Providing evidence to improve practice

Action: Captive breeding salamanders (including newts) Amphibian Conservation

Key messages

  • Four of six replicated studies (including four small studies) in Japan, Germany, the UK and USA found that eggs were produced successfully in captivity, in one case by one captive-bred female. Two found that production of eggs depended on tank habitat or was more successful in semi-natural compared to laboratory conditions. Captive-bred salamanders were raised to yearlings or a small number of larvae or adults in captivity. One review found that four salamander species bred successfully in captivity, but slimy salamanders produced eggs that did not hatch.
  • One replicated study in Japan found that 60% of Japanese giant salamander eggs survived to hatching in captivity. Two replicated studies (including one small study) in Mexico and the USA found that larval development, body condition and survival of captive-bred amphibians were affected by water temperature, density and whether they were raised under laboratory or semi-natural conditions.


Supporting evidence from individual studies


A small, replicated study in 1979–1980 of captive Texas blind salamander Typhlomolge rathbuni in Cincinnati Zoo, USA (Maruska 1982) found that the species bred successfully in captivity. Three clutches of 8–21 eggs were produced. Three larvae from the first clutch survived and all 14 eggs from the third clutch hatched. Embryonic development required a constant water temperature of 20–21°C as lower temperatures resulted in deformities. Adults had been in captivity since 1975 and were moved to separate tanks once eggs were produced.



A small, replicated study in 1981 of captive Pyrenean mountain salamanders Euproctus asper asper in the UK (Wisniewski & Paull 1982) found that eggs were produced by one of two pairs in captivity. Eight eggs were produced by the pair in June 1982. Three of the eggs hatched one month later. One tadpole survived to at least six months having eaten the other two. Mating behaviour had been observed all year. Two pairs were obtained in 1981 and housed in two aerated 30 x 20 x 20 cm tanks. Gravel substrate and cover were provided.



A review of captive breeding programmes (Maruska 1986) found that a number of amphibian species have been bred successfully in captivity. Salamander species that were bred successfully in captivity were: Texas blind salamander Typhlomolge rathbuni, Tennessee cave salamander Gyrinophilus palleucus, Japanese giant salamander Andrais japonicas and Anderson's salamander Ambystoma andersoni. Slimy salamander Plethodon glutinosus produced eggs but they did not hatch.



A replicated study in 1978–1988 of captive Japanese giant salamanders Andrias japonicas in a zoo in Hiroshima, Japan (Kuwabara et al. 1989) found that the salamanders bred successfully in captivity. Between one and three females produced eggs in each of the three breeding groups each year. A total of 36 egg masses were produced between 1979 and 1988 (336–2434 eggs/mass). Survival to hatching was approximately 60% (range: 0–97%). By 1988, there were 1,035 captive larvae and young. Males were observed eating eggs in the smaller enclosures. Between 1978 and 1983, three breeding groups were established with 2–4 males and three females. Groups were housed in two to four connected outdoor tanks (90 x 70 x 45 cm) with sand and water. Larvae were reared in separate outdoor tanks (65 x 38 x 15 cm).



A small, replicated study in 1990–1994 of Texas salamanders Eurycea neotenes at the Dallas Aquarium, USA (Roberts, Schleser & Jordan 1995) found that captive breeding was successful under certain conditions. In 1991, the female in a planted tank deposited 19 eggs. Eggs were transferred to a dark tank and four hatched. After one year, two of the three surviving captive-bred salamanders laid fully developed eggs. No further reproductive behaviour was seen for 1.5 years. The one original and one captive-bred female placed in an artificial aquifer laid eggs in 1–2 years. Larvae left in the aquifer were not predated by the parents over two months. Three pairs of wild-caught salamanders were housed in separate 4 L aquaria with water flow (22°C). One had gravel substrate, one contained plants and the other partially buried rocks. Fourteen additional animals were housed in a 189 L aquarium with water flow and pipe sections. One original and one captive-bred pair were placed in an artificial 1.2 m long aquifer.



A replicated study in 2003 of Mexican axolotls Ambystoma mexicanum in Xochimilco, Mexico (McKay 2003) found that survival was lower but growth and body condition greater in captive-bred animals reared in semi-natural compared to laboratory conditions. Survival was significantly lower from day 10 to 30 under low-maintenance, semi-natural conditions in canals than under high-maintenance laboratory conditions (8–42% vs 88–98%). Survival was independent of density. By day 30, axolotls reared in canals were significantly larger than those in the laboratory. Those raised at low densities were significantly larger than those at high densities (canal: 0.44 vs 0.39; lab: 0.17 vs 0.11 g). Body condition was significantly better under semi-natural conditions. Average maintenance costs/axolotl/year for seven captive-breeding facilities were US$14–340. One hundred and fifty larvae were divided between six aquaria (45 L; 17–19°C) and six cages in a canal (45 L; 19–24°C) at densities of either 5 or 20 larvae. Aquaria had artificial plants and were cleaned every 10 days. The canal had filter systems to prevent aquatic predators from entering. Larvae were measured every 10 days.



A replicated study in 1994–2004 in the Luhe valley, Germany (Kinne 2004) found that captive breeding in an outdoor enclosure was significantly more effective than in indoor tanks for great crested newts Triturus cristatus and smooth newts Triturus vulgaris. Initially, 24 of each species were housed in 100 L indoor tanks (one male and three females/tank). In following years, eggs produced in an outdoor enclosure were collected and transferred indoors for rearing. Sixty great crested newt and 90 smooth newt larvae and juveniles were released into two created ponds annually.



A small, replicated study in 2009–2011 of southern dwarf sirens Pseudobranchus axanthus at the Central Florida Zoo and Botanic Gardens, USA (Stabile 2012) found that breeding occurred in captivity. Single eggs were recorded in December 2010 and larvae in February 2011. The 12 larvae observed were moved to a separate tank and all survived to at least nine months. Nine wild caught animals were housed in two aquaria (38 L) with sand and leaf litter substrate. In May 2010, two males and two females were moved to outdoor cattle troughs filled with rainwater, with sand and leaf litter substrate. Local aquatic plants were added and invertebrates colonized naturally.


Referenced papers

Please cite as:

Smith, R.K., Meredith, H. & Sutherland, W.J. (2017) Amphibian Conservation. Pages 9-65 in: W.J. Sutherland, L.V. Dicks, N. Ockendon & R.K. Smith (eds) What Works in Conservation 2017. Open Book Publishers, Cambridge, UK.