Poppies: Feed or 'No Feed'?

The Tortoise Mythbusters - Episode 8

Poppies are often listed as a 'no feed', but certainly some species are eaten by tortoises in the wild. We take a look into this and try to explain what is going on.

Poppies - 'feed or no feed' ? It's not quite that simple... firstly, there are different species of poppies and tortoises appear to have distinct preferences, even when more than one species occurs in the same habitat. While there is general agreement that Welsh poppies (Meconopsis cambrica) and California poppies (Eschscholzia californica) are safe to feed in moderation for tortoises, there is much debate on some of the other genera and species, especially the archetypical 'Papaver' family.

Papaver hybridium

Papaver hybridium flower

Papaver rhoeos flower

So, here we have two species regularly found in Testudo graeca and Testudo hermanni hermanni habitats. Papaver hybridum (the 'rough poppy') and Papaver rhoeas (commonly known as the 'field poppy', 'corn poppy' or 'Flanders poppy'). Tortoises do not appear to eat the former, bypassing it when grazing, however, they will eat Papaver rhoeas enthusiastically and sometimes in quite large quantities. It is impossible to know for sure why this is so. Something about the taste? Or an instictive or learned survival mechanism? We just don't know for certain.

Despite the fact that the species of poppy that is commonly eaten by tortoises (Papaver rhoeas) being described as "causing acute liver toxicity and intoxication with different effects (nausea, restlessness, dyspnoea, contractions unconsciousness, numbness, etc.)", that of course, is based upon the reaction of humans and mammals, however, and it cannot necessarily be assumed to have such effects upon tortoises.

A Testudo hermanni hermanni grazes in a field rich with Papaver rhoeas in Northern Spain.

Different species of animals vary widely in their susceptibilities to individual plant toxins. Most lists of toxic plants presented to reptile keepers are derived from studies (or experiences) based upon humans and commercial livestock. Although such lists are useful, they are by no means entirely reliable when applied to reptiles. These different responses are explained by fundamental differences in diet, by differences in the anatomy and chemistry of digestive tracts, and by differences in feeding behaviours. Certainly, one major factor that influences the effect of ingestion of potentially toxic plant materials is the herbivorous reptiles reliance upon microbial digestion. This is paralleled in mammals by ruminants such as cattle, goats and sheep. Species that utilise a primarily microbial mode of digestion, whether mammalian or reptilian, possess a large fore stomach, or hindgut in the case of reptiles, where toxins become well diluted and are absorbed relatively slowly. The rate of absorption in such species is typically far slower than in non-ruminants such as pigs, dogs, cats, horses or human beings. Microbial action in the rumen (or hindgut) can also destroy certain toxic plant compounds, although, conversely, it can also have a negative effect by promoting the release of toxic constituents as plant cellular material is degraded and 'bound in' toxins released.

The most important factors in species differences in plant toxin susceptibility are probably the detoxification mechanisms specific to each species. These mechanisms can even vary within animals of the same species, or between different geographical populations of the same species, e.g., some (but not all) rabbits possess an enzyme known as atropinesterase that counters the effects of deadly nightshade; goats also possess an atropinesterase that makes them less susceptible to poisoning by the same plant. Deer are said to be able to feed on yew and rhododendron, and grey squirrels and American box turtles upon the deadly Amanita mushroom. Galapagos tortoises also consume highly toxic fruits in quantities sufficient to kill a horse. Among lizards, the ground iguanas Cyclura cornuta and C. carinata are known to feed extensively upon the fruits and leaves of the manchineel (Hippomane mancinella) and the fruit of the poisonwood tree (Metopium toxiferum), both of which contain powerful alkaloid toxins (Wiewandt, 1977 and Iverson, 1979).

Although it is commonly assumed that goats can feed with impunity upon almost anything, this is not true. Under natural conditions, the feeding pattern of the flock is determined by its leader, so that identical plants are eaten by all members at the same sites. Flock members learn to avoid the plants actively discriminated against by their leader – goats not kept as a flock, or kids separated from their mothers at an early age show little discrimination and are far more likely to succumb to plant poisoning. It is not known how widely learned toxin avoidance behaviour is in reptiles.

This indicates that lists based upon mammalian reactions are absolutely not entirely reliable when applied to herbivorous reptiles.

Most 'no feed' lists often also fail to discriminate between differing toxicity properties of separate plant parts in sufficient detail; it is well known that potato and tomato leaves are inedible, while the fruits and tubers are edible – but it is not so often realised that many plants vary in toxicity seasonally, with toxin levels in some plants rising dramatically during episodes of drought. Even altitude can affect toxin levels in some plant species. In nature, the animal’s feeding behavior often varies to accommodate these changes or differences, for example, Dipsosaurus dorsalis consumes only the flowers of the creosote bush (Larea divaricata) and ignores the toxic leaves (Mares and Hulse, 1978). In captivity, however, it is clear that reptiles are not necessarily able to discern dangerous plant toxins if offered unknown varieties – for example, tortoises have died as a result of eating daffodils and green iguanas have died after consuming azalea and oleander - therefore, although 'poison plant lists' are technically somewhat inadequate, they should certainly send a clear warning signal unless specifically contradicted by other evidence.

Testudo horsfieldii

Testudo horsfieldii (the Russian tortoise) is another tortoise that is known to consume poppy species extensively (Papaver pavoninum - which also highly toxic to mammals). Indeed, one paper concluded that "Steppe tortoises avoid feeding on grass and feed mostly on plant species that are usually highly toxic to mammals, suggesting a limited (if any) competition for food between tortoises and wild or domestic ungulates" (Legarde, et al, 2003).

Interestingly, wild buttercups also form, at certain times of year, a very high proportion of the wild diet of this same species. This is another plant that is widely listed on 'no feed' lists based upon its toxicity to mammals.

The main basis for this is a single paper published in 1979 (Holt, et. al. ) this reported upon one case where a tortoise died shortly after eating buttercups (Ranunculus sp.), and noted that "there was a 1.5 cm diameter, hemorrhagic area in the mucosa of the greater curvature of the gastric fundus. No other lesions were found. The owners admitted feeding large numbers of buttercup flowers to the tortoise immediately prior to death and the gastric lesions were similar to those described in buttercup poisoning cases in mammals". However, subsequently it has proved difficult to locate any similar cases. Also, many, many keepers have reported that their tortoises routinely consumed buttercups over several decades with no apparent ill effect whatever. We also, at the Tortoise Trust in Wales, UK, had quite a large number of buttercups appear seasonally, in many of the Testudo pens and tortoises were observed eating them. Again, over the course of 10+ years no ill effects were observed and no mortalities occurred. There are numerous reports of toxicity in humans and mammals, but it is far from clear if they really are a problem to tortoises. Interestingly, one report noted that Testudo hermanni hermanni kept in semi-captivity in France that had high parasitic loads preferably selected plants of the Ranunculaceae family, possibly due to its natural 'vermifuge' effects. (Longepierre,et. al., 1999).

Practical summary

It is very clear that most 'no feed' lists are far from entirely accurate or reliable. There are also broad generalisations that are certainly not always true. For example it is often said that tortoises never fraze upon anything that 'grows from a bulb' - however, Testudo graeca graeca in both Spain and North Africa do sometimes eat the leaves of Allium triquetrum the 'three cornered garlic'. Obviously there are some highly toxic plants with proven damaging or even lethal effects upon tortoises (Daffodil is one of these), but equally, we also find wild tortoises regularly consuming quite large amounts of other plants that appear on such lists. In this case, keepers should certainly not panic or rush to exclude these particular 'exceptions to the rule' from the diet. If there are some poppies, or buttercup for example, in the tortoise enclosure, we would leave them there. We would not advise deliberately feeding them, but if they are there, the realistic chances that they will cause harm are absolutely minimal.

References and further reading:

Holt, P. E., Cooper, J. E., Needham, J. R. Diseases of Tortoises: A review of seventy cases. Journal of Small Animal Practice, 20:269, 1979.

Iverson, J. B. (1979) Behaviour and Ecology of the Rock Iguana, Cyclura carinata. Ph.D. dissertation, Univ. Florida, Gainsville.

Longepierre, S. & C. Grenot (1999): Some effects of intestinal nematodes on the plant foraging behaviour of Testudo hermanni in the south of France. In: Current Studies in Herpetology (eds. C. Miaud & R. Guyetant). SEH, Le Bourget du Lac, pp. 277-284.

Mares, M. A. and Hulse, A. C. (1978) Patterns of some vertebrate communities in creosote bush deserts. In: Creosote Bush: Biology and Chemistry of Larrea in New World Deserts. Dowden, Hutchinson and Ross, Stroudsburg, Pa.

Wiewandt, T.A. (1977) Ecology, behavior, and management of the Mona Island ground iguana Cyclura stejnegeri. Ph.D. dissertation, Cornell Univ. Ithaca, NY.

Frederic Lagarde, Xavier Bonnet, Johanna Corbin, Brian Henen, Ken Nagy, Baktjior Mardonov and Guy Naulleau. Foraging behaviour and diet of an ectothermic herbivore: Testudo horsfieldii. ECOGRAPHY 26: 236–242, 2003

To receive updates and notifications of new articles, videos and other important news please consider 'Subscribing' to the Tortoise Trust. There is even a free option available. This supports our work and allows us to make this information available. The Tortoise Trust also offers comprehensive online courses that cover all aspects of behaviour, ecology and captive care.

(c) 2024 A. C. Highfield/Tortoise Trust