N5
2015 Southeastern Naturalist Notes Vol. 14, No. 1
R.L. Hill, J.R. Mendelson III, and J.L. Stabile
Direct Observation and Review of Herbivory in Sirenidae
(Amphibia: Caudata)
Robert L. Hill1, Joseph R. Mendelson III1,*, and Jennifer L. Stabile2,3
Abstract - We provide the first direct observation of intentional herbivory in sirenid salamanders and
provide a literature summary that indicates that the phenomenon has been well documented, but often
misinterpreted. Here we report direct observation of herbivory in Siren lactertina (Greater Siren)
and Pseudobranhcus axanthus (Southern Dwarf Siren) and review indirect evidence of the same for
S. intermedia (Lesser Siren). We believe that at least two species of sirenid salamanders are facultative
herbivores, the only salamanders known to show such behavior, and this behavior may apply to
all known species of Sirenidae.
Introduction. The two species of the genus Siren are large (TL to >900 mm), fully
aquatic salamanders native to the midwestern and southeastern regions of the United
States, including Texas and the Rio Grande drainage (Petranka 1998). Sirenid salamanders
lack rear limbs, have an elongate eel-like body, and retain several larval characteristics
as adults. While both adult-form and larval salamanders generally are accepted to be carnivorous
(Wells 2007:741), the diets of wild Siren lacertina österdam (Greater Siren) and
S. intermedia Barnes (Lesser Siren) have been reported numerous times and include various
invertebrates, small vertebrates, and substantial amounts of plant matter and muddy
substrate material (Bennett and Taylor 1968, Davis and Knapp 1953, Luhring 2007, Noble
and Marshall 1932, Scroggin and Davis 1956). Dunn (1924) reported observations of large
quantities of plant material in the guts of a series of field-collected S. lacertina and suggested
that the animals are at least partially herbivorous. Nevertheless, subsequent reports and
reviews generally have qualified their observations of plant material in the gastrointestinal
tracts of wild-caught Siren spp. with the caveat that the material passes undigested through
the digestive tract (Davis and Knapp 1953) and with the implied or explicit notion that the
plant material is ingested incidental to capture of animal prey (e.g., Noble and Marshall
1932, Scroggin and Davis 1956). The diet of Pseudobranchus axanthus Netting and Goin
(Southern Dwarf Siren) has been poorly studied, but all reports include only invertebrate
prey and captive individuals consuming eggs of conspecifics (reviewed by Petranka 1998);
there are no reports on the diet of Pseudobranchus striatus (LeConte) (Northern Dwarf
Siren). Here we report direct observation of captive sirenid salamanders purposefully
ingesting plant material and make the argument that at least some members of the family
should be properly classified as facultative herbivores, or simp ly omnivores.
Observations. Our initial observation was of a large adult Greater Siren in an exhibit
enclosure at Zoo Atlanta. The exhibit contains a variety of live angiosperm plants, green
algae, small rocks, and submerged wooden props. The exhibit also houses two Sternotherus
odoratus (Latreille in Sonnini and Latreille) (Eastern Musk Turtle). This new exhibit enclosure
was constructed in July 2013, and standard routine observation of animal welfare was
conducted after animals were introduced.
During the observation period one of the authors (R.L. Hill) observed the siren consuming
a mat of green algae that had become dislodged from a submerged wooden prop. The
1Zoo Atlanta, 800 Cherokee Avenue SE, Atlanta, GA 30315. 2Albuquerque BioPark, 903 10th Street
SW, Albuquerque, NM 87102.3Current address - San Antonio Zoo, 3903 North Saint Mary’s Street,
San Antonio, TX 78212. *Corresponding author - jmendelson@zooatlanta.org.
Manuscript Editor: Max Nickerson
Notes of the Southeastern Naturalist, Issue 14/1, 2015
2015 Southeastern Naturalist Notes Vol. 14, No. 1
N6
R.L. Hill, J.R. Mendelson III, and J.L. Stabile
individual was observed initially to be foraging through the sand substrate (consistent with
observations by Altig 1967), although we could not determine if any substrate was actually
ingested. During this bout of foraging, the animal came in contact with the algal mat and
immediately took it into its mouth, masticated it momentarily, spit it out, then immediately
consumed it again with more mastication and finally swallowed it entirely. There were no
invertebrates inhabiting this exhibit at that time, so there is little chance that the algal mat
was ingested as a means of ingesting invertebrates (e.g., snails) that may have been suspended
in it.
A few days later, we introduced a large piece of boiled romaine lettuce into the enclosure,
held in a fully submerged position by a piece of wood. The next day we observed that
about half of the piece of lettuce had been consumed overnight, and the entire piece had
been consumed after a second night. Nevertheless, we could not confirm that the salamander—
rather than the turtles—had consumed the plant material.
Similar trials were conducted with another individual Greater Siren at Albuquerque
BioPark, in the absence of other animals, and documented with night-vision videography.
Footage obtained on multiple nights indicated intentional consumption of plant material by
the Greater Siren in the absence of any animal prey. In these trials, several large pieces of
boiled iceberg lettuce were introduced to the enclosure. Immediately upon placing the vegetation
in the enclosure, the animal was observed rapidly foraging through the sand similar
to observations by Zoo Atlanta. During these foraging sessions, the individual was observed
consuming all of the lettuce resting on the substrate. Several pieces of frozen fingerling trout
had been consumed by the Greater Siren the day before introducing the lettuce, potentially
ruling out hunger as a possible reason for ingesting the vegetation. Trial offerings of other
plants (Echionodorus sp., Bacopa caroliniana (Walter) B.L. Rob [Blue Waterhyssop], and
uncooked romaine) at Albuquerque BioPark were not successful; the animal showed some
interest in moving the plant material with its snout, but did n ot ingest the plants.
Subsequent overnight trials with the animal at Zoo Atlanta, in the absence of the turtles,
also indicated intentional plant consumption. Additionally, we offered pieces of boiled romaine
lettuce with forceps to the Zoo Atlanta specimen during daylight hours. The animal
showed great interest and consumed multiple pieces of lettuce without hesitation (Fig. 1).
The Greater Siren was not able to consume one especially large piece whole, but did attempt
on numerous occasions to masticate and swallow it. However, all pieces smaller than this
were readily consumed completely. As this specimen is fed other food items from forceps
on occasion, we cannot rule out a conditioned response to the forceps rather than the food
item for this particular trial.
In a trial with a group of Southern Dwarf Siren at Albuquerque BioPark, an adult specimen
initially began exploring a piece of boiled lettuce by moving it about with its snout,
eventually taking a small piece into its mouth and masticating, but it did not appear to completely
ingest it. Subsequently a juvenile in the same enclosure emerged, foraged around
and beneath the piece of lettuce, and then was observed to directly ingest a piece of lettuce.
We note particularly that, typical of animals held in zoos, the individuals observed in our
study were in very good body condition and were not starved nor food-deprived in any way
that may have elicited feeding responses to non-preferred food items. In no instance during
trials or routine maintenance did we observe undigested plant materials passed by study
animals at either institution.
Discussion. Our direct observations of intentional consumption of plant material, in the
absence of animal prey, by Greater Siren and Southern Dwarf Siren are unambiguous and
support the original declaration by Dunn (1924) that Siren spp. may be at least partially
N7
2015 Southeastern Naturalist Notes Vol. 14, No. 1
R.L. Hill, J.R. Mendelson III, and J.L. Stabile
herbivorous. We emphasize that most of our observations involved the animals ingesting
the plant material directly from the substrate, and do not represent a conditioned feeding
response to any item offered on forceps. All animals observed in these trials were long-term
captives and maintained on a general carnivore diet consisting of crayfish, bloodworms,
earthworms, small fish, and commercial aquatic turtle pellets.
Ironically, evidence for herbivory in the salamander genus Siren has been well documented,
but evidently understated, with the exception of Dunn (1924), as other authors
claimed that plant material in the guts of Siren spp. was incidental to capture of animal
prey. Certainly, given what is known of the ecology of sirenid salamanders, it is possible
and probable that capture of animal prey does sometimes result in the incidental ingestion of
plant material. Our observations indicate that plant material is also intentionally consumed
by sirens in captivity. Thus, the premise of herbivory in these animals being inconsequential
nutritionally and entirely secondary to their putative carnivor ous diet is doubtful.
Altig (1967) directly observed captive Siren ingesting substrate; his observations are
consistent with multiple observations of muddy substrate in the intestines of Siren spp.
Consumption of muddy substrate is consistent with detrivory, in contrast with presumed
herbivory, as described by Altig et al. (2007). Hanlin (1978:360) provided a secondhand
observation of a Greater Siren directly consuming plant material (Elodea). Ultsch
(1973:305) also reported a similar second-hand observation and further stated that “The
possibility that plant material may be utilizable should not be discounted until enzymatic
studies have been done, and the possibility of gut symbionts has been ruled out.” Pryor et
al. (2006) addressed that notion and described the capacity for hindgut fermentation of plant
material in Greater Sirens, thus providing the requisite information suggested by Ultsch
(1973). But Pryor et al. (2006) did not comment on the purposeful ingestion of plant material
in the absence of coincidental animal prey.
Figure. 1. An adult Siren lacertina (Greater Siren) in the collection at Zoo Atlanta consuming boiled
romaine lettuce offered to it with forceps. Photograph © David Brothers.
2015 Southeastern Naturalist Notes Vol. 14, No. 1
N8
R.L. Hill, J.R. Mendelson III, and J.L. Stabile
We note that Scroggin and Davis (1956) reported that gut contents in their study contained
a higher volume of plant material than invertebrates and that Pryor et al. (2006)
described the intestinal anatomy and microbial content to be similar to that of some herbivorous
fishes. The observations by Davis and Knapp (1953) that the plant material in the
intestines of their sample animals “… appeared to be unaffected by digestive fluids” and the
statement by Scroggin and Davis (1956) that “the anatomy of the digestive tract of Siren belies
the statement that it is herbivorous” are in contrast to our own reported here, and are not
supported by the physiological and anatomical determinations of Pryor et al. (2006). Hanlin
(1978) opined that some plant material he observed in the intestines of S. lacertina appeared
to be partially digested, but he reinforced the concept that plant material was ingested in
direct association with animal prey. While previous authors have emphasized that ingestion
of plant material by sirens may be secondarily related to their pursuit of macroinvertebrates
(e.g., snails) situated on the plant material, we note that there typically also would be a
wealth of micro-organisms (e.g., fungi or bacteria) also situated on the surfaces of the plants
that could be a source of nutrition for the sirens as well (i.e., the epilithon described by
Altig et al. [2007]). Consumption of plant material as a means of ingesting epilithon does
not constitute herbivory, and specific studies would be required to determine to what extent
sirenids are digesting any epilithon. Regardless of their ability to digest epilithon, Pryor et
al. (2006) demonstrated that sirenids likely can digest plant material, and our observations
that undigested plant material does not pass through the digest ive tract indicate herbivory.
We have documented intentional herbivory in the absence of animal prey in captive
Greater Sirens and Southern Dwarf Siren. In conjunction with a review of the literature
regarding Lesser Sirens (Bennett and Taylor 1968, Davis and Knapp, 1953, Noble and Marshall
1932, Scroggin and Davis 1956) and reasonable conjecture regarding Northern Dwarf
Sirens, we suggest the possibility that all four species of sirenid salamanders may be facultative
herbivores, making them the only known herbivorous or omnivorous salamanders.
Thus, among amphibians, we posit that at least 2, and perhaps all, species of sirenids should
be considered alongside the Brazilian hylid frog Xenohyla truncata (Izecksohn) (da Silva
and de Britto-Pereira 2006, see also Das 1996) as the only known facultative non-larval herbivorous
amphibians. While our first-hand observations of herbivory in these salamanders
are novel (but see pers. comm. reported by Hanlin 1978, and Ultsch 1973), our review of
the literature indicates that the phenomenon was noted but understated in previous studies,
with the exception of the initial proposition of herbivory in a salamander by herpetologist
E.R. Dunn in 1924.
Acknowledgments. We are grateful to D. Brothers for photographic assistance and to R.
Altig and H.R. da Silva for comments on the manuscript.
Literature Cited
Altig, R. 1967. Siren intermedia nettingi in a spring-fed swamp in southern Illinois. American Midland
Naturalist 77:239–241.
Altig, R., M.R. Whiles, and C.L. Taylor. 2007. What do tadpoles really eat? Assessing the trophic
status of an understudied and imperiled group of consumers in freshwater habitats. Freshwater
Biology 52:386–395.
Bennett, C., and R.J. Taylor. 1968. Notes on the Lesser Siren, Siren intermedia (Urodela). Southwestern
Naturalist 13:455–457.
da Silva, H.R., and M.C. de Britto-Pereira. 2006. How much fruit do fruit-eating frogs eat? An investigation
on the diet of Xenohyla truncata (Lissamphibia: Anura: Hylidae). Journal of Zoology
270:692–698.
Das, I. 1996. Folivory and seasonal changes in diet in Rana hexadactyla (Anura:Ranidae). Journal of
Zoology 238:785–794.
N9
2015 Southeastern Naturalist Notes Vol. 14, No. 1
R.L. Hill, J.R. Mendelson III, and J.L. Stabile
Davis, W.B., and F.T. Knapp. 1953. Notes on the salamander Siren intermedia. Copeia 1953:119–121.
Dunn, E.R. 1924. Siren: A herbivorous salamander? Science 59:145.
Hanlin, H.G. 1978. Food habits of the Greater Siren, Siren lacertina, in an Alabama Coastal Plain
pond. Copeia 1978:358–360.
Luhring, T.M. 2007. Siren lacertina (Greater Siren) diet. Herpetological Review 38:317.
Noble, G.K., and B.C. Marshall. 1932. The validity of Siren intermedia LeConte, with observations
on its life history. American Museum Novitates 532:1–17.
Petranka, J.W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press,
Washington, DC. 587 pp.
Pryor, G.S., D.P. German, and K.A. Bjorndal. 2006. Gastrointestinal fermentation in Greater Sirens
(Siren lacertina). Journal of Herpetology 40:112–117.
Scroggin, J.B., and W.B. Davis. 1956. Food habits of the Texas Dwarf Siren. Herpetologica
12:231–237.
Ultsch, G.R. 1973. Observations on the life history of Siren lacertina. Herpetologica 29:304–305.
Wells, K.D. 2007. The Ecology and Behavior of Amphibians. University of Chicago Press, Chicago,
IL. 1148 pp.