Rear declining species in captivity
Overall effectiveness category Awaiting assessment
Number of studies: 13
Background information and definitions
Captive-breeding can be an important method for boosting small population sizes, or for ensuring the continued persistence of a species while its habitat is restored to a suitable condition in the wild (Schultz et al. 2008). It is likely to be more successful if founding individuals are taken from large populations, and the size of the captive population remains stable over time (Crone et al. 2007).
CAUTION: Before removing butterflies and moths from the wild to take into captivity, the impact of the removal of individuals from the donor site must be considered, to avoid harming existing populations.
For studies on the release of captive-bred butterflies and moths into the wild, see “Release captive-bred individuals to the wild”.
Crone E.E., Pickering D. & Schultz C.B. (2007) Can captive rearing promote recovery of endangered butterflies? An assessment in the face of uncertainty. Biological Conservation, 139, 103–112.
Schultz C.B., Russell C. & Wynn L. (2008) Restoration, reintroduction, and captive propagation for at-risk butterflies: A review of British and American conservation efforts. Israel Journal of Ecology & Evolution, 54, 41–61.
Supporting evidence from individual studies
A replicated, paired, controlled study in 1964 in a fen in Cambridgeshire, UK (Duffey 1968) reported that semi-wild large copper Lycaena dispar batavus caterpillars reared at high density on small great water dock Rumex hydrolapathum plants had lower survival than caterpillars reared at lower density or on larger plants. Results were not tested for statistical significance. On small plants, the survival of large copper caterpillars reared in groups of 12/plant was 35% (67/192 survived), compared to 65% (31/48 survived) for caterpillars in groups of three/plant. However, on large plants the survival of caterpillars in groups of 12/plant was 81% (39/48 survived) compared to 75% (9/12 survived) for caterpillars in groups of three/plant. The author reported that small plants with 12 caterpillars/plant were abandoned after all the leaves had been eaten, before the caterpillars were fully grown. In a fen with a semi-wild large copper colony, four batches of 10 great water dock plants were selected. In each batch, eight plants were 50 cm tall (3 leaves/plant) and two were >100 cm high (9–20 leaves/plant). In May 1964, three or 12 large copper caterpillars were placed onto each plant, and the plants were covered with a 6-mm plastic mesh cage to exclude birds and mammals. In July 1964, all surviving caterpillars and pupae were counted.Study and other actions tested
A controlled study in 1968 at a research station in Cambridgeshire, UK (Duffey 1977) reported that large copper butterflies Lycaena dispar batava laid more eggs in a cage kept in a greenhouse than in a cage kept outside, and eggs in the greenhouse had a higher hatching success. Results were not tested for statistical significance. The number of eggs laid in a cage kept in a greenhouse (498 eggs) was higher than the number laid in a cage kept outside (126 eggs). In addition, the proportion of eggs which hatched was higher in the greenhouse (91%) than outside (40%). In summer 1968, two cages (5.40 × 1.65 × 1.80 m) were constructed from 1 × 1 cm mesh. One was kept in a greenhouse and the other was placed outside. Each cage contained 15 female and 23 male large coppers, and 20 potted great water dock Rumex hydrolapathum plants. From 17 May–14 August 1968, the mean maximum temperature in the greenhouse cage (23.8°C) was higher than in the outside cage (16.8°C), but the mean minimum temperature was similar (greenhouse: 8.7°C; outside: 8.6°C). Eggs were counted daily.Study and other actions tested
A controlled study in 1979–1980 in a laboratory in the UK (Thomas & Wardlaw 1990) found that large blue Maculinea arion caterpillars reared in ant Myrmica spp. nests without a queen present were more likely to survive than caterpillars reared in nests with a queen. The survival of caterpillars in nests without queen ants (10 out of 26 caterpillars) was higher than in nests with queen ants present (6 out of 39 caterpillars). The authors reported that caterpillars in Myrmica scabrinodis nests had lower survival than caterpillars in nests of the other species (data not presented). In 1979 and 1980, a total of 65 Myrmica ant colonies were established, containing 20–1,137 workers/colony depending on nest design (see paper for details). In each of 26 nests, 1–6 queen ants (depending on colony size) were present, and the other 39 nests did not contain queens. Most nests were Myrmica sabuleti, but four colonies were established with each of Myrmica rubra, Myrmica ruginodis and Myrmica scabrinodis. After >1 week, one newly moulted caterpillar was introduced to each nest. Caterpillar survival was monitored for >2 weeks.Study and other actions tested
A controlled study (years not given) in a laboratory in the UK or France (location not clear) (Elmes et al. 1991) found that the survival of mountain Alcon blue Maculinea rebeli caterpillars reared in captivity differed between ant Myrmica spp. species, but not between colonies with or without queens present. Mountain Alcon blue caterpillars reared with Myrmica schencki (10 out of 99 survived to pupation) had higher survival rates than caterpillars reared with Myrmica sabuleti (4/78 survived), Myrmica scabrinodis (2/43 survived), Myrmica rubra (1/112 survived), Myrmica ruginodis (4/71 survived) or Myrmica sulcinodis (1/24 survived). The survival of caterpillars reared for three weeks in colonies with queens was 31–89%, compared to 43–78% without queens (statistical significance not assessed, see paper for details on each ant species). Over five years, >800 mountain Alcon blue caterpillars were introduced to 120 ant colonies kept in small plastic “Brian” nests (no further details provided). Colonies were collected from France and England, fed fruit flies Drosophila spp. and sucrose, and kept at a constant temperature which was adjusted weekly to mimic natural temperatures. Caterpillars were collected on their food plant, and placed into the foraging areas of the ants after emergence. Caterpillar survival was monitored for between three weeks and 10 months (to pupation) in the ant nests.Study and other actions tested
A replicated, controlled study (years not given) in a laboratory in Spain (Thomas et al. 1993) found that mountain Alcon blue Maculinea rebeli caterpillars reared in ant Myrmica rubra nests at low density had higher survival rates than caterpillars reared at higher densities. The survival of mountain Alcon blue caterpillars reared at low density (five caterpillars/nest: 18 of 20 caterpillars survived) was higher than for caterpillars reared at higher densities (10 caterpillars/nest: 29/40 survived; 25 caterpillars/nest: 22/100 survived). At higher densities, more caterpillars survived in ant nests founded from a colony containing winged females (10: 19/20 survived; 25: 17/50 survived) than from a colony without winged females (10: 10/20 survived: 25: 5/50 survived), but surviving individuals were lighter in colonies with winged females (22–27 mg) than without them (37–50 mg). Caterpillars reared at low density weighed 39–46 mg. Two Myrmica rubra nests in the Pyrenees were excavated and used to establish 12 colonies, each containing 50 workers and 10 ant larvae. Colonies were kept in “Brian” nests with abundant food (no further details provided). When excavated, one nest contained a large number of winged females while the other contained none. In August, on the evening of their final moult, 160 mountain Alcon blue caterpillars were collected from the same site, and introduced to the ant colonies at three densities: five, 10 and 25 caterpillars/colony. From October–March, nests were overwintered in a cool room, after which caterpillars grew for another eight weeks until pupation. The survival and weights of caterpillars were recorded before pupation.Study and other actions tested
A review in 1998 (Wardlaw et al. 1998) reported that four species of large blue butterfly Maculinea spp. were bred in captivity using ant Myrmica spp. colonies, with varying success. Alcon blue Maculinea alcon and mountain alcon blue Maculinea rebeli caterpillars were successfully reared in captive ant colonies on different occasions over 20 years (data not presented). Large blue Maculinea arion caterpillars were sometimes reared successfully using two methods, but 13 other nest designs failed (data not presented). Scarce large blue Maculinea teleius caterpillars were reared for up to eight months, using common red ant Myrmica rubra and Myrmica scabrinodis in two nest designs. Wild butterfly eggs were collected on flowering stems of food plants (gentian Gentiana spp., wild thyme Thymus spp., oregano Origanum spp., great burnet Sanguisorba officinalis). Ant colonies (see paper for six species) were collected by excavating nests. Caterpillars were introduced to ant colonies after they dropped from the flower heads. A variety of rearing methods were used, with different nest box designs (see paper for details).Study and other actions tested
A site comparison study in 1997–1998 in a laboratory in the UK (Nicholls & Pullin 2000) found that wild- and captive-laid large copper Lycaena dispar batavus eggs and caterpillars had similar survival in captivity. Both the survival to overwintering of wild-laid eggs (19 of 20 caterpillars), and the overwinter survival of these caterpillars (4 of 19 caterpillars), were statistically similar to the survival to overwintering (15 of 20 caterpillars) and overwinter survival (3 of 15 caterpillars) of captive-laid eggs. In September 1997, twenty wild-laid eggs were collected from a lowland bog in the Netherlands, and 20 captive-laid eggs were obtained from a 25-year-old glasshouse-reared colony at Woodwalton Fen. Eggs were reared to overwintering under controlled conditions (10 hours light, 14 hours dark, 20°C) in a laboratory. Immediately before overwintering, caterpillars were transferred to great water dock Rumex hydrolapathum pot plants and maintained in an overwinter environment (10 hours light, 14 hours dark, 5°C) for 20 weeks. Emergence was stimulated by increasing light by 15 minutes, and increasing temperature by 2°C, every two days for eight days, and survival was recorded.Study and other actions tested
A study (year not specified) in two captive-rearing facilities in Warwickshire and Oxfordshire, UK (Lewis & Thomas 2001) reported that large white Pieris brassicae were successfully reared in captivity for >25 years, but found some morphological changes occurred. A population of large white were bred in captivity for >25 years (100–150 generations). However, long-term captive-bred butterflies were heavier (1.9 g) than butterflies in their third generation in captivity (1.8 g), and had smaller, shorter and broader wings (see paper for details). Captive-bred females laid more eggs (340 eggs/female) than females new to captivity (30 eggs/female). Caterpillars in a long-term captive population, originally caught in southern England >25 years ago, were reared on a synthetic diet at 23–25°C. Adults were kept in cages (45 × 80 × 48 cm) with 150–200 adults/cage. Wild egg batches were collected in Glamorgan, UK, and reared through two generations in captivity in the same conditions. Data were collected on the number of eggs laid in the first 16 days after emergence by 15 females kept with 15 males of each group, and the weight and wing size of freshly emerged adults (number not specified).Study and other actions tested
A controlled study in 2003–2006 in two captive-breeding facilities in Washington, USA (Schultz et al. 2009) found that captive-reared Puget blue butterflies Icaricia icarioides blackmorei were smaller than wild-born individuals, and caterpillars kept in refrigerators overwinter had lower survival than other treatments. There was no significant difference between the survival to adulthood of eggs collected from the wild (17/200 eggs) and eggs laid in captivity (39/548 eggs), or of caterpillars kept in environmental chambers (49/514 caterpillars) or outdoor enclosures (49/450) overwinter. However, all 308 caterpillars kept in refrigerators overwinter died. Captive-reared butterflies were smaller than wild-caught butterflies, but adult size was similar between all captive treatments (see paper for details). In June 2003, forty-eight female butterflies were collected from the wild and 39 laid 1,879 eggs in captivity. Overwinter, surviving caterpillars were kept in one of three treatments: a refrigerator, an environmental chamber with light, humidity and temperature approximating optimal rearing conditions, or an outdoor enclosure experiencing ambient conditions (see paper for housing details). In 2004, surviving caterpillars were reared on netted sickle-keeled lupine Lupinus albicaulis. In spring 2005, sixty female butterflies were collected from a second site and 51 laid 548 eggs in captivity. In addition, lupine leaves with 200 wild-laid eggs were collected and reared in captivity. All caterpillars overwintered in outdoor enclosures. In 2004–2006, captive-reared and wild-caught adults were weighed and measured.Study and other actions tested
A study in 2004–2007 in six captive-breeding sites in Cumbria, UK (Porter & Ellis 2011) reported that a captive population of marsh fritillary Euphydryas aurinia increased in size over two years. Results were not tested for statistical significance. Two years after 250 caterpillars were taken into captivity, the captive population was estimated at 50,000 caterpillars. In September 2004, the only two caterpillar webs (containing 155 individuals) remaining locally were taken into captivity. In addition, 95 caterpillars from 19 populations (five from each location) in west Scotland were collected. Caterpillars were checked for infection with the parasitoid Cotesia melitaearum. Caterpillars were kept at six separate locations, and reared in natural conditions using large netted cages and pot-grown devil’s-bit scabious Succisa pratensis, supplemented with garden varieties of honeysuckle Lonicera spp., snowberry Symphoricarpos albus and wild honeysuckle Lonicera periclymenum. The number of caterpillars in the captive population was estimated in spring 2007.Study and other actions tested
A replicated, site comparison study in 2014 in a laboratory in Poland (Witek et al. 2016) found that scarce large blue Maculinea teleius caterpillars reared by ants Myrmica scabrinodis from sites where the butterfly occurs survived longer than caterpillars raised by ants from sites where the butterfly does not occur, but all caterpillars ultimately died. The survival of scarce large blue caterpillars raised in ant colonies collected from sites where scarce large blue occurs was higher than in colonies collected from sites where scarce large blue does not occur (data presented as model results). However, no caterpillars survived >35 days. In August 2014, ten ant colonies were collected from each of four wet meadows, 110–470 km apart: two where scarce large blue and other ant parasites occurred and two where they did not. Each colony (50 old and 50 young workers with 15 ant larvae) was placed in a plastic box (20 × 12 × 7 cm) containing a patch of wet plaster covered by a flowerpot saucer with an entrance notch. Great burnet Sanguisorba officinalis stems were collected from one site, and placed in water with the flowerheads bagged in eight bunches of 25 stems. Bunches were shaken each morning to collect fourth instar caterpillars, and one caterpillar was placed in each ant colony. Fifteen ant larvae were added to each colony each week as food. The survival of caterpillars was checked every 1–2 days until all caterpillars had died.Study and other actions tested
A replicated, randomized, paired, controlled study in 2014–2016 in a greenhouse in Iowa, USA (Pocius et al. 2017) reported that monarch butterflies Danaus plexippus were successfully reared in captivity for 12 generations, and found that caterpillar survival differed between milkweed Asclepias species. A population of monarchs was bred in captivity for 12 generations. However, more caterpillars fed on butterfly milkweed Asclepias tuberosa (75%) or poke milkweed Asclepias exaltata (72%) survived to adulthood than caterpillars fed on tall green milkweed Asclepias hirtella (31%) or prairie milkweed Asclepias sullivantii (36%). In May–June 2014, a total of 253 wild monarch eggs and young caterpillars were collected. Caterpillars were fed on common milkweed Asclepias syriaca in the summer, and a tropical milkweed Asclepias curassavica in the autumn and winter. Adults were tested for parasites Ophryocystis elektroscirrha before being allowed to mate. In the 13th generation, individual, newly hatched caterpillars were placed on an 8-week-old milkweed plant grown from seed. Thirty-six blocks, each containing one plant of nine milkweed species (butterfly, poke, tall green, prairie, common, swamp Asclepias incarnata, showy Asclepias speciosa, whorled Asclepias verticillata and honeyvine Cynanchum leave milkweed), were placed in a pop-up cage (57 × 37 × 55 cm) and netting in a greenhouse. From day 12, cages were checked daily, and pupae were moved to a laboratory until emergence.Study and other actions tested
A replicated, randomized, controlled study in 2014–2015 on a farm in Galilee, Israel (Berman et al. 2018) found that captive spring webworm Ocnogyna loewii caterpillars fed vegetation from grazed paddocks had a similar growth rate to caterpillars fed vegetation from ungrazed paddocks. Over five days, the growth rate of caterpillars fed on vegetation from cattle-grazed paddocks (0.12 mg/mg/day) was similar to caterpillars fed vegetation from ungrazed pastures (0.11 mg/mg/day). Sixty wild, fourth instar caterpillars were collected and weighed, and placed in individual plastic containers (12 cm diameter, 8 cm height) with a perforated lid. Caterpillars were randomly divided into six groups, and fed daily with fresh plants from one of six paddocks (three grazed, three ungrazed). Caterpillars were re-weighed after five days, and their growth rate calculated.Study and other actions tested