Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Remove or control fish using rotenone Amphibian Conservation

Key messages

  • Three studies (including one replicated study) in Sweden, the UK and USA found that eliminating fish using rotenone increased numbers of amphibian species, abundance and recruitment or newt populations.
  • One review in Australia, the UK and USA found that fish control, which included using rotenone, increased breeding success for four amphibian species.
  • Two replicated studies in Pakistan and the UK found when rotenone was applied, many frogs died and a small number of newts showed symptoms of negative effects.


Supporting evidence from individual studies


A replicated, before-and-after study in 1970 of three ponds in Mymensingh, Pakistan (Haque 1971) found that rotenone treatment to eradicate fish resulted in death of frogs. It was reported that many frogs died following application of rotenone, but that a similar number escaped death by moving to the shore. Fish were affected within 5 minutes of application. Approximately 110 kg of fish were removed from the ponds. There was no significant difference between the effects of the three treatment concentrations. Rotenone was added to three ponds in concentrations of 1.0, 1.5 and 2.0 parts per million in May 1970. Fish were collected and ponds monitored for two days following treatment.



A before-and-after study in 1992 of an artificial pond in woodland in England, UK (McLee & Scaife 1992/1993) found that great crested newts Triturus cristatus and smooth newts Triturus vulgaris colonized following the removal of sticklebacks Gasterosteus aculeatus using rotenone. Larvae of both species were observed in the pond two months after treatment. Released toad tadpoles survived and metamorphosed in the pond. The concrete tank had sloping walls and a water depth of 90 cm. It contained approximately 2,000–3,000 sticklebacks. Rotenone was applied (5%; 0.2 mg/L) in May 1992 and seven days later the pond was dredged to remove dead fish. Over 100 toad tadpoles were then released into the pond. Aquatic plants were also introduced.




A controlled study in 1977–1984 in two lakes in south western Sweden (Aronsson & Stenson 1995) found that fish elimination using rotenone resulted in a rapid increase in the smooth newt Triturus vulgaris population. Newts colonized within two years of fish removal. Between 1977 and 1980 the breeding population increased from 2,000 to almost 10,000 individuals. Following fish stocking in 1979 with 2,000 roach Rutilus rutilus, newt numbers declined to below 900 by 1984. No newts were found in an adjacent (50 m) lake without fish removal. Rotenone was applied in 1973. Newts were sampled using a capture-recapture survey from May to June in 1977–1984. Forty-two cage traps were uniformly distributed around the removal lake. Traps were set in the untreated lake from 1978–1983.



A review of fish control programmes from 1992 to 1998 of two ponds in England, UK and one in Australia and Alabama, USA (Watson 2002) found that breeding success increased for dusky gopher frogs Rana sevosa, green and golden bell frogs Litoria aurea, great crested newts Triturus cristatus and smooth newts Lissotriton vulgaris. Egg masses of the gopher frogs increased from 10 to 150. At one site in England both newt species re-colonized and reproduced in a treated pond in the first year following stickleback (Gasterosteidae) elimination (2,000–3,000 fish). At the second site in England, although great crested newt adults and eggs were recorded following stickleback removal, no larvae were seen. Fish were recorded at two of the sites within a few years of treatment. At the first English site, rotenone (5%) was applied, dredge netting undertaken and aquatic plants introduced to an isolated concrete pond (104 m2) in May 1992. At the other site, rotenone and electrofishing were undertaken in 1996. In Alabama a pond was drained, fish removed and rotenone added in 1992. On Kooragang Island, Australia, rotenone was added to a pond to remove non-native plague minnows Gambusia holbrooki in 1998.



A replicated, before-and-after site comparison study in 2000–2002 of four ponds in a Nature Preserve in Illinois, USA (Mullin, Towey & Szafoni 2004) found that amphibian abundance and recruitment increased after fish control using rotenone (see also Towey 2007, Walston & Mullin 2007). Overall, numbers of amphibians increased by 411% in the two treated ponds compared to 165% in two existing fishless ponds. Recruitment increased by 873% in treated and 219% in historically fishless ponds. Abundance increases were greater in treated compared to fishless ponds for smallmouth salamanders Ambystoma texanum (610 vs 82%), American toad Bufo americanus (206 vs 190%), bullfrog Rana catesbeiana (101 vs 40%) and southern leopard frog Lithobates sphenocephalus (950 vs 325%). Wood frog Rana sylvatica increased by the same amount in treatment and controls (188 vs 188). Rotenone was applied to the two ponds (3–7 parts per million) with introduced native fish in December 2001. Amphibians were monitored in these two ponds and two without fish by using drift-fencing and pitfall traps from May 2000 to December 2002. Call surveys were also undertaken.



A replicated study in 2005–2006 of 39 ponds in a nature reserve in England, UK (Piec 2006) found that rotenone application to eliminate sticklebacks Pungitius pungitius had a direct negative effect on a small number of newts at the time of application. Nine great crested newts Triturus cristatus (one adult; eight larvae) and 12 smooth newts Triturus vulgaris (seven adult; five larvae) were negatively affected, 19 from one pond. Additional newts were potentially affected but not found. Eight of the affected newts (38%; 5 crested newts) survived a 48-hour observation period in clean water and were released into nearby untreated ponds. Populations in the nature reserve were estimated at 30,000 adult great crested newts and several thousand smooth newts. Rotenone was applied (2.5%; 3 parts per million) in December 2005 using sprayers. Seventeen ponds received a second application (2 parts per million) in January 2006. Most ponds were hand netted prior to treatment in an attempt to remove newts; 14 newts were found in five ponds.



A continuation of a replicated, before-and-after, site comparison study (Mullin, Towey & Szafoni 2004) in 2000–2002 (Towey 2007) found that recruitment of three amphibian species increased after fish elimination using rotenone (see also Walston & Mullin 2007). Recruitment (emerging metamorphs per breeding adult) increased significantly for smallmouth salamanders Ambystoma texanum (from 0 to 1–11), wood frog Rana sylvatica (0 to 1–2) and in one of two ponds American toad Bufo americanus (0 to 15). Recruitment tended to become higher than in two historically fishless ponds (salamanders: 0–1; wood frog: 0–0.5; American toad: 1–10). Numbers of emerging metamorphs increased significantly at experimental ponds for salamanders (0 to 20–205), wood frog (0–2 to 2–15) and American bullfrog Rana catesbeiana (35–42 to 47–50), but not American toad (0–2500 to 100–1700). Numbers of adults captured did not differ with treatment in experimental (before: 2–24; after: 5–44) and fishless ponds (before: 4–68; after: 16–84), apart from American toad which decreased in treatment ponds (before: 20–130; after: 2–80). Amphibians were monitored before (2001) and after (2002) treatment using drift-fencing with pitfall traps (7.5 m apart). Fish were eliminated, apart from bullhead catfish Ameiurus melas in one pond.



A continuation of a replicated, before-and-after, site comparison study (Mullin, Towey & Szafoni 2004, Towey 2007) in 2001–2004 (Walston & Mullin 2007) found that amphibian diversity and smallmouth salamander recruitment increased significantly after fish elimination using rotenone. Species relative abundance increased from 0.2 to 0.7 and became similar to that in historically fishless ponds (0.5–0.6). Small-mouth salamanders became the most abundant species in both treatment (41%) and fishless ponds (54%). American toad had been most abundant before fish removal (treatment: 91%; fishless: 67%). Although fish elimination did not result in increased salamander size at metamorphosis (42 vs 37 mm), it resulted in a significantly longer larval period (12% increase) and increased reproductive success (proportion of juveniles to breeding females: 0.3 vs 16.0). In fishless ponds larval period decreased 7% and recruitment was similar (0.2 vs 2.5). Numbers of juveniles increased significantly in treated (12 to 861) and fishless ponds (29 to 400). Amphibians were monitored before (2001) and after (2002–2004) treatment. One pond received a second application of rotenone to eliminate black bullhead catfish Ameiurus melas in January 2003.


Referenced papers

Please cite as:

Smith, R.K., Meredith, H. & Sutherland, W.J. (2018) Amphibian Conservation. Pages 9-65 in: W.J. Sutherland, L.V. Dicks, N. Ockendon, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2018. Open Book Publishers, Cambridge, UK.