Release captive-bred individuals to the wild

How is the evidence assessed?
  • Effectiveness
    not assessed
  • Certainty
    not assessed
  • Harms
    not assessed

Study locations

Key messages



  • Abundance (10 studies): Six studies (including one before-and-after study) in the UK, the USA and Poland and Slovakia reported that captive-bred butterfly populations released as eggs, caterpillars, pupae and adults (sometimes into managed habitat or alongside translocated individuals) persisted for 2–28 years and increased in abundance (sometimes with continued captive-rearing of wild-laid caterpillars or supplemented by further releases). Two studies (including one review) in the UK reported that captive-bred large copper and belted beauty moth populations released as caterpillars (sometimes into managed habitat) died out one, two or 12 years after release, or required further releases to survive. One replicated study in the UK reported that three of 10 captive-bred barberry carpet moth populations released as caterpillars (and in one case as adults) established, and at least one persisted for five years. One review across the UK and Ireland found that 25% of captive-bred and translocated butterfly populations survived for >3 years, but 38% died out in that time, and only 8% were known to have survived for >10 years.
  • Reproductive success (2 studies): One study in the UK reported that after the release of a captive-bred population of large copper, the number of eggs laid/female increased over the first three years. One before-and-after study in the UK reported captive-bred adult pearl-bordered fritillaries released into coppiced woodland successfully bred at least once.
  • Survival (3 studies): Three studies (including two replicated, site comparison studies and one review) in the UK and the UK and the Netherlands found that released, captive-bred large copper caterpillars had a lower survival rate than captive, wild or translocated caterpillars.


About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A review in 1929–1966 in three fens in Cambridgeshire and Norfolk, UK (Duffey 1968) reported that released large copper Lycaena dispar batavus caterpillars had lower survival rates than captive caterpillars, and three released populations ultimately died out or required additional releases to survive. Results were not tested for statistical significance. One population of captive-bred large copper survived for 12 years after release, until the fen was drained, but a second population died out two years after release. A third population was maintained for over 30 years by continued releases. In this population, survival from the egg stage to caterpillars in spring in the released population was 4.4%, compared to 5.1% in the captive population, but the survival of caterpillars from spring to pupation was 15.1% in the released population, compared to 79.1% in the captive population. In winter 1929–1930, a 3-ha fen in Cambridgeshire was cleared and planted with great water dock Rumex hydrolapathum. In May 1930, ‘a sufficient number’ of large copper caterpillars were placed on marked plants, and a second release was conducted in 1931 or 1932. In 1942 the fen was drained. In June–July 1949, eighty adults were released on a fen in Norfolk. From the 1930s–1966, a semi-wild population was maintained at a second Cambridgeshire fen, by regular (becoming annual) releases of captive-bred caterpillars. From 1961–1966, the survival of released and captive caterpillars was estimated each year.

    Study and other actions tested
  2. A study in 1970–1976 in a fen in Cambridgeshire, UK (Duffey 1977) reported that after the release of captive-bred large copper butterflies Lycaena dispar batava, the number of eggs laid/female and the number of caterpillars emerging after hibernation increased over three years, and the population survived for at least six years. Results were not tested for statistical significance. One year after the first release of adult butterflies, 111 caterpillars emerged from hibernation. In the second year, 427 caterpillars emerged, and in the third year 1,344 caterpillars emerged. The number of eggs laid/female increased from 4.85 in the first year to 89–100 in the fourth year. Six years after the first release, eggs were widely distributed across the site. In late summer 1970, caterpillars and pupae from two captive-bred populations were placed in muslin cages across a fenland nature reserve, from which 517 males and 551 females were released. In spring 1971–1973, wild-hatched caterpillars were collected and reared to pupation in muslin cages, and additional releases from captive stock were made (344–554 males/year, 208–446 females/year). Wild-hatched caterpillars were reared in cages again in later years. No details are given on how the eggs and caterpillars were counted and collected.

    Study and other actions tested
  3. A study in 1984–1989 in a woodland in Essex, UK (Warren 1991) reported that captive-bred heath fritillary Mellicta athalia released into a coppiced woodland survived and the population increased. Results were not tested for statistical significance. Two years after the release of 53 adult heath fritillaries, the population was nearly 3,000 adults (when the extent of breeding habitat was at its maximum), but stabilized at around 500 adults after five years. In 1984, a total of 53 captive-bred adult heath fritillaries (31 females, 22 males) were released into a coppiced woodland (coppicing commenced in 1980) containing around 4 ha of hostplant (common cow-wheat Melampyrum pratense). From 1985–1989, butterflies were surveyed annually on timed counts along a zig-zag route covering the known flight areas. The total yearly population was estimated by multiplying the peak population count by three.

    Study and other actions tested
  4. A review from pre-1900–1988 across the UK and Ireland (Pullin 1996) found that at least a quarter of reintroduced butterfly populations survived for over three years, but only 8% were known to survive for more than 10 years after release. Of 274 documented reintroductions of native butterflies, 68 populations (25%) were known to have survived for more than three years, and 21 (8%) were known to survive for more than 10 years. However, 103 populations (38%) died out within three years of release. The remaining reintroductions were either poorly documented (73 releases) or occurred too recently to determine success (30 releases). Twenty-five releases which aimed to reinforce existing populations are not included. Records of all documented releases of butterflies in the UK and Ireland were compiled by Oates & Warren (1990), and their success up to 10 years after release was updated by this study. At least 29% of releases were of captive-bred butterflies. No further details were provided.

    Oates M.R. & Warren M.S. (1990) A review of butterfly reintroductions in Britain and Ireland. Joint Committee for the Conservation of British Insects/World Wildlife Fund, Godalming, United Kingdom.

    Study and other actions tested
  5. A replicated, site comparison study in 1995–1998 in three fens in Norfolk, UK and nine in Overijssel, the Netherlands (Nicholls & Pullin 2000a, same study as Nicholls & Pullin 2000b) found that released, captive-bred large copper Lycaena dispar batavus caterpillars had lower overwinter survival rates than wild caterpillars. The overwinter survival of released captive-bred caterpillars in the UK (0–7%/year: 30/1,440 caterpillars found) was lower than the overwinter survival of a wild population in the Netherlands (19–42%/year: 41/139 caterpillars found). Post-winter survival to pupation of released captive-bred caterpillars in the UK was 49–56% (215/420 caterpillars found). A captive population of large copper was maintained for >25 years. From September–October 1995–1997, captive-bred caterpillars were released in seven 50 × 50 m plots across three sites in Norfolk. Five caterpillars were placed on each of 7–40 randomly chosen great water dock Rumex hydrolapathum plants/plot/year (total: 1,440 caterpillars). In April–May 1996–1998 plants were searched for surviving caterpillars. In April–May 1996 and 1997 a further 15–70 caterpillars/plot/year (total: 420 caterpillars) were released, and monitored through to pupation. In 1996–1997, at nine fens within a 3,500-ha lowland bog in the Netherlands, wild large copper eggs were counted on every water dock encountered along a transect through each fen. Plants were revisited three weeks later, and in April the following year, to record caterpillar survival.

    Study and other actions tested
  6. A replicated, paired, site comparison study in 1997–1998 in a fen in Norfolk, UK (Nicholls & Pullin 2000b, same study as Nicholls & Pullin 2000a) found that released captive-bred large copper Lycaena dispar batavus caterpillars had lower overwinter survival rates than translocated wild caterpillars. The overwinter survival of released captive-bred caterpillars (1 of 95 caterpillars found) was lower than the overwinter survival of translocated wild caterpillars (8 of 95 caterpillars found). In September 1997, captive-laid eggs were obtained from a 25-year-old glasshouse-reared colony at Woodwalton Fen, and wild-laid eggs were collected from a 3,500-ha lowland bog in the Netherlands. Eggs from both sources were reared to overwintering in the laboratory. A total of 95 captive-bred and 95 wild caterpillars were placed on to 19 pairs of great water dock Rumex hydrolapathum (5 caterpillars/plant) in an open fen in Norfolk. In May 1998, after late flooding, surviving caterpillars were counted on each plant.

    Study and other actions tested
  7. A study in 2002–2003 in a coastal sand dune in Caernarvonshire, UK (Howe et al. 2004) reported that a population of captive-bred belted beauty moth Lycia zonaria britannica did not survive the first year after release. One year after the first release of caterpillars, no adults were found at the site. Results for the second year were not presented. In April 2002 and 2003, three to eight gravid females were collected from an 89-ha sand dune system. Their eggs were kept, and the caterpillars were reared on bird’s-foot trefoil Lotus corniculatus. In late May 2002 and 2003, a total of 2,030 caterpillars were released (number/year not specified) at 22 locations across a 7.4-ha sand dune (21.5 km from the capture site), most of which was managed as a nature reserve. In spring 2003, the release site was searched for emerging adults. The authors also reported that two releases of small numbers of caterpillars to a site in Merseyside, UK, in the early 1990s, were unsuccessful (no further details provided).

    Study and other actions tested
  8. A replicated study in 1987–early 2000s (exact year not given) in 10 woodlands and hedgerows in England, UK (Waring 2004) reported that three out of 10 released populations of captive-bred barberry carpet moth Pareulype berberata established in the wild. At three sites where hundreds of barberry carpet moth caterpillars, and in one case adults, were released, self-maintaining populations were established. One population had survived for five years. However, at one site where thousands of caterpillars were released in multiple attempts, some breeding took place but the population died out after three generations. The author suggested that common barberry Berberis vulgaris bushes needed to be trimmed to generate new growth. At another site where thousands of caterpillars were released on to cultivated barberry (Berberis thunbergii and Berberis ottawensis), the population failed to establish, despite caterpillars using these species in captivity. From 1987, but especially from the late 1990s, small numbers of barberry carpet moths (adults and caterpillars) were collected from the wild. Captive-bred caterpillars were released in multiple batches at 10 sites in Wiltshire, Northamptonshire, Lincolnshire, Suffolk, Cambridgeshire and Norfolk with established barberry bushes. At one site adults were also released. No monitoring details were provided.

    Study and other actions tested
  9. A study in 1977–2005 in an alpine meadow in Colorado, USA (Boggs et al. 2006) reported that a captive-bred and translocated population of Gillette's checkerspot Euphydryas gillettii released outside the species’ native range survived for 28 years, but only increased in size and colonized new sites after 22–25 years. For 21 years after the release of 83 egg clusters, the population size fluctuated between 24 and 143 adults, and remained confined to the release site. However, four years later, the population was estimated at >3,000 adults, and covered 70.4 ha. After a further three years, the population had declined to 150 adults at the release site, but two other habitat patches (0.3 and 0.6 ha) remained occupied (13–153 adults/ha). In July 1977, eggs and adult females were collected in Wyoming, and kept in a laboratory where more eggs were laid. In July–August 1977, eighty-three wild- and captive-laid egg and caterpillar clusters (~10,000 individuals from ~40 females) were released on to bearberry honeysuckle Lonicera involucrata in a 2.25-ha meadow. Details of the translocation taken from Holdren & Ehrlich (1981). In June–July 1978–1989 and 2002–2005, adult butterflies at the release site were caught and uniquely marked every 1–7 days. Recapture rates of marked butterflies were used to estimate the population size in years with sufficient data (1981–1986, 2002–2005). In 1978–1989, 2002 and 2004–2005, egg clusters and/or caterpillar webs were counted throughout the season at the release site, and in 2003–2005 at two newly colonized sites. The relationship between number of egg clusters and adult population was used to estimate the population size in the remaining years. From 1978–1987, areas surrounding the release site were searched for egg clusters or caterpillar webs, and from 2002–2005 a larger area was searched for adults.

    Holdren C.E. & Ehrlich P.R. (1981) Long range dispersal in checkerspot butterflies: transplant experiments with Euphydryas gillettii. Oecologia, 50, 125–129.

    Study and other actions tested
  10. A before-and-after study in 1991–2003 in a limestone montane grassland reserve in Poland and Slovakia (Adamski & Witkowski, 2007) found that releasing captive-bred Apollo butterflies Parnassius apollo frankenbergeri increased both the population size and the number of occupied sites. Eleven years after the release of captive-bred butterflies began, a population of Apollo butterflies contained ~1,000 individuals, compared to 30 individuals the year before the first release. The number of sub-populations increased from one to more than 12. However, when the number of butterflies in a sub-population was high, the sub-population size in a given year was lower when more captive-bred butterflies had been released the previous year (data presented as model results). In 1991, butterflies from a remnant population were taken into captivity and bred (number captured not specified). From 1992–2001, between 22 and 658 butterflies/year were released across four areas (2,917 butterflies released in total). In 1995, the captive population was supplemented with butterflies from another population to increase genetic diversity. The species’ host plant, stonecrop Sedum maximum, was abundant at the release sites. From 1991–2003, suitable sites were visited twice/week for 1–3 hours during the flight period, and population size was estimated by marking and recapturing all butterflies observed.

    Study and other actions tested
  11. A replicated study in 2007–2009 in four wet grasslands in Cumbria, UK (Porter & Ellis 2011) reported that three out of four released populations of marsh fritillary Euphydryas aurinia increased in size over three years. Results were not tested for statistical significance. Three years after the release of 3,500–22,900 caterpillars/site at three sites, the number of caterpillar webs was 47–240 webs/site, compared to 16–112 webs/site one year after release. However, at a fourth site where 11,000 caterpillars were released, the number of webs remained low (1–10 webs/year). From before 2004, four sites were managed by a combination of scrub control (all sites), removal of trees (one site), reinstating cattle or pony grazing at 0.4 livestock units/ha (all sites), water level management (two sites) and cutting and removal of course grasses (one site) to increase the area of suitable habitat for marsh fritillary (estimated at 7.4–20.0 ha/site in 2007). In March–April 2007, a total of 42,000 caterpillars were released at four sites (3,500–22,900 caterpillars/site). Caterpillars were placed in groups of ~100 on clusters of devil’s-bit scabious Succisa pratensis. From mid-August 2007–2009, every patch of devil’s-bit scabious at each site was searched for caterpillar webs.

    Study and other actions tested
  12. A before-and-after study in 2007–2011 in a woodland in East Sussex, UK (Hoare et al 2012) reported that captive bred pearl-bordered fritillary Boloria euphrosyne released into coppiced woodland bred at least once. Results were not tested for statistical significance. In 2011, one year after the release of some captive-bred adult pearl-bordered fritillaries (numbers not provided) into an area where they had previously gone extinct, their offspring were recorded at the site. At some time between 2007 and 2010 surveys showed that there was no longer a population of pearl-bordered fritillary in Rother Woods, so during this time coppicing management was reinstated (exact management dates not provided). Captive-bred adults were released into the site in 2010 and a survey in 2011 recorded adults of the next generation.

    Study and other actions tested
  13. A study in 2009–2016 in six coastal grasslands in Essex, UK (Gardiner et al 2017) reported that released captive-bred Fisher’s estuarine moths Gortyna borelii lunata established populations and bred at all sites, in some cases for seven years. Six populations resulting from the release of captive Fisher’s estuarine moths were present for up to  seven years, and breeding was confirmed at all sites (sometimes in all seven years). Numbers of individuals and years of confirmed breeding events at each site are not provided. In March–April 2009–2016, blades of grass containing eggs from a captive breeding programme were placed next to hog’s fennel Peucedanum officinale plants at each of the six sites. Later each year some caterpillars which were not needed for the continuation for the captive breeding programmes were also introduced to each site. Two sites received eggs and caterpillars in two years, and four only in one year (years of release at each site not provided). Between mid-July and mid-August each year hog’s fennel plants at each site were searched for distinctive feeding signs to confirm presence of the moth.

    Study and other actions tested
Please cite as:

Bladon A.J., Bladon, E. K., Smith R.K. & Sutherland W.J. (2023) Butterfly and Moth Conservation: Global Evidence for the Effects of Interventions for butterflies and moths. Conservation Evidence Series Synopsis. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Butterfly and Moth Conservation

This Action forms part of the Action Synopsis:

Butterfly and Moth Conservation
Butterfly and Moth Conservation

Butterfly and Moth Conservation - Published 2023

Butterfly and Moth Synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.

Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust