Present address: Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11, The University of Sydney, NSW 2006, Australia.
An enzyme in snail saliva induces herbivore-resistance in a marine alga
Article first published online: 15 NOV 2006
Volume 21, Issue 1, pages 101–106, February 2007
How to Cite
COLEMAN, R. A., RAMCHUNDER, S. J., MOODY, A. J. and FOGGO, A. (2007), An enzyme in snail saliva induces herbivore-resistance in a marine alga. Functional Ecology, 21: 101–106. doi: 10.1111/j.1365-2435.2006.01210.x
- Issue published online: 15 NOV 2006
- Article first published online: 15 NOV 2006
- Received 30 June 2006; revised 27 August 2006; accepted 29 August 2006Editor: Frank Messina
- 1It is well understood that herbivory can cause plants to elevate production of defensive chemicals in their tissues. One of the key questions in understanding patterns of potential coevolutionary links between plant and herbivore is ‘what switches these induced plant defences on?’ Until cues are identified, understanding the evolutionary and ecological significance of defences in the context of the plant is difficult.
- 2We induced host plant resistance in a marine macroalga (Ascophyllum nodosum) in the absence of herbivory by application of α-amylase, known to exist in mollusc saliva. There was a demonstrable change in the behaviour of a subsequent herbivore (Littorina obtusata) consistent with herbivore induction, i.e. reduced consumption, more but smaller meals and greater movement. We also produced a concomitant increase in the level of phlorotannins, compounds associated with defence against herbivory.
- 3Such changes in herbivore behaviour and plant chemistry provide evidence that brown algae and higher plants utilize similar salivate-based signals in the induction of defence against herbivory. Thus the possibility that such responses could be either highly conserved or a convergent strategy is discussed.
Plants defend themselves against herbivory in many ways (see Karban & Baldwin 1997 for review). Most, if not all, theory concerning herbivore-induced defences/resistance in plants is based on examples from plants in terrestrial systems (Karban & Baldwin 1997; Kessler & Baldwin 2001). Despite this, there is clear evidence that grazing herbivores also cause chemical changes such as elevated production of terpenoids in marine algae from a variety of groups (Pohnert 2004) but there has been less emphasis on ecological consequences of herbivore behaviour (Borell, Foggo & Coleman 2004; Amsler & Fairhead 2006). We recently demonstrated such changes in chemistry and subsequent alterations in herbivore behaviour in the brown alga Ascophyllum nodosum (L.) Le.Jol grazed by the littorinid mollusc Littorina obtusata (L.) (Borell et al. 2004). Previous studies had identified herbivore-induced changes in chemistry and reduced consumption in other species of fucoid algae (Amsler & Fairhead 2006). These changes are consistent with theories of induced defence, which predict greater herbivore mobility on herbivore-induced tissue plants (Edwards & Wratten 1987; Coleman, Barker & Fenner 1996) and appear analogous to those that occur in phylogenetically distinct terrestrial systems where plants are grazed by invertebrates such as insects and gastropod molluscs (Karban & Baldwin 1997). As induced defences bear costs (Agrawal 2001; Cipollini, Purrington & Bergelson 2003), theory predicts plants must enable such defences only when under attack (Agrawal & Karban 1999). Plants must therefore be able to recognize herbivore attack to warrant an induced response (Borell et al. 2004). To date, the identity of this signal is known in only one herbaceous plant family when it is grazed by lepidopteran insect larvae (Mattiacci, Dicke & Posthumus 1995; Alborn et al. 1997; Shen, Zheng & Dooner 2000). Induction of defence could occur by mechanical damage (Green & Ryan 1972; Karban & Baldwin 1997); however, mechanical damage alone usually does not initiate a wound-induced response in A. nodosum (Pavia & Toth 2000; Borell et al. 2004) and other algae (Amsler & Fairhead 2006). This is consistent with theoretical predictions (Edwards & Wratten 1987; Baldwin 1990) in that the rocky intertidal and/or shallow subtidal habitat carries a high risk of abrasion damage (Shanks & Wright 1986); therefore, some additional signal from a grazer should be required to initiate a wound-induced defence. Interestingly, Pavia & Toth (2000) demonstrated that not only is the biotic link essential to induce defences in A. nodosum, but also that the biotic link is dependent on the grazer in that grazing by snails will induce defences but amphipod grazing will not.
One possible signal is the presence of digestive enzymes in secretions produced as herbivores feed (Alborn et al. 1997). Saliva has been shown to modify plant growth patterns (Bergman 2002), which may have long-term herbivore-resistance implications. A specialized role for saliva in triggering defences has been demonstrated in two different terrestrial plant species (Mattiacci et al. 1995; Alborn et al. 1997); however, these defences are indirect as they affect the herbivore via natural enemies. This has not been shown in algae, although the products of digestion of alginates can trigger an oxidative burst in some Laminariales (Küpper et al. 2002). Similarities in chemical mediators of growth and defence occur in angiosperms and algae from several different groups, for example oxilipins such as jasmonic acid have been shown to elicit changes in concentrations of polyphenolics in a wide variety of plant and algal taxa (Karban & Baldwin 1997; Arnold et al. 2001; Pohnert 2004). Despite such evidence for either conserved or converged mechanisms, there is no evidence at present for any such parallel in the mechanism by which the plants detect threats from grazers. Many gastropods secrete salivary amylase and other carbohydrases (Vieira & Ladeira 1965; Fretter & Graham 1994; Moura, Terra & Ribeiro 2004), marine snails such as littorinids are no exception (Fretter & Graham 1994; Park et al. 1999; Ermakova et al. 2001). As snails graze, saliva is released into the oral mucus (Davies & Hawkins 1998). Plants grazed on by snails will therefore be exposed to salivary enzymes. We investigated the possible activity of salivary amylase as an elicitor of induced defence in an intertidal brown alga. Rather than apply the enzyme cue and search for a chemical response we chose to use a bioassay approach as this was a more reliable method of checking for induction (Coleman et al. 1996; Borell et al. 2004) given the potentially unstable and uncertain chemistry involved (Kubanek et al. 2004). A model of herbivore-induced defence developed for terrestrial plants predicts that herbivores respond to elevated levels of plant defensive compounds by unambiguously modifying their feeding behaviour (Edwards & Wratten 1987; Coleman et al. 1996). Previous work has shown this to be the case (Pavia & Toth 2000; Borell et al. 2004).
Amylase may be present in locomotory as well as oral mucus secreted by L. obtusata, thus the presence of grazers could signal a potential grazing threat to the plant. Alternatively, mechanical damage might be needed to allow the enzyme to penetrate the tissues. Using real L. obtusata saliva on its own would be technically extremely difficult (Mark Davies pers. comm.), so we chose to use an artificial representation. We predicted that α-amylase, the most frequently found amylase in saliva, either with or without accompanying mechanical damage, would elicit a response from plants similar to that produced by herbivory by L. obtusata. We predicted that subsequent L. obtusata grazers on induced plants would be more mobile on damaged/induced plants, they would take more frequent but smaller meals (Borell et al. 2004) and would consume less plant tissue than individuals fed on control plants. While possibly not being the agent which reduces feeding (Deal et al. 2003; Amsler & Fairhead 2006), phlorotannins correlate well with resistance to herbivory (Pavia & Toth 2000; Borell et al. 2004) and are related to many induced-defence chemicals in terrestrial plants (Karban & Baldwin 1997). We therefore predicted that the concentration of phlorotannins in A. nodosum tissues in previously exposed to α-amylase would be similar to those in A. nodusum previously grazed by L. obtusata.
Experiments were carried out between May and October 2004. Mature A. nodosum plants were collected from the field (near Plymouth, UK) with holdfasts intact, cleaned of epibiota and acclimated in seawater in the laboratory for a month at 15 °C and L : D 16 h : 8 h. Four weeks before the trials, primary shoot sections of 10 plants were exposed to one of four treatments lasting 2 weeks (Borell et al. 2004): (I) control (no action); (II) α-amylase 0·035 U encapsulated in 40 µL agarose gel spread over 8–10 mm2; (III) mechanical damage plus α-amylase [8–10 mm2 of tissue was abraded (Borell et al. 2004) then 40 µL of amylase suspension was applied]; and (IV) herbivore-induction by three L. obtusata constrained using a mesh bag on a 20 cm long section of shoot. Seawater was changed frequently to reduce any effects of ammonium salts excreted by the snails (Taylor & Rees 1998).
To ensure that the enzyme was restricted to the areas that had been artificially damaged, it was encapsulated in an agarose gel. Preliminary trials indicated that, compared with control (untreated) plants, application of agarose on its own did not cause a different response in the behaviour of L. obtusata placed on different sections of the plant 2 weeks after application. Extra low-temperature gelling agarose (Sigma A-5030, gel point 8–17 °C) was mixed at 3% (w/v) with seawater, and this suspension was then autoclaved under normal conditions to ensure the agarose was completely melted. From this stock agarose solution, aliquots of 500 µL were removed and dispensed into Eppendorf tubes. These were kept at 4 °C, and the agarose re-melted at 70 °C before use. An ammonium sulphate suspension of α-amylase (Sigma A-2643, from porcine pancreas) was used, which had a specific activity of 1·370 U mg−1 protein, and 32 mg mL−1 of protein. The working stock solution was derived by adding 1 µL of enzyme suspension to 99 µL of seawater, thus reducing the activity to 438·4 U mL−1. From this stock solution of α-amylase, 1 µL was removed and added to 500 µL of melted 3% agarose at 30 °C to give a final concentration of 0·88 U mL−1. The amylase suspension was applied to an area approximately 8–10 mm2 (undamaged or damaged); thus 40 µL of the melted amylase-containing agarose was applied to each area. This amounted to the application of 0·035 U of amylase per area.
The following experiments were performed to test the predictions. To test the prediction of greater mobility of herbivores on tissue from induced plants, treated sections of the plants were removed and an individual snail, starved for 48 h, was allowed to move freely about the plant for 100 min. The location of the snail was noted continuously and changes in position were detected with reference to an acetate grid fixed over the test arena (Coleman et al. 1996; Borell et al. 2004). Differences in mean number of relocations on each tissue were tested by a one-way anova run on WinGMAV5 (EICC, University of Sydney), heteroscedasticity was checked by Cochran's C and significant factors were separated by SNK tests (Underwood 1997). We tested the prediction of smaller but more frequent meal sizes by allowing starved individual L. obtusata to feed for 100 min on undamaged excised 10 cm sections of primary shoot placed in a 14 cm Petri dish lined with filter paper dampened with sea water. The resulting feeding scars were counted and measured using a dissecting microscope and 1 mm2 eyepiece graticule. We used a log-likelihood (G) test of independence adjusted by Williams’ correction (Sokal & Rohlf 1995) to test the null hypothesis that the size-frequency distribution of feeding scars of different size was independent of treatment. The prediction that the snails fed on tissue from induced plants would consume less was tested by allowing a snail to feed on excised undamaged sections of primary shoot from a treated plant. Each section was placed in a 14 cm Petri dish lined with filter paper dampened with sea water and an individual snail was allowed to forage for 100 min. Previous work had shown autogenic water loss to be insignificant with this method (Borell et al. 2004). Differences in mean amount eaten were assessed using anova as above. Phlorotannin concentrations were determined using a modified Folin–Ciocalteau method (Borell et al. 2004) from 1 g of plant material from a different section of the plant. Differences in mean concentration of phlorotannin were tested using anova as above. There were 10 replicates in every experiment.
Application of amylase caused a significant increase in the amount of movement by the snails (Table 1) although this was less than the amount of movement on herbivore-induced tissue (Fig. 1). Mechanical damage did not appear necessary for amylase to initiate a response. The frequency of meals of different sizes was significantly dependent on treatment (Gadj = 35·56, 12 d.f., P < 0·001). The frequency distribution of feeding scars on abraded plants with amylase was most similar to that of damage inflicted by snails feeding on herbivore-induced tissue. The feeding scar frequency distribution from L. obtusata on plants exposed to amylase alone was more similar to that of abraded + amylase than to the frequency distribution of damage on naïve plants (Fig. 2). Treatments significantly affected the amount of algal tissue consumed (Table 1). In this case, abrasion substantially increased the effect of amylase in that the amount of tissue consumed was identical to the amount consumed by snails feeding on tissue from plants exposed to prior herbivory but 17 times less than control plants (Fig. 3). Phlorotannin concentrations were also significantly affected by treatment (Table 1), with a clear ranking of herbivore-induced tissues containing most phlorotannins followed by abraded + amylase, amylase alone then naïve plants. SNK tests showed all treatments to be significantly different from each other (Fig. 4).
|Heteroscedasticity||Moves C = 0·49, NS||Consumption C = 0·48, NS||Phlorotannin C = 0·50, NS|
|Treatment||3||3690·00||28·01||< 0·001||0·02||137·62||< 0·001||42·41||29·65||< 0·001|
Application of α-amylase caused a consistent decline in food acceptability for subsequent gastropod herbivores. This decline correlates well with other studies using animal-based induction. With respect to the amount consumed, the abraded + amylase treatment caused an identical reduction in amount, relative to control plants, as that of L. obtusata feeding on tissues from herbivore-induced tissues. These values are very similar to an earlier study (Borell et al. 2004). The levels of phlorotannin detected in all treatments were less than in previous studies (Pavia & Toth 2000; Borell et al. 2004); however, the difference between phlorotannin concentrations in herbivore-induced tissue and naïve tissue is consistent between this work and previous studies (Pavia & Toth 2000; Borell et al. 2004). Tissue from plants abraded with amylase was closer to that of herbivore-induced plants in terms of phlorotannin concentration than to abraded tissues from those earlier studies, whereas tissues from plants exposed amylase alone were more similar to artificially damaged tissues in earlier studies (Pavia & Toth 2000; Borell et al. 2004) than tissues from herbivore-induced plants.
Our results clearly indicate that α-amylase can act to induce resistance to herbivory in A. nodosum. Previous attempts at artificially inducing resistance in A. nodosum have met with little or no success (Pavia & Toth 2000; Borell et al. 2004; Rohde, Molis & Wahl 2004) indicating that either the mechanical damage did not accurately mimic the process of herbivory (Baldwin 1990) or that some cues were missing. The herbivore mobility model (Edwards & Wratten 1987; Coleman et al. 1996; Borell et al. 2004) predicted that compared to herbivores feeding on naïve plants, herbivores attacking induced plants would: (1) be more mobile; (2) take more but smaller meals; and (3) would consume less plant tissue. For movement, regardless of mechanical damage, application of amylase caused a herbivore behavioural response intermediate between the responses of herbivores on naïve plants and those on induced plants, which indicated that amylase caused a systemic reduction in acceptability. In respect of tests of consumption (2 and 3 above), abrading tissues and applying amylase gave results identical to those from herbivores feeding from tissues from induced plants. Combining these results from patterns of phlorotannin production indicate that while amylase application does not directly mimic prior herbivory, it does cause enough of a systemic response to cause feeding patterns of subsequent herbivores to be no different from herbivores feeding on induced plants. Thus salivary carbohydrases are capable of triggering the induced response.
Given that oligoalginates, derivatives of the degradation of alginate, can elicit a defensive oxidative burst in many brown algae (Küpper et al. 2002) it seems likely that it is the products of the action of α-amylase on polysaccharides that elicit the response in A. nodosum rather than the enzyme itself. Some effect of α-amylase was seen regardless of whether the surface of the plant was abraded or not, thereby eliminating intracellular polysaccharide. The principal polysaccharide components of brown algal cell walls are cellulose, alginates and fucoidan (van den Hoek, Mann & Jahns 1995). However, α-amylase is a 1,4-α-d-glucan-glucanohydrolase requiring three or more α(14) linked d-glucose residues, and would not be expected to affect any of these: cellulose contains β glycosidic linkages; alginates contain neither glucose nor α(14) linkages (van den Hoek et al. 1995); and fucoidan does not contain glucose (Berteau & Mulloy 2003). It seems therefore that the α-amylase must be acting on either a minor polysaccharide or a glycoprotein component of the cell wall. However, surprisingly α-amylase is known to bind to alginates (Sardar & Gupta 1998) suggesting that a direct signalling role for the enzyme is perhaps a possibility after all. The possibility that highly soluble sugars are involved may provide an underlying mechanism for the observation of waterborne cues being passed from plant to plant (Toth & Pavia 2000) in the same systems used here.
This is the first experimental demonstration of a component of salivate switching on direct herbivore-induced defence; a recent correlational study inferred a salivate rôle in nicotine production in Nicotiana tabacum (Musser et al. 2006). Previous evidence from terrestrial insect–herbaceous plant interactions showed indirect interactions via changes in plant-produced volatiles attracting the natural enemies (Mattiacci et al. 1995; Alborn et al. 1997; Coleman et al. 1997; Mattiacci et al. 2001), rather than a direct impact on the herbivore. Direct impacts of induced responses to herbivory have been shown to reduce herbivory by up to 50% (Agrawal 1998). Our results show that although dicotyledenous plants and marine macroalgae diverged over 400 Ma, they have either evolved analogous responses to switch on defences via a simple conserved trigger mechanism or have independently evolved this trait.
Despite the evolutionary divergence and ecological differences between angiosperms and green algae and their respective habitats, they are often attacked by phylogenetically similar organisms. Of the potential herbivores present in terrestrial and intertidal marine habitats, molluscs stand out as the most abundant and potentially the most damaging (Hawkins et al. 1992; Hanley, Fenner & Edwards 1995). It is significant therefore that molluscs and insects, the most common plant herbivores on land, both secrete glucose oxidases in saliva. Our data indicate that such enzymes from molluscs as well as insects can elicit induced defences. Other herbivores of marine algae include crustaceans such as amphipods and isopods. There is, to date, no evidence that we are aware of that any marine peracarid crustaceans produce salivary amylases (J.I. Spicer, pers. comm.), and it is significant that crustacean grazing has been shown to be less reliable in producing elevated chemical defences than mollusc herbivory (Pavia & Toth 2000).
In conclusion, we have demonstrated elicitation of induced defences in a brown alga by a grazer via a carbohydrase enzyme. A single previous study in a higher plant system showed induction of defence, albeit indirectly via the next trophic level, as a consequence of a salivary carbohydrase (Mattiacci et al. 1995). This trigger in common between two disparate plant lineages suggests questions about the evolution of this trait. This is either evolved in parallel, in which case the adaptation has been significant enough to arise in two utterly different lineages despite the costs of developing and deploying such strategies (Agrawal & Karban 1999), or it is possible that using a simple cue is an effective, highly conserved strategy, and that the two groups have retained the use of carbohydrases as a signal most representative of grazer attack.
This work was funded by the authors and the University of Plymouth. We thank Richard Ticehurst and Nick Crocker for technical support. During writing and analysis, RAC received support from the Australian Research Council via the Special Research Centres Programme. This paper was much improved by considered comments from Frank Messina and two anonymous reviewers.
- 1998) Induced responses to herbivory and increased plant performance. Science 279, 1201–1202. (
- 2001) Phenotypic plasticity in the interactions and evolution of species. Science 294, 321–326. (
- 1999) Why induced defenses may be favored over constitutive strategies in plants. The Ecology and Evolution of Inducible Defenses (eds R.Tollrian & C.D.Harvell), pp. 45–61. Princeton University Press, Princeton, NJ. & (
- 1997) An elicitor of plant volatiles from beet armyworm oral secretion. Science 276, 945–949. , , , , & (
- 2006) Defensive and sensory chemical ecology of brown algae. Advances in Botanical Research 43, 1–91. & (
- 2001) Evidence for methyl jasmonate-induced phlorotannin production in Fucus vesiculosus (Phaeophyceae). Journal of Phycology 37, 1026–1029. , , , & (
- 1990) Herbivory simulations in ecological research. Trends in Ecology and Evolution 5, 91–93. (
- 2002) Can saliva from moose, Alces alces, affect growth responses in the sallow, Salix caprea? Oikos 96, 164–168. (
- 2003) Sulfated fucans, fresh perspectives: structures, functions, and biological properties of sulfated fucans and an overview of enzymes active toward this class of polysaccharide. Glycobiology 13, 29R–40R. & (
- 2004) Induced resistance in intertidal macroalgae modifies feeding behaviour of herbivorous snails. Oecologia 140, 328–344. , & (
- 2003) Costs of induced responses in plants. Basic and Applied Ecology 4, 79–85. , & (
- 1996) Cabbage (Brassica oleracea var. Capitata) fails to show wound-induced defence against adapted and non-adapted herbivores? Oecologia 108, 105–112. , & (
- 1997) Relative effect of different host feeding site on long-range host location and electroantennogram response in the parasitoid Cotesia glomerata L. (Hymenoptera: Braconidae). Journal of Applied Entomology 121, 487–494. , , & (
- 1998) Mucus from marine molluscs. Advances in Marine Biology 34, 1–71. & (
- 2003) Galactolipids rather than phlorotannins as herbivore deterrents in the brown seaweed Fucus vesiculosus. Oecologia 136, 107–114. , , & (
- 1987) Ecological significance of wound-induced changes in plant chemistry. Insects–Plants (eds V.Labeyrie, D.Farbes & D.Lachaise), pp. 213–218. W.H. Junk, Dordrecht. & (
- 2001) Proteins of brown seaweeds as inhibitors of endo-13-beta-d-glucanases of marine invertebrates. Biochemistry-Moscow 66, 188–194. , , , & (
- 1994) British Prosobranch Molluscs: Their Functional Anatomy and Ecology, 2nd edn. The Ray Society, London. & (
- 1972) Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175, 776–777. & (
- 1995) An experimental field-study of the effects of mollusk grazing on seedling recruitment and survival in grassland. Journal of Ecology 83, 621–627. , & (
- 1992) Plant–animal interactions on hard substrata in the North-East Atlantic. Plant–Animal Interactions in the Marine Benthos (eds D.M.John, S.J.Hawkins & J.H.Price), pp. 1–32. Clarendon Press, Oxford. , , & (
- 1995) Algae: an Introduction to Phycology. Cambridge University Press, Cambridge. , & (
- 1997) Induced Responses to Herbivory. University of Chicago Press, Chicago, IL. & (
- 2001) Defensive function of herbivore-induced plant volatile emissions in nature. Science 291, 2141–2144. & (
- 2004) Ambiguous role of phlorotannins as chemical defenses in the brown alga Fucus vesiculosus. Marine Ecology Progress Series 277, 79–93. , , & (
- 2002) Oligoalginate recognition and oxidative burst play a key role in natural and induced resistance of sporophytes of laminariales. Journal of Chemical Ecology 28, 2057–2081. , , , & (
- 1995) β-glucosidases: an elicitor of herbivore-induced plant odor that attracts host-searching parasitic wasps. Proceedings of the National Academy of Sciences of the United States of America 92, 2036–2040. , & (
- 2001) Systemically induced plant volatiles emitted at the time of ‘danger’. Journal of Chemical Ecology 27, 2233–2252. , , , , & (
- 2004) The functional organization of the salivary gland of Biomphalaria straminea (Gastropoda: Planorbidae): Secretory mechanisms and enzymatic determinations. Journal of Molluscan Studies 70, 21–29. , & (
- 2006) Ablation of caterpillar labial salivary glands: Technique for determining the role of saliva in insect–plant interactions. Journal of Chemical Ecology 32, 981–992. , , , & (
- 1999) Amylase polymorphism of Littorina brevicula from polluted and unpolluted sites, Korea. Bulletin of Environmental Contamination and Toxicology 63, 633–638. , , & (
- 2000) Inducible chemical resistance to herbivory in the brown seaweed Ascophyllum nodosum. Ecology 81, 3212–3225. & (
- 2004) Chemical defense strategies of marine organisms. Topics in Current Chemistry 239, 179–219. (
- 2004) Regulation of anti-herbivore defence by Fucus vesiculosus in response to various cues. Journal of Ecology 92, 1011–1018. , & (
- 1998) Alginate beads as an affinity material for alpha amylases. Bioseparation 7, 159–165. & (
- 1986) Adding teeth to wave action: the destructive effects of wave-borne rocks on intertidal organisms. Oecologia 69, 420–428. & (
- 2000) A maize sesquiterpene cyclase gene induced by insect herbivory and volicitin: Characterization of wild-type and mutant alleles. Proceedings of the National Academy of Sciences of the United States of America 97, 14807–14812. , & (
- 1995) Biometry, 3rd edn. W.H. Freeman, New York. & (
- 1998) Excretory products of mobile epifauna as a nitrogen source for seaweeds. Limnology and Oceanography 43, 600–606. & (
- 2000) Water-borne cues induce chemical defense in a marine alga (Ascophyllum nodosum). Proceedings of the National Academy of Sciences of the United States of America 97, 14418–14420. & (
- 1997) Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variance. Cambridge University Press, Cambridge. (
- 1965) Amylase of snail Australorbis glabratus (Mollusca: Planorbidae). Comparative Biochemistry and Physiology 14, 281–&. & (