Timing of development of the middle ear of Anura (Amphibia)

Summary The opercularis system and tympanum-stapes complex of the anuran middle ear develop at different times relative to metamorphosis. In early larvae, the fenestra ovalis is represented by a large lateral opening in the otic capsule filled with connective tissue. At later larval stages, but well before metamorphosis, a cartilaginous operculum begins to form at the posterior margin of the fenestra ovalis, and proceeds to expand to fill all except the anterior part of the fenestra. The opercularis muscle forms along with the levator scapulae superior muscle at the anteromedial edge of the developing suprascapular cartilage of the shoulder girdle. The muscle fibers extend anteroventrally towards the operculum and otic capsule, and, just before emergence of the forelimbs, that portion that will form the opercularis muscle inserts on the lateral surface of the operculum. At this stage, when the metamorphosing frogs first show terrestrial habits, the opercularis system is complete and presumably functional. Timing of development of the tympanum-stapes complex is more variable. The stapes begins as a cartilaginous condensation in the anterior part of the fenestra ovalis, and develops laterally to eventually contact the epidermis and dermis that together will form the tympanum. Meanwhile a middle ear cavity and tympanic annulus form to complete the complex. In several species, especially those that metamorphose at a smaller body size, the tympanum-stapes complex is quite incomplete by the end of metamorphosis, and in Hyla crucifer it takes about 60 days to fully develop. The presence of a complete opercularis system by the start of terrestrial activity is consistent with an hypothesized seismic function of the system. The independent timing of development of the opercularis system and tympanum-stapes complex does not support functional hypotheses linking the opercularis system with modulation of responsiveness of the tympanum-stapes complex to aerial sound. Newly metamorphosed frogs with poorly developed tympanum-stapes complexes are presumably either insensitive to aerial sound or employ alternate mechanisms for transmission of sound energy to the inner ear, possibly involving the opercularis system.     

Trait independence primes convergent trait loss

The repeated, independent evolution of traits (convergent evolution) is often attributed to shared environmental selection pressures. However, developmental dependencies among traits can limit the phenotypic variation available to selection and bias evolutionary outcomes. Here, we determine how changes in developmentally correlated traits may impact convergent loss of the tympanic middle ear, a highly labile trait within toads that currently lack adaptive explanation. The middle ear's lability could reflect evolutionary trade‐offs with other skull features under selection, or the middle ear may evolve independently of the rest of the skull, allowing it to be modified by active or passive processes without pleiotropic trade‐offs with other skull features. We compare the skulls of 55 species (39 eared, 16 earless) within the family Bufonidae, spanning six hypothesized independent middle ear transitions. We test whether shared or lineage‐specific changes in skull shape distinguish earless species from eared species and whether earless skulls lack other late‐forming skull bones. We find no evidence for pleiotropic trade‐offs between the middle ear and other skull structures. Instead, middle ear loss in anurans may provide a rare example of developmental independence contributing to evolutionary lability of a sensory system.     

Biomechanical and neurophysiological studies on audition in eared and earless harlequin frogs (Atelopus)

Tissue displacement of various body surfaces and the auditory midbrain sensitivities to sound were measured in Atelopus species with or without a tympanic middle ear (“eared” and “earless”, respectively). Tissue displacement (vibration) of body regions was measured by laser Doppler vibrometer . The body wall directly overlying the lung is most dramatically displaced by sound pressure in all species tested. The otic (lateral head) region showed low displacement in earless species, but significant displacement to high-frequency sound in eared species. Peak tissue displacement of the body wall occurred within the frequency range of each species' advertisement vocalization. Peak tissue displacement of the otic region of the eared species also occurred within these frequencies. Multi-unit neurophysiological recordings of the auditory midbrain (torus semicircularis) also were obtained. Auditory sensitivity curves showed three distinct regions of sensitivity at low, middle, and high frequencies, the latter located within the frequency range of each species' advertisement vocalization. The correlation between auditory midbrain sensitivity and tissue displacement of the body wall region at advertisement vocalization frequencies, suggests that the body wall/lungs serve as the route of sound transfer to the inner ear in earless species and possibly in the eared species as well. Accepted: 4 April 1998     

Comparative morphology of the amphibian opercularis system: I. General design features and functional interpretation

The morphology of the opercularis system of anuran and caudate amphibians suggests that it acts to produce motion of the operculum that in turn produces fluid motion within the inner ear. The operculum and opercularis muscle form a lever system, with a narrow connection between the operculum and otic capsule acting as a fulcrum about which the operculum moves in response to forces applied via the muscle. The opercula of many species possess a muscular process on which the muscle inserts, thereby increasing the moment arm through which the muscle acts. The tonicity of the opercularis muscle allows tensile forces produced by substrate vibration or other mechanical energy applied to the forelimb to be effectively transmitted to the operculum; the elasticity of the connective tissue holding the operculum in place should act to return the operculum to its original position. The opercularis systems of frogs and non-plethodontid salamanders are similar structurally and functionally; that of plethodontid salamanders is structurally distinct but also functions as a lever system. Fluid motion produced by opercular motion could stimulate various end organs of the inner ear; the saccule, lagena, and amphibian papilla are in close approximation and wave energy could directly affect their otoconial or tectorial structures. In those anurans with a tympanic ear, the stapedial footplate and operculum articulate, but this articulation allows both to move independently. The stapes-tympanum complex and opercularis system therefore appear to be independent functional systems, and it is unlikely that the opercularis system modulates middle ear responsiveness. The general design of the opercularis system is consistent with a function in reception of substrate vibrations.     

Middle-ear development: II. Morphometric changes in the conducting apparatus of the chick.

The ontogeny of various middle-ear structures was examined in 11 groups of chicks between 10 days embryonic and adult. Measurements of the tympanic membrane surface area and height, columella length, and that of the columella footplate, annular ligament, and oval window area were obtained using video micrographs and computer digitization techniques. The oval window matures first at 53 days post-hatching, whereas the columella achieves adult size at 74 days. The tympanic membrane surface area is the last middle-ear variable studied to reach adult size (79 days post-hatch). The columella increases its length from 0.63 mm (10 days embryonic) to 2.73 mm in the adult. The tympanic membrane area expands by 280% whereas the columellar footplate area increases by 11x. As a result, the pressure amplification of the middle ear due to the tympanic membrane/columellar footplate area ratio improves by over 400%. These data further contribute to our understanding of the functional development of the middle ear.     

Analysis of chick (Gallus gallus) middle ear columella formation

Background

The chick middle ear bone, the columella, provides an accessible model in which to study the tissue and molecular interactions necessary for induction and patterning of the columella, as well as associated multiple aspects of endochondral ossification. These include mesenchymal condensation, chondrogenesis, ossification of the medial footplate and shaft, and joint formation between the persistent cartilage of the extracolumella and ossified columella. Middle and external ear defects are responsible for approximately 10% of congenital hearing defects. Thus, understanding the morphogenesis and the molecular mechanisms of the formation of the middle ear is important to understanding normal and abnormal development of this essential component of the hearing apparatus.

Results

The columella, which arises from proximal ectomesenchyme of the second pharyngeal arch, is induced and patterned in a dynamic multi-step process. From the footplate, which inserts into the inner ear oval window, the shaft spans the pneumatic middle ear cavity, and the extracolumella inserts into the tympanic membrane. Through marker gene and immunolabeling analysis, we have determined the onset of each stage in the columella's development, from condensation to ossification. Significantly, a single condensation with the putative shaft and extracolumella arms already distinguishable is observed shortly before initiation of five separate chondrogenic centers within these structures. Ossification begins later, with periosteum formation in the shaft and, unexpectedly, a separate periosteum in the footplate.

Conclusions

The data presented in this study document the spatiotemporal events leading to morphogenesis of the columella and middle ear structures and provide the first gene expression data for this region. These data identify candidate genes and facilitate future functional studies and elucidation of the molecular mechanisms of columella formation.     

Lung-based hearing in an “earless” anuran amphibian

The mechanisms of hearing in the fire-bellied toad Bombina orientalis, an “earless” species of amphibian that lacks a standard tympanic middle ear, were studied using laser Doppler vibrometric and neurophysiological techniques. Laser vibrometry demonstrated that the anterolateral body wall overlying the lung is much more responsive to sound than the lateral head surface overlying the inner ear. Covering the lateral body wall with silicone grease dramatically decreased auditory midbrain sensitivity at all frequencies examined, elevating thresholds by 20–25 dB. Filling the lungs with oxygenated saline produced similar decrements in hearing sensitivity, and both manipulations strongly suggest that the lung is the primary route of sound reception in this species. The precise route of transfer of sound energy from the body wall and lungs to the inner ear remains unclear. The lung-based hearing system of “earless” fire-bellied toads may represent the retention of the first auditory mechanism used by early tetrapod vertebrates for detection of airborne sound. Accepted: 10 December 1998     

Middle ear transmission in the grass frog, Rana temporaria

The anuran middle ear serves to transmit eardrum vibrations to the inner ear. In order to do this efficiently, the eardrum and middle ear must operate as an impedance transformer matching the low impedance of air to the higher impedance of the fluid-filled inner ear. In amniotes, one of the mechanisms used to achieve impedance transformation is to have the middle ear work as a force-amplifying lever system. Here, we present evidence that the grass frog middle ear also implements a lever system. The columellar footplate, which sits in the oval window, is firmly connected to the otic capsule along its ventral edge. Therefore, simple in-out movement of the columella is prevented while a rotational movement around the footplate's ventral edge is possible. The latter movement pattern was confirmed by laser vibrometry measurements of eardrum and footplate vibrations. The results showed that the footplate vibrations were 20–30 dB weaker than those of the eardrum and that the two structures vibrated 180° out of phase (at low frequencies). The lever ratio was approximately 6, i.e. somewhat higher than lever ratios reported for amniotes. Hence, the middle ear lever probably makes a significant contribution to impedance matching in frogs. Accepted: 1 July 1997     

Role of the opercularis muscle in seismic sensitivity in the bullfrog Rana catesbeiana

The inner ear of anuran amphibians appears to be exceptionally sensitive to substrate vibration. The opercularis system, consisting of an opercularis muscle running from the shoulder girdle to a movable, cartilaginous operculum lying next to the inner ear, has been hypothesized to be involved in driving these seismic responses. Removal of the opercularis muscle of adult bullfrogs, Rana catesbeiana, caused clear decreases in microphonic responses of the inner ear to vibrations from 20-250 Hz and 0.05-5.0 cm/sec2 accelerations. Degree of decrease in responsiveness was variable between individuals and between different frequencies of stimulation, ranging up to 90% reduction at certain frequencies and in certain specimens. Decreases were most marked at lower frequencies below about 50 Hz. Additional removal of the levator scapulae superior muscle, which runs alongside the opercularis muscle from the shoulder girdle to ventrolateral portions of the otic capsule, also tended to depress responses, although this effect was substantially less (generally less than 10%) and also less consistent. As the opercularis muscle appears to be derived from the levator scapulae musculature, it is speculated that primitively seismic sensitivity was enhanced by a muscular connection that could transmit motion from the forelimb to the otic region, responsiveness being further enhanced by the subsequent evolution of the specialized opercularis system.     

Competing tadpoles: Australian native frogs affect invasive cane toads (Rhinella marina) in natural waterbodies

The cane toad (Rhinella marina) is one of the most successful invasive species worldwide, and has caused significant negative impacts on Australian fauna. Experimental work in the laboratory and in mesocosms has shown that tadpoles of native frogs can affect survival, size at metamorphosis and duration of larval period of cane toad tadpoles. To test if these effects occur in nature, we conducted a field experiment using two temporary ponds where we set up enclosures with tadpoles of native green tree frogs (Litoria caerulea) and cane toads in treatments with a range of densities and combinations. The presence of green tree frog tadpoles significantly decreased the growth rate of toad tadpoles and increased the duration of their larval period in both ponds; in one pond, frog tadpoles also significantly reduced the body length and mass of metamorph toads. Toad tadpoles did not have any significant negative effects on green tree frog tadpoles, but there was strong intraspecific competition within the latter species: increased frog tadpole density resulted in increased larval period and reduced survival, growth rate and size at metamorphosis for frogs at one or both ponds. Our results are encouraging for the possibility of using native frogs as one component of an integrated approach to the biological control of cane toads.     

Histochemical studies on the amphibian opercularis muscle (Amphibia)

Summary Histochemical studies of the opercularis muscle of the bullfrog (Rana catesbeiana) and the tiger salamander (Ambystoma tigrinum) provide evidence that the opercularis muscle of anurans is a specialized, tonic portion of the levator scapulae superior muscle. Staining results for myosin adenosine triphosphatase (ATPase) and succinate dehydrogenase (SDH), combined with measurements of muscle fiber diameters, demonstrate that the opercularis/levator scapulae superior muscle mass of both the tiger salamander and bullfrog consists of an anterior tonic portion, a middle fast oxidative-glycolytic (FOG) twitch portion, and a posterior fast-glycolytic (FG) twitch portion. In R. catesbeiana the tonic fibers represent 57.3% of the fiber total and (because they have relatively narrow diameters) about 29% of the cross-sectional area of the muscle mass, and form that part of the muscle (=opercularis muscle) that inserts on the operculum. In Ambystoma the tonic fibers represent only 8.8% of the fiber total and represent about 4% of the cross-sectional area. In the tiger salamander, the entire levator scapulae superior muscle inserts on the operculum and therefore represents the opercularis muscle. The bullfrog differs from the tiger salamander, therefore, in that the anterior tonic part of the opercularis/levator scapulae superior complex is greatly enlarged and the insertion on the operculum is limited to these tonic fibers. No evidence of a columellar muscle was found in R. catesbeiana. Previous reports of one in this species and in other anurans may be based on the tripartite nature of the opercularis/levator scapulae superior muscle mass. The middle FOG portion of the muscle may have been considered a muscle distinct from the anterior tonic portion (=opercularis muscle) and the posterior FG portion.     

Biomechanics of vibration reception in the bullfrog,Rana catesbeiana

The opercularis system (OPS) of amphibians consists of an opercularis muscle that connects the shoulder girdle skeleton to the operculum, a movable element in the oval window of the otic capsule. The role of the OPS in reception of vibrations was examined in bullfrogs (Rana catesbeiana) tested in various postures that manipulated differential motion between the shoulder girdle (the origin of the opercularis muscle) and skull (including the inner ear). Amplitude and phase relationship of motions of the suprascapular cartilage of the shoulder girdle and the posterior skull were also measured during these tests. 1. Microphonic responses to vertical vibrations from 25-200 Hz were typically highest when frogs were in a normal, sitting posture with the head held off the vibrating platform. Responses from animals in which the head directly contacted the platform were often less (by up to 10 dB at certain frequencies). Responses from all test positions were highest at lower frequencies, especially between 50-100 Hz. 2. Suprascapular accelerations were typically highest in the normal, sitting posture, and at lower frequencies (50-75 Hz) were often greater than that of the vibrating platform by up to 8 dB. The shoulder girdle skeleton of the bullfrog is therefore readily affected by vertical substrate motion. 3. The amplitude of microphonic responses in the different test postures did not correspond well with head acceleration. Rather, response amplitude corresponded best with the absolute difference between shoulder and head motion. For example, in the normal posture, suprascapular motion was much greater than head motion, and responses were relatively high. If only the head was vibrated, head motion was high and shoulder motion low, and responses also were relatively high. If the head and body were vibrated together, their motions were similar, and responses to the same platform accelerations were often reduced. Phase differences between shoulder and head motions were small at the frequencies examined and may be of little functional significance. The importance of differences in shoulder and head motion suggests that the resulting differential motion of the operculum and inner ear fluids can produce waves that stimulate appropriate end organs (such as the saccule). 4. Removal of the opercularis muscle reduced responses up to 18 dB at certain frequencies in some of the test postures. The most significant reductions were observed in those postures with a significant difference between shoulder and head motion (such as the normal posture).(ABSTRACT TRUNCATED AT 400 WORDS)     

Differential Senescence of Maize Hybrids following Ear Removal : II. Selected Leaf

In conjunction with a study of the effects of ear removal on the senescence of whole maize (Zea mays L.) plants, visual symptoms and associated changes in constituent contents and activities of a selected leaf (first leaf above the ear) were determined. Leaves were sampled from field-grown eared and earless Pioneer brand 3382, B73 × Mo17, and Farm Services brand 854 maize hybrids at nine times during the grainfilling period.

Visual symptoms indicated the following sequence and rate of senescence: earless B73 × Mo17 > earless P3382 » eared B73 × Mo17 » eared P3382 ≤ earless FS854 > eared FS854. All earless hybrids showed increases in leaf dry weight and sugar content; however, the increases were transitory for P3382 and B73 × Mo17, but continuous throughout the grain-filling period for FS854, indicative of continued photosynthetic activity of the latter. All earless hybrids exhibited similar and transitory starch accumulation patterns. Thus, FS854 was an exception to the concept that carbohydrate accumulation accelerates leaf senescence. Ear removal resulted in accelerated losses of reduced N, phosphoenolpyruvate and ribulose bisphosphate carboxylases, phosphorus, chlorophyll, nitrate reductase activity, and moisture for P3382 and B73 × Mo17 plants. In contrast, the loss of all components (except phosphorus) was similar for the selected leaf of earless and eared FS854.

Although the loss of nitrate reductase activity, reduced N, and carboxylating enzymes accurately reflected the development of senescence of the selected leaf, the rate of net loss of reduced N and carboxylating enzymes appeared to be regulated. We deduced that the rate of flux of N into the leaf was a factor in regulating the differing rates of senescence observed for the six treatments; however, we cannot rule out the possibility of concurrent influence of growth regulators or other metabolites.

     

Auditory development in the mouse: Structural maturation of the middle ear

The development of middle-ear structures in the mouse was examined in nine groups of pups between 1 and 45 days of age. The area of the tympanic membrane (pars tensa and pars flaccida), the length of the lever arms of the malleus and incus, the surface area of the oval window, and the volume of the bulla all showed systematic changes during neonatal life. The area of the oval window reached maturity first and the lever arms achieved 90% of their adult size on day 11. The tympanic membrane achieved the same criterion on day 18. These data help us further to understand the processes that contribute to the functional ontogeny of the middle ear.     

An invasive species imposes selection on life‐history traits of a native frog

As well as their direct ecological impacts on native taxa, invasive species can impose selection on phenotypic attributes (morphology, physiology, behaviour, etc.) of the native fauna. In anurans, body size at metamorphosis is a critical life‐history trait: for most challenges faced by post‐metamorphic anurans, larger size at metamorphosis probably enhances survival. However, our studies on Australian frogs (Limnodynastes convexiusculus) show that this pattern can be reversed by the arrival of an invasive species. When metamorph frogs first encounter invasive cane toads (Bufo marinus), they try to eat the toxic invader and, if they are able to do so, are likely to die from poisoning. Because frogs are gape‐limited predators, small metamorphs cannot ingest a toad and thus survive long enough to disperse away from the natal pond (and thus from potentially deadly toads). These data show that larger size at metamorphosis can reduce rather than increase anuran survival rates, because larger metamorphs are more easily able to ingest (and thus be poisoned by) metamorph cane toads. Our results suggest that patterns of selection on life‐history traits of native taxa (such as size and age at metamorphosis, seasonal timing of breeding and duration of pondside aggregation prior to dispersal) can be modified by the arrival of an invasive species. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100 , 329–336.     

Neurometamorphosis of the ear in the gypsy moth,Lymantria dispar,and its homologue in the earless forest tent caterpillar moth,Malacosoma disstria

The adult gypsy moth, Lymantria dispar (Lymantriidae: Noctuoidea) has a pair of metathoracic tympanic ears that each contain a two-celled auditory chordotonal organ (CO). The earless forest tent caterpillar moth, Malacosoma disstria (Lasiocampidae: Bombycoidea), has a homologous pair of three-celled, nonauditory hindwing COs in their place. The purpose of our study was to determine whether the adult CO in both species arises from a preexisting larval organ or if it develops as a novel structure during metamorphosis. We describe the larval metathoracic nervous system of L. dispar and M. distria, and identify a three-celled chordotonal organ in the anatomically homologous site as the adult CO. If the larval CO is severed from the homologue of the adult auditory nerve (IIIN1b1) in L. dispar prior to metamorphosis, the adult develops an ear lacking an auditory organ. Axonal backfills of the larval IIIN1b1 nerve in both species reveal three chordotonal sensory neurons and one nonchordotonal multipolar cell. The axons of these cells project into tracts of the central nervous system putatively homologous with those of the auditory pathways in adult L. dispar. Following metamorphosis, M. disstria moths retain all four cells (three CO and one multipolar) while L. dispar adults possess two cells that service the auditory CO and one nonauditory, multipolar cell. We conclude that the larval IIIN1b1 CO is the precursor of both the auditory organ in L. dispar and the putative proprioceptor CO in M. disstria and represents the premetamorphic condition of these insects. The implications of our results in understanding the evolution of the ear in the Lepidoptera and insects in general are discussed. © 1996 John Wiley & Sons, Inc.     

The Evolution of the Tetrapod Middle Ear in the Rhipidistian-Amphibian Transition

From a survey ot the structure of the skull in rhipidistianfishes and early labylinthodont Amphibia and of the mechanismof hearing in these two groups, an account of the evolutionof the tetrapod middle ear is presented. The overall modificationof the otic region of the skull during the rhipidistian-amphibiantransition is analyzed in terms of changes in different organsystems in response to different selective pressures (affecting,for example, the feeding, respiratory, and locomotory mechanisms).These changes are seen to occur in a completely integrated pattern.Considerations of the different requirements for sound receptionunder water and in air, in connection with this correlated progressionof evolutionary change in the otic region of the head, revealthe manner in which the hyomandibular, spiracular diverticulum,and operculum of rhipidistian fishes became modified to formthe stapes, the tympanic cavity, and the outer portion of thetympanum, respectively, of tetrapods.     

After the crash: How do predators adjust following the invasion of a novel toxic prey type?

The ability of a native predator to adjust to a dangerously toxic invasive species is key to avoiding an ongoing suppression of the predator's population and the trophic cascade of effects that can result. Many species of anurophagous predators have suffered population declines due to the cane toad's (Rhinella marina: Bufonidae) invasion of Australia; these predators can be fatally poisoned from attempting to consume the toxic toad. We studied one such toad‐vulnerable predator, the yellow‐spotted monitor (Varanus panoptes: Varanidae), testing whether changes to the predator's feeding behaviour could explain how the species persists following toad invasion. Wild, free‐roaming lizards from (1) toad‐naïve and (2) toad‐exposed populations were offered non‐toxic native frogs and slightly toxic cane toads (with parotoid glands removed) in standardized feeding trials. Toad‐naïve lizards readily consumed both frogs and toads, with some lizards displaying overt signs of illness after consuming toads. In contrast, lizards from toad‐exposed populations consumed frogs but avoided toads. Repeated encounters with toads did not modify feeding responses by lizards from the toad‐naïve populations, suggesting that aversion learning is limited (but may nonetheless occur). Our results suggest that this vulnerable predator can adjust to toad invasion by developing an aversion to feeding on the toxic invader, but it remains unclear as to whether the lizard's toad‐aversion arises via adaptation or learning.     

Genes induced during the early developmental stages of the Cane Toad, Bufo (Chaunus) marinus

Metamorphosis, a critical stage in the development of toads and frogs, involves rapid levels of morphological change. In the current study, we have used microarray analysis to identify shifts in gene expression between tadpole and toadlet stages of the cane toad, Bufo (Chaunus) marinus. Here, we report on nine genes that show the greatest induction during metamorphosis; the gut-associated gastrokine and trefoil factor, blood components haemoglobins alpha/beta, apolipoprotein and serum albumin, a nasal gene olfactomedin, a lens gene gamma-crystallin, and a novel gene with low homology to frog harderin. We present both temporal and spatial expression patterns of these genes identified in developing and adult cane toads. This study extends our knowledge of the molecular basis of toad metamorphosis, and not only offers insights to the genes induced during the general remodelling that occurs but also reveals possible targets for control and manipulation of amphibian pest species, for example, the cane toad in Australia.     

Behavioural responses of reptile predators to invasive cane toads in tropical Australia

The ecological impact of an invasive species can depend on the behavioural responses of native fauna to the invader. For example, the greatest risk posed by invasive cane toads (Rhinella marina Bufonidae) in tropical Australia is lethal poisoning of predators that attempt to eat a toad; and thus, a predator's response to a toad determines its vulnerability. We conducted standardized laboratory trials on recently captured (toad‐naïve) predatory snakes and lizards, in advance of the toad invasion front as it progressed through tropical Australia. Responses to a live edible‐sized toad differed strongly among squamate species. We recorded attacks (and hence, predator mortality) in scincid, agamid and varanid lizards, and in elapid, colubrid and pythonid snakes. Larger‐bodied predators were at greater risk, and some groups (elapid snakes and varanid lizards) were especially vulnerable. However, feeding responses differed among species within families and within genera. Some taxa (notably, many scincid and agamid lizards) do not attack toads; and many colubrid snakes either do not consume toads, or are physiologically resistant to the toad's toxins. Intraspecific variation in responses means that even in taxa that apparently are unaffected by toad invasion at the population level, some individual predators nonetheless may be fatally poisoned by invasive cane toads.