, Xu et al. (2024) wrote about farming mandarin fish Siniperca chuatsi:

Consequently, farming Chinese perch using an artificial diet has become a viable solution to the challenges associated with live fish feeding and has been implemented in several aquaculture farms in Chinese provinces, including Jiangxi, Hubei, and Guangdong [3,5]. However, a critical issue with this farming method is that the technology for the domestication of Chinese perch to eat artificial diet is not yet mature. If the domestication process fails, the Chinese perch will refuse to eat the artificial diet.

(EDIT August 22, 2025): We could therefore support further work in this direction.

Another option could be to substitute feed animals that suffer less when raised or eaten. However, there are some important constraints on options:

1. Their sizes as feed may need to correspond to the size of the mandarin fish (with a feed to mandarin fish length ratio of roughly half (FAO table 4), or around one quarter to half (Yao & Li, 2018, table 3.7.1), or 0.12 to 0.53 (Yao & Li, 2018)),

2. They may need to be motile to be targeted by mandarin fish.^[7]

3. They must be cheap enough.

This may not leave any feasible options that are better for animals:

1. Most other live common feeds in larviculture for fish and crustaceans, like brine shrimp and rotifers, are probably too small for mandarin fish when the mandarin fish grow large enough. Even insects would probably be too small.

2. Bivalves and tunicates may not trigger consumption or might trigger regurgitation.^[8]

3. Decapod shrimp like whiteleg shrimp and larger crustaceans would be too expensive.
4. Other vertebrates could suffer more in expectation than fish as live feed, assuming we are more confident in their sentience.

Small wild-caught fish, e.g. those typically caught for fishmeal, may be preferable to artificially propagated ones, but this is unclear. They would also need to be kept live through and past capture.

Alternatives for maintaining water quality

Lower economic value carp are often stocked in polyculture ponds to maintain water quality (Wang et al., 2016, He et al., 2018, Yao & Li, 2018). In particular, “silver carp and bighead carp clean water through filtering plankton” (He et al., 2018). I have not estimated how many fish are used this way, or whether farms would still do this if not for farming mandarin fish. We may be able to design other methods to maintain water quality that don’t require these fish, or promote the use of other animals that matter less morally, e.g. bivalves, for maintaining water quality. On the other hand, if the alternatives are cheaper — or otherwise more profitable, perhaps they bring in more revenue, too —, this could increase the aquaculture of the higher-value farmed species. Given the scale of fish fry production and substitution between farmed animals, this could be a tradeoff worth making, but it would be worth investigating possible tradeoffs further before proceeding to support such an intervention.

Reducing mandarin fish production

Alternatively, to reduce live feed production and use for mandarin fish, we could try to reduce mandarin fish production. This could be difficult to target specifically, especially being in China. Mandarin fish account for only around 1% of China’s farmed fish production, 0.5% of China’s total aquaculture production, and even less as a share of animal consumption,^[9] so general promotion of plant-based foods or substitutes for people would have little impact on mandarin fish farming and live feed, e.g. discounted to on the order of 0.5% or less of the shift from animal products to plant-based products, assuming the decrease is uniform. On the other hand, more targeted reduction of mandarin fish farming seems less tractable in China, given obstacles to and risks for advocacy and charities there (Echo Sun/Animal Charity Evaluators, 2018, Animal Ethics, 2021) and the potential to undermine animal advocacy in China more generally, but this could be worth investigating further.

Welfare reforms for artificially propagated fish

Welfare reforms for feed fish and artificially propagated fish in general during rearing could also be worth investigating further. One direction for influence could be through the World Organisation for Animal Health, which works on developing international animal welfare standards and capacity to support welfare improvements in cooperation with governments. This would likely exclude the welfare of the animals when eaten by mandarin fish, assuming mandarin fish would still be live fed.

1. ^{^}

    Around 298,057 tonnes of mandarin fish farmed out of 28,788,609 tonnes of farmed fish and 61,536,375 tonnes of all aquaculture production in China in 2015, according to Fishcount. Meat consumption (presumably terrestrial vertebrates only and excluding seafood) is around 100 million tonnes per year in China (Rouzi, 2024, New Zealand Ministry of Foreign Affairs and Trade, 2023, Grimmelt et al., 2023).

2. ^{^}

    Newton et al. (2021) wrote:

   Instead of post-harvest supply chain transformation, data suggest that in the face of stagnating prices for traditional species, successful risk mitigation strategies of Chinese inland aquaculture were to diversify species production to better match variable consumer demand. Silver carp were especially low in value. Whereas silver carp made up the majority of Chinese carp production for human consumption prior to the 1980s (FAO 2020), they were often used as a direct feed input for more highly valued species in Hubei at the time of the survey. Adult silver carp were being minced for swamp eel (Monopterus albus) feed or the fry as live feed for Mandarin fish (Siniperca chautsi).

3. ^{^}

    Wang et al. (2016) wrote:

   In general, freshwater aquaculture in China almost always utilizes models incorporating principles of polyculture: as a means of not only increasing productivity but also improving the utilization of feed resources, allochthonous and autochthonous, which in turn reduces nutrients in the effluent (Wang et al. 2014). The majority of the [Chinese mitten crab] farms (92 %) stocked high-valued, carnivorous, mandarin fish (Siniperca chuasti) of average body length of 4–10 cm at densities ranging from 150 to 900 individuals ha−1. These fish were expected to forage on the naturally recruited wild fish. It was also a very common practice for all the farms surveyed to stock silver carp (Hypophthalmichthys molitrix Valenciennes 1844) and/or big head carp (Aristichthys nobilis Richardson 1845) of average weight in the range of 50–250 g at a density range of 300–750 individuals ha−1. These species feed on plankton and are known to facilitate maintenance of water quality. These practices are also utilized in turtle culture in China and known to be very effective (Wang et al. 2007b; Shi et al. 2008).

    

   Normally harvested silver carp, bighead carp and other small fish are frozen and used as trash fish for the next rearing cycle. However, the more highly valued mandarin fish is sold on capture.

^{^}

 He et al. (2018) wrote:

Polyculture is typical of traditional Chinese aquaculture, and also widely adopted in soft‐shelled turtle culture during the pond culture stage. Although there are many species suitable for this system, the main species include fish such as silver carp, bighead carp, yellowtail catfish, crustaceans, such as firewater giant prawn (Macrobrachium rosenbergi), river shrimp, crab and white‐leg shrimp (Table 3.13.1).

Stocking different species based on their habits enables full use of the water column and available natural food organisms, and the species can benefit from each other. For example, silver carp and bighead carp clean water through filtering plankton, crab and shrimp clean the bottom organic matter by feeding on feed residues, and turtles feed on crustacean molts, as well as dead fish and/or dead shrimp.

^{^}

 10 USD/1000 fingerlings * 1 fingerling / 5 g of fingerlings * 1000 g/kg of fingerlings = 2 USD/kg of fingerlings. If the fingerlings are 1 g each, this would be 10 USD/kg of fingerlings.

On the other hand, grass carp fry cost around 5.7–16% of sales for grass carp culture for direct human consumption in China (Xie et al., 2018, table 2.1.2), so 5.7–16% of the 1.55-2.64 USD/kg of market weight carp above. Or, across fish (mostly carp) aquaculture in China, fry/fingerlings cost 38.7–47% of the cost of production (Yuan, 2007, p.85, table 35; Rola & Hasan, 2007, pp.15–18, tables 16–19). If

1. grass carp are harvested at 1.5kg each ("The marketing size for grass carp is 1-1.5 kg and 1.5-2.5 kg in ponds and cages respectively.” (FAO, 2024)),

2. the survival rate from fry to harvest is 50%

3. harvested grass carp sales prices are 2 USD/kg, and

4. grass carp fry cost farmers 10% of carp sales,

then grass carp fry cost

10% * 2 USD/kg of harvest carp * 1.5 kg of harvest carp / harvest carp * 0.5 harvest carp / fry = 0.6 USD / fry = 600 USD/1000 fry, over 40x more than the 6.76 to 13.5 USD/1000 fingerlings figure from Kumar et al., 2008, table 3.

^{^}

 Lindsay (1984) wrote:

Fish that break up food by means of pharyngeal teeth (cyprinids) or other modifications to the buccal cavity (wrasse, rays) had low chitinase activity in spite of consuming chitin in the diet while those fish that tend to gulp prey whole (salmonids, gadoids, perch, eel, red mullet, gurnards, mackerel) had high activity. This suggests that the primary function of gastric chitinase is to disrupt the chitinous envelope of prey allowing access to the soft inner tissues by the digestive juices. This role would also be functional in those fish destined to be piscivorous as adults but which gulp down relatively large invertebrate prey when young.

Mandarin fish are Perciformes, or “perch-like”, and so related to perch. They are also not cyprinids. So, it’s likely that they “gulp prey whole”.

Some other Perciformes species also swallow prey whole, including multiple Cichla species (Gonzalez Neves dos Santos et al., 2011), Centropomus undecimalis (Sánchez Soto, 2020) and (Catarino & Zuanon, 2010). On the other hand, barracuda are Perciformes, too, and they tear larger prey before swallowing, but they have teeth particularly adapted for doing so, and they swallow smaller prey whole (Bester), unlike mandarin fish.

^{^}

 The feeding strategies of mandarin fish suggest to me that they target motile animals specifically. Yao and Li (2018) wrote:

Their feeding strategy differs at different stages of their lifecycle, using a cruising search‐and‐attack strategy in the larval stage, and a hide‐and‐ambush strategy in the adult stage (Li et al. 2014a; Yao and Liang 2015).

Mo et al. (2019) wrote:

Mandarin fish has a very peculiar feeding habit that they exclusively feed on live fish rather than formulated feeds, and that they detect and capture prey by using their vision and lateral-line mechanoreception. This species has a low visual acuity and preys slowly by stalking.

Furthermore, if mandarin fish regurgitate already dead food (FAO, a, FAO, b), they might not tell the difference between very unresponsive or slow-reacting animals and dead, and so regurgitate non-motile animals, too.

^{^}

 See the previous footnote.

^{^}

 Around 298,057 tonnes of mandarin fish farmed out of 28,788,609 tonnes of farmed fish and 61,536,375 tonnes of all aquaculture production in China in 2015, according to Fishcount. Meat consumption (presumably terrestrial vertebrates only and excluding seafood) is around 100 million tonnes per year in China (Rouzi, 2024, New Zealand Ministry of Foreign Affairs and Trade, 2023, Grimmelt et al., 2023).

Show all footnotes