by guest on July 11, 2016 ...angelo.berkeley.edu/wp-content/uploads/Kupferberg... · 1988, Berven and Chadra, 1988) can deter-mine tadpole growth and timing of meta-morphosis. These

AMER. ZOOL., 37:146-159 (1997)

The Role of Larval Diet in Anuran Metamorphosis1

SARAH J. KUPFERBERG

Department of Integrative Biology, University of California Berkeley, California 94720

SYNOPSIS. FOCUS on tadpole diet and foraging behavior offers potential forintegrating ecological and endocrinological approaches to understanding anu-ran metamorphosis. Natural larval diets vary widely in relative amounts ofprotein, carbohydrate and lipid, factors known to influence thyroid hormonefunction, which in turn is essential for metamorphosis to occur. Previous workhas shown that tadpoles fed high protein diets grow and develop quickly. Thispattern is consistent with findings from other classes of vertebrates that someaspects of thyroid function, e.g., activity of 5'D monodeiodinase, an enzymethat converts thyroxine (T4) to the more potent metamorphosis-inducer 3,5,3'-triiodothyronine (T,), are proportional to availability of dietary protein. Infield experiments at a northern California river, I found that nutritionalvariation among algal taxa routinely consumed by tadpoles (chlorophytes,diatoms, cyanobacteria, etc.) influenced metamorphosis several ways. Growthrate was positively correlated with percent protein content of a food type.Tadpoles fed filamentous green algae with epiphytic diatoms developed morequickly and metamorphosed at larger sizes than tadpoles raised on other di-ets. The dominant diatoms in this system are high in protein because theyhost nitrogen fixing cyanobacterial endosymbionts. Resource quality also me-diated the effects of competition and predation. Changes in abundance of highquality algae caused by invading bullfrog (R. catesbeiana) tadpoles explainedmost of their competitive effects on metamorphosis of native tadpoles (Hylaregilla and (Rana boylii). In choice experiments, tadpoles foraged selectivelyon the algal foods that promoted most rapid growth and development. In thepresence of garter snake predators, however, tadpoles avoided such patchesthus decreasing growth. Direct examination of thyroid hormone productionin tadpoles consuming different diets may reveal a proximate mechanism link-ing diet quality to size at and time to metamorphosis.

INTRODUCTION food resources (Brockelman, 1969; De-The basic premise of this review is that Benedictis 1974; Wilbur, 1980; Travis,

an examination of tadpole diet and forag- 1 9 8 4 ^ Kupferberg 1996). Variables directlying behavior can reveal links between what influenced by diet, e.g., consumption rateecologists and endocrinologists know about (J § ' d a y ) and size specific growth rateanuran metamorphosis. Ecologists have ex- (g " d ay ' g tadpole), have been identi-amined external influences on metamorpho- fied a s k e y predictors in physiological (Pan-sis including food quantity (Wilbur, 1977a, d i a n a n d Marian, 1985a) and ecologicalb; Berven and Chadra, 1988), changes in (Wilbur and Collins, 1973) models of am-growth history (Travis, 1984; Alford and phibian metamorphosis. Other environmen-Harris, 1988), food quality (Steinwachser t a l f a c t o r s w h i c h c a n v a r y independently ofand Travis, 1983; Ahlgren and Bowen, f o o d availability, e.g., temperature, may in-1991; Pfennig et a I 1991; Kupferberg et d i r e c t l y influence metamorphosis by alter-al., 1994), diet composition (Crump, 1983, i ng feeding rates (Calef, 1973; Pandian and1990; Pfennig, 1990), and competition for Marian, 19856; Warkentin, 1992a, 1992ft).

Interactions between food availability and'. temperature have also been shown to affect

From the Symposium Amphibian Metamorphosis: i o n o f g e n e s controlling levels ofAn Integrative Approach presented at the Annual Meet- F 6 6

ing of the American Society of Zoologists, 27-30 De- paedomorphosis in the Mexican axolotlcember 1995. at Washington, D.C. (Voss, 1995).

146

by guest on July 11, 2016http://az.oxfordjournals.org/

Dow

nloaded from

http://az.oxfordjournals.org/

LARVAL DIET AND ANURAN METAMORPHOSIS 147

Endocrinologists, on the other hand, haveprimarily examined internal influences onmetamorphosis, identifying thyroid hor-mone (TH) as essential to this transforma-tion (reviewed by Galton, 1988), and pro-lactin (PRL) as essential to maintaining lar-val structures and promoting growth (Whiteand Nicoll, 1981; Delidow 1989; Tata1991). Molecular biologists have identifiednuclear TH receptors as transcription fac-tors that regulate cascades of gene expres-sion governing specific cell death, prolifera-tion, and differentiation (reviewed by Shi,1994). External stressors, such as pond dry-ing and crowding, can affect neurotransmit-tors important in the thyroid axis, such asthe stress hormone corticosterone, which inturn affect TH (Denver, 1997). The exter-nal factor influencing TH function that Iwill focus on is the food ingested by tad-poles.

In this paper, I review the effects of dietquantity and composition on thyroid andprolactin function in mammals, birds, andfish, because there is no research directlyaddressing diet effects on thyroid or prolac-tin in larval amphibians. The patterns ob-served from the general vertebrate litera-ture, however, are consistent with the resultsof many investigations of diet effects onanuran metamorphosis. I argue that a focuson diet composition is pertinent to under-standing metamorphosis because the vari-ous foods consumed by larval anurans varywidely in ratios of protein, carbohydrate,and lipid, factors shown to influence thyroidfunction in other vertebrates. I then presentsome of my published and unpublished re-sults on the ways in which interactionsamong algae, larval anuran grazers, and gar-ter snake predators influence survival to,time to, and size at metamorphosis. I showthat independent of quantity (because tad-poles were fed ad libitum), protein contentof diverse algal taxa correlates with rapidgrowth and metamorphosis; that tadpolesfeed selectively on the algae that promotegrowth and development; and, that this se-lectivity changes in the presence of preda-tors. I also show that the effects of tadpoleson the composition of algal assemblagescan explain the competitive interactionsamong tadpoles. Finally, I end with sugges-

tions for future research which could di-rectly address the effects of diet on thyroidfunction in larval anurans.

CONNECTIONS BETWEEN DIET ANDTHYROID FUNCTION

Food quantityFood quantity and patterns of nutrient in-

take affect several aspects of thyroid func-tion in a variety of vertebrates (reviewed byEales, 1988). In mammals (most studiesdone on humans and rats) starvation leadsto decreased levels of triiodothyronine (T3)and thyroxine (T4) in the blood, decreasedproduction of thyroid stimulating hormone(TSH), and decreased activity of 5'D, themonodeiodinase enzyme which converts T4to T3 in the liver. The starvation responsecan be triggered by a decrease in carbohy-drates, and the carbohydrates eaten in asingle meal after starvation stimulate thehypothalamo-pituitary-thyroid axis. De-creasing caloric intake decreases the bind-ing capacity of T3 receptors. Overfeeding,on the other hand, increases plasma T3 be-cause of increased conversion of T4 to T3.In birds and fish similar responses havebeen observed. Starvation leads to de-creased production of T4 from the thyroidand decreased conversion of T4 to T3, andcarbohydrates in feedings after starvationcause plasma T4 levels to surge (Himick etal., 1991).

Effects of food qualityMost investigations of food quality and

thyroid function have focused on varyingthe proportions of carbohydrates, lipids, andproteins in rations. Dietary protein appearsto be central in triggering diet-inducedchanges in thyroid function. In humans,protein deprivation leads to decreases in se-rum T3 by altering the 5' monodeiodinationof T4 (Marine et al., 1991). In rats, elimi-nation of protein from the diet is sufficientto reduce hypothalmic thyroid releasinghormone (TRH) and pituitary TSH gene ex-pression, leading to decreased plasma T3(Shi et al., 1993). In trout (Oncorhyncusmykiss), low amounts of dietary protein de-press T3, and the activity of 5'D is propor-tional to the amount of dietary protein


Dow

nloaded from


148 SARAH J. KUPFERBERG

(Eales et al., 1992). When the ratio of pro-tein to lipid is increased in artificial rationsfor trout, the amount of T3 increased (Ealeset al., 1990), whereas when protein washeld constant and carbohydrates and lipidsvaried, no effects on TH were observed. Notonly is the proportion of protein important,but the amino acid composition of the pro-tein also influences thyroid function (Rileyet al., 1993).

Because conversion of T4 to T3 is a keystep in the TH induction of amphibian meta-morphosis and maximal conversion occursat mid-metamorphic climax (Buscaglia etal., 1985), I will speculate that effects offood quantity and quality on metamorpho-sis observed in the ecological experimentsreviewed below may be mediated throughthis pathway.

ECOLOGICAL STUDIES ON TADPOLE DIETS

Effects of food quantity on growth andtiming of metamorphosis

The amount of food available is oftenthe cause for density dependent effects onanuran life history traits and survivorship(Brockelman, 1969; DeBenedictis, 1974;Wilbur, 1980; Travis, 1984). Diet ma-nipulations have confirmed that food quan-tity (Wilbur, 1977a, b; Alford and Harris,1988, Berven and Chadra, 1988) can deter-mine tadpole growth and timing of meta-morphosis. These results are consistentwith the pattern in other vertebrates, de-scribed above, that level of food ration candetermine up-regulation of TH and THreceptors.

In addition to quantity per se, the se-quence of food abundance and scarcity dur-ing larval ontogeny influences the tradeoffsbetween growth and differentiation (Alfordand Harris, 1988; Tejedo and Reques, 1994;Audo et al., 1995). In these experiments,tadpoles starved, or on low rations early inontogeny, caught up with high food con-trol tadpoles to metamorphose at a similarsize, but at a later time. This result is pre-dicted by Wilbur and Collins' (1973) mo-del that tadpoles optimally delay meta-morphosis by decreasing development ratewhen size-specific growth rate is above acertain threshold, as in an improving food

environment. My hypothesis suggests thatif high food level up-regulates TH, thenincreasing resource abundance would ac-celerate development and tadpoles wouldmetamorphose promptly. Because nutritionis likely to influence secretion of manyhormones other than TH (e.g., growth hor-mone, insulin, insulin-like growth factor,and PRL), which have a range of positive,negative, or unknown effects on metamor-phosis, a diet based endocrine explanationof this phenomenon is potentially verycomplex. PRL is of particular interest be-cause it clearly inhibits metamorphosis, butits response to feeding and/or fasting ishighly variable. If PRL concentration wereto increase as a result of feeding in tadpoles,as has been reported for humans (Carlson,1989), cattle (McAttee and Trenkle, 1971)and horses (DePew et al., 1994; Sticker etal., 1995), then metamorphosis would bedelayed. Conversely, diet restriction de-creases PRL in rats (Atterwill et al. 1989)and goats (Kloren et al., 1993). In tilapia,a teleost fish, fasting appears to have theopposite effect, causing augmentation ofPRL release by pituitary cells in vitro(Rodgers et al., 1992). Other studies reportno effect of food manipulation on PRL(sheep: Landefeld et al., 1989; Thomas etal., 1990; rat: McGuire et al., 1995). Foramphibians, I have found no evidence onhow diet affects PRL. An alternative endo-crine explanation for delayed metamorpho-sis in an improving food environment hasbeen suggested by Wassersug (1997). Highrates of ingestion may result in tadpolesdosing themselves with a metamorphic in-hibitor, i.e., epidermal growth factor, thatis secreted in their oral mucus (Lee et al.,1993).

Tadpoles facing food deprivation late indevelopment metamorphose generally at thesame time as high food controls but at asmaller size (Alford and Harris, 1988;Tejedo and Reques, 1994; Audo et al.,1995). A possible reason for this lack ofplasticity in metamorphic timing is thatonce tadpoles reared on abundant foodsreach a certain developmental stage, devel-opmental rate becomes fixed (Travis, 1984;Hensley, 1993). If starvation decreases pro-duction of TH in tadpoles as it does in the


Dow

nloaded from



vertebrate examples reviewed above, thequestion remains how does differentiationcontinue? Perhaps internal stores of energyand nutrients replace the reduced rations toallow TH synthesis to occur.

Effects of food quality on growth andmetamorphosis

Larval anuran diets are diverse and varywidely across taxa and environments. Thefoods consumed by tadpoles include vas-cular plants (Storre, 1994), vascular plantbased detritus (Diaz-Paniagua, 1985), at-tached algae or periphyton, a combinationof filamentous green algae, cyanobacteria,diatoms, bacteria, and desmids (Dickman,1968; Kupferberg et al., 1994), ultraplank-ton (Wassersug, 1972), planktonic greenalgae (Seale, 1980) and cyanobacteria(Seale and Beckvar, 1980), precipitates ofdissolved organic matter (Ahlgren and Bo-wen, 1991), protozoans (Nathan and James,1972), pollen (Wagner, 1986), mosquitolarvae (Pimm and Kitching, 1987; Blau-stein and Kotler, 1993), fairy shrimp (Pfen-nig, 1990; Pfennig et al., 1991), other tad-poles (Crump, 1983, 1986; Polis and Mey-ers, 1985; Pfennig, 1990), and other am-phibian eggs (Polis and Meyers, 1985;Lannoo et al., 1987; Jungfer and Schier-sari, 1995). The differences in nutritionalquality among the above foods are obviouswhen one considers the differences be-tween plant and animal tissues. Plant tissueis generally higher in carbohydrates andlower in lipids and protein than animalmatter. Less obviously, there is also sub-stantial variation in the nutrient content ofprimary producers. For example, filamen-tous green algae store excess photosyn-thate as carbohydrate whereas diatomsstore excess photosynthate as lipid (Boldand Wynne, 1985). Aquatic and terrestrialprimary producer nutrient content is alsodetermined by growing conditions, suchthat food quality is lowest when plantsgrow under nutrient shortage (Goldman etal., 1979; Chapin et al., 1990). A growingbody of literature is showing that aquaticherbivores can be limited by the mineral(e.g., nitrogen or phosphorous) content oftheir food (Sterner, 1993, and referencescited therein).

Are tadpole growth and development pro-tein limited? Several lines of evidence sug-gest that they are. Laboratory studies inwhich tadpoles were fed diets in which bothcarbohydrate and protein were varied, eitherin terms of lab rations or in amounts of ani-mal protein included in diet, demonstratethat increased protein enhances eithergrowth or development. Steinwachser andTravis (1983) found that growth and devel-opment of Hyla chrysoscelis tadpoles wasproportional to the ratio of protein to car-bohydrate in the diet, not overall rationquantity. For Rana clamitans tadpoles,growth was proportional to percent protein,but the tadpoles did not develop (i.e.,change Gosner stage) during the one weekfeeding trial. Pandian and Marian (1985a)found that Rana catesbeiana tadpoleswhose rations were supplemented with tubi-fex worms had shorter larval periods andgrew faster than tadpoles fed ad libitumnon-animal diets. Nathan and James (1972)found that addition of protozoans to a boiledlettuce diet decreased time to metamorpho-sis in Bufo tadpoles. Although tadpoles fedonly lettuce also successfully metamor-phosed, the addition of live protozoans inmud or of sterilized mud increased size at,and decreased time to metamorphosis. Inthe sterilized mud treatment, tadpoles wereseen swimming upside down at the water'ssurface consuming surface films, which theauthors suggested were high in protein andlipids. Additional support for this proteinlimitation hypothesis will be presented inthe section below describing results fromfeeding trials I conducted in a northern Cali-fornia river.

Does the addition of animal protein totadpole diet enhance growth and develop-ment or is the consumption of animal hor-mones the proximate mechanism? Recentexperiments on the effects of tadpole-tadpole cannibalism, which has been docu-mented in at least 11 anuran species (Polisand Meyers, 1985), offer some insight re-garding this question. Hyla pseudopumatadpoles fed conspecifics grow faster andare larger at metamorphosis than tadpolesfed isocaloric diets of heterospecifics, butdo not reach metamorphosis in signifi-cantly shorter periods of time (Crump,


Dow

nloaded from



1990). Carnivorous morphs of spadefoottoad tadpoles (Scaphiopus bombifrons, S.intermontanus, and S. multiplicatus), how-ever, do develop and metamorphose morerapidly than omnivorous morphs that con-sume a combination of algae, detritus, andfairy shrimp (Pfennig, 1990). The largercarnivore morph is triggered by ingestionof shrimp or other tadpoles. Bragg (1956)postulated that the enhanced developmentobserved among cannibals is due to eitheriodine or thyroxine in the prey. Pfennig(1992) confirmed that exposure to exog-enous thyroid hormone indeed producescarnivore morphs. He concluded that inges-tion of other tadpoles, which contain T4,or shrimp, which contain diiodotyrosine(T2), could be the proximate mechanismfor stimulating the development of the car-nivore morph. It is not clear from his in-vestigation, however, if the quantity of T2in the shrimp eaten by tadpoles was equiv-alent to the amount of exogenous T4 re-quired to produce the carnivorous mor-phology. It would be interesting to offerearly stages of spadefoot toad tadpoles achoice between algae and anastrocans todetermine whether they selectively ingesthigh TH foods that induce the carnivorousmorphology. The observation that thyrox-ine causes the same results as a shrimpdiet, however, does not rule out the possi-bility of an indirect causal pathway throughprotein or other limiting nutrients con-tained in the prey affecting the cannibal'srate of TH production. Crump (1986) alsohypothesized that cannibals receive a per-fect balance of nutrients when she observedpremetamorphic Cuban treefrogs, Osteopi-lus septentronalis, preying on older indi-viduals with four limbs. As a possiblemechanism, she cited the increased proteinconversion efficiency (ratio of the amountof amino acids in a tadpole's food to theamount of amino acid in its whole carcass)when tadpoles are fed conspecifics as op-posed to other high protein food sourcessuch as fish meal (Nagai et al., 1971). Todefinitively separate protein effects fromexogenous TH effects, further experimentsin which tadpoles are fed isoproteinaceousdiets and TH metabolism is directly mea-sured are needed.

INTERACTIONS BETWEEN TADPOLES ANDALGAE IN A NORTHERN CALIFORNIA RIVER:

EXAMPLES OF DIET IMPACTS ONMETAMORPHOSIS

Natural history and study systemMy hypothesis regarding protein limita-

tion as a connection between the endocrineand nutritional controls of anuran metamor-phosis stems, in part, from manipulations ofalgal diets and tadpole competition con-ducted in the South Fork of the Eel River(Angelo Coast Range Reserve, MendocinoCo., California (39°44'N, 123°39'W). Theriver has a winter flood/summer drought hy-drograph typical of rivers in Mediterraneanclimates. Like most western rivers, withlimited deciduous canopy cover, the rockbedded river is sun lit, and most productiv-ity comes from algae which bloom duringthe low flows of spring and summer. Nitro-gen is one of the important factors limitingalgal growth in this system (Hill andKnight, 1988; Power, 1992). During the lowflow period, the river is used for breedingby three species of anurans: yellow leggedfrogs, Rana boylii, Pacific treefrogs, Hylaregilla, and non-indigenous bullfrogs, Ranacatesbeiana. Despite their hundred year his-tory in California, bullfrogs are only pres-ently invading the study reach.

The available food resources for tadpoleconsumers include the dominant filamen-tous green alga, Cladophora glomerata, andthe diatoms that grow as epiphytes onCladophora. The epiphytes can completelycover the filaments of the host algae withmean (± 1 SE) diatom biovolume of180,238 ± 18,353 u^/mm of algae (Kupfer-berg, 1997). Tadpoles, R. boylii in particu-lar, can effectively remove diatoms fromCladophora, decreasing average epiphytebiovolume to 78,526 ± 13,343 \i?lmm ofalgae. Diatoms increase the nutritionalvalue of Cladophora for two reasons. Themost common epiphytes, Epithemia spp.,contain nitrogen fixing cyanobacterial en-dosymbionts, so the protein content ofCladophora with epiphytes is 11.3% vs.5.8% without (nutritional data from Kupfer-berg et al., 1994). Epiphytes store excessphotosynthate as lipid rather than carbohy-drate so Cladophora with epiphytes has 1%


Dow

nloaded from



450-

400

350

^ 300rag,£ra 250-

200

150

100-

50"

mass at metamorphosis

time to metamorphosis

< • :"I

10 20% protein

30 40

FIG. 1. Size at, and time to, metamorphosis of Hyla regilla tadpoles fed diets of varying protein content. Eachpoint represents the mean for a diet treatment (n = 6 replicates/treatment and 5 tadpoles/replicate). Error barsindicate ± 1 SE.

fat and 2% non-structural carbohydrates byweight, compared to .56% fat and 2.9% car-bohydrates in Cladophora without epi-phytes. Additional resources are: other fila-mentous green algae, such as Oedogoniumwhich supports only a small epiphyte com-munity (7.8% protein); mucous producingtaxa such as Mougeotia (11.6% protein) andZygnema (21% protein) which have no epi-phytes; and flocculent material containingdetritus, algae, protozoa, and bacteria (.44%protein).

Algal quality affects growth andmetamorphosis

The variation in resource quality amongthese foods influences tadpole growth, de-velopment, and survival to metamorphosis.In field diet studies with H. regilla (Kupfer-berg et al., 1994) fed the above foods aswell as a control food, Tetra Reptomin®(35% protein), size at metamorphosis wassignificantly correlated (Pearson's r = 0.96,P = 0.0006) with protein content (Fig. 1).Although tadpoles fed high protein dietsmetamorphosed sooner than tadpoles fedlow protein diets (Fig. 1), the correlationwas not significant (r = —0.59, P = 0.16).In these experiments total food quantity wasnot controlled, tadpoles were fed ad libitum.Tadpoles were thus free to compensate for

low protein content of a food by eatingmore of it. If compensation did occur therelationship between protein content andperformance would be weakened. An ad li-bitum feeding regime therefore provides aconservative test of the effects of proteincontent.

R. boy Hi were raised in a similar experi-ment (Kupferberg, 1996), to determine ifthe relative profitabilities of foods werealike. Food treatments were Cladophorawith a heavy growth of epiphytes, Clad-ophora without epiphytes, Zygnematales(including Mougeotia and Spirogyra spp.),and finely ground Reptomin®. Tadpolesfrom a single clutch of eggs were kept inthe river in flow through enclosures (12.7liter, 30 cm diameter plastic buckets withtwo windows, 23 X 31 cm, covered with 1mm fiberglass mesh, five tadpoles perbucket (70/m2), ambient density = 29.4 ±30.8/m2, range = 0 - 112/m2, n = 14).Tadpoles were weighed each week andmean per capita weight calculated for eachreplicate. Dead tadpoles were replaced withsize matched individuals from extra repli-cates. Treatment effects on size were com-pared using profile analysis. High proteindiets generated significant differences intadpole growth rate and total size (Table 1).The significant interaction between date and


Dow

nloaded from



TABLE 1. Profile analysis o/Rana boylii growth trajectories fed different quality diets, and Bonferroni multiplecomparison probabilities of logarithmically transformed sums of tadpole mass on days 14, 28, 36, 50, and 64.

Source Mean square Wilks' lambda

DateDate X Treatment

TreatmentError

Analysis of differences (MANOVA)0.0390.153

Analysis of totals

412

(ANOVA)

, 13,34

3

80.342.59

9.29

0.49 X0.014

0.86 X

io-8

io-30.3590.039 16

Treatment

Reptominepiphytized Cladophoraclean CladophoraZygnematales

Reptomin

10.80.00060.06

Multiple comparisonsEpiphytized Cladophora

10.021

Clean Cladophora

10.3

Zygnematales

I

treatment indicates that the growth trajecto-ries were not parallel (Fig. 2). The trajecto-ries of tadpoles fed commercial food or al-gal food rich in epiphytic diatoms weresteeper than trajectories of tadpoles fed al-gae either cleaned of epiphytes or not sup-porting an epiphyte assemblage. The sig-nificant date effect merely shows that tad-poles grew. The multiple comparisons of

mean total size indicate that Cladophoracleaned of its epiphytes is the worst algalfood but that Cladophora with its diatomepiphytes is not significantly different froma high quality commercial food (Table 1).

The conclusion from these growth stud-ies is that if tadpoles just ate what was avail-able, then seasonal and spatial variation inresources that vary in nutritional content

Reptomin®

epiphytized Cladophora

clean Cladophora

Zygnematales

14I i i

28 42time (days)

FIG. 2. Growth trajectories of Rana boylii tadpoles fed algal and control diets. See Table 1 for comparisonsamong treatments.


Dow

nloaded from



CO

Ea.CO•a15

6 l

4 •

• Rana boylii• Hy/a regillaO Algae alone

EpiphytizedCladophora

CleanCladophora

Oedogonium Mougeotia

Food typeFIG. 3. Algal weights after 1 week in the presence or absence of Hyla regilla and Rana boylii tadpoles in acafeteria food selectivity experiment. Histograms joined by horizontal bars are not significantly different fromeach other (Tukey's HSD P > 0.05).

could have a large effect on size and tim-ing of metamorphosis. Selective foragingby tadpoles among these food types couldfurther reinforce the importance of dietquality such that size at metamorphosis ismaximized and time to metamorphosisminimized.

Effects of tadpole foraging behavior onmetamorphosis: Food preference andpredator avoidance

Do tadpoles feed selectively on the foodswhich promote early metamorphosis andlarge size at metamorphosis? Based on gutanalyses of tadpoles in well mixed algalpond environments (Farlowe, 1928; Jens-sen, 1967; Heyer, 1972, 1976; Nathan andJames, 1972; Seale, 1980; Diaz-Paniagua,1985; Loschenkohl, 1986), laboratory feed-ing trials on suspended particles (Wasser-sug, 1972; Seale and Beckvar, 1980), andon the morphological adaptations associatedwith suspension feeding (Wassersug, 1975),tadpoles have been considered indiscrimi-nate feeders. In an environment where al-gal food resources are patchily distributedand grow attached to substrates, as in ariver, there is the potential for selectiveforaging. I conducted field experiments

(Kupferberg, 1996) to determine whethertadpoles chose the foods they grew best onand how that behavior was changed bypredators.

Tadpoles preferred algae with relativelyhigh protein contents. Preferences amongfour types of algae, Cladophora with andwithout epiphytes, Mougeotia, and Oedogo-nium, were compared by placing 5 g dampmass of each type in flow through bucketenclosures. Enclosures were stocked withgroups of 5 R. boylii or 5 H. regilla tadpoles(n = 10 enclosures/tadpole species) or al-gae only (n = 3). After 7 days, algae wereharvested and reweighed after spinning thealgae in a lettuce spinner 50 revolutions(Hay, 1986). For each alga the amount re-maining in Rana, Hyla, and control repli-cates was compared with one-way ANOVA.Grazing treatments with R. boylii and H.regilla significantly reduced the mass ofboth Cladophora with epiphytes (F22o

=

7.1, P = 0.005) and Mougeotia (F220 = 4.0,P = 0.03) relative to controls (Fig! 3). Tad-poles avoided the lower protein foods, withno significant differences observed betweentadpole and algae only treatments: cleanCladophora (F2 20 = 0.74, P = .5) or Oe-dogonium (F2 20 = 3.3, P = .06).


Dow

nloaded from



I experimentally examined how preda-tors, western aquatic garter snakes (Tham-nophis atratus, formerly T. couchii) affectedH. regilla tadpole numbers and selectiveforaging behavior (Kupferberg, 1996). Inreplicated (n = 24) fenced 1 m diameterpools dug into gravel bars, I observedwhether tadpole patch choice follows therule of minimizing the ratio of mortalityrisk, (JL, to foraging gain, g (Werner and Gil-liam, 1984; Werner, 1986). Enclosures werestocked with weighed tadpoles, 260 gCladophora in a mat form, and 260 g ofMougeotia which took the shape of a float-ing cloud. Twelve randomly chosen pondsreceived a similarly sized garter snake. Onday 9 tadpole feeding activity and locationin Cladophora, Mougeotia, or sediment,was repeatedly spot checked. After the fi-nal spot check, tadpoles and algae were har-vested and weighed. Response variableswere the number of survivors, tadpolegrowth, proportion actively feeding, andproportion in each substrate type. Snake ef-fects were evaluated with Bonferroni ad-justed Mests.

Garter snakes did not significantly reducethe number of tadpoles but did alter patchchoice. Tadpoles spent significantly (t =4.4, P = 0.002) more time in low food qual-ity sediments when a snake was present(66 ± 3%) than when snakes were excluded(53 ± 1%). Avoidance of Cladophora in thepresence of garter snakes significantly (t =- 5 . 1 , P = 0.0003) decreased tadpolegrowth by 28% (snake absent = 45.7 ± 1.5vs. snake present 32.9 ± 2 . 1 mg/day). Inmortality risk experiments, however, therewere no significant differences in the num-bers of tadpoles consumed among patchtypes, and garter snakes were not size se-lective. Therefore, patch choice did not fol-low the rule of minimizing (x/g. The changein patch choice was likely due to the suble-thal effect of tadpoles being 56% less ac-tive in the presence of a snake, a commontadpole response to the presence of preda-tors (Lawler, 1989; Werner, 1991; Skelly,1994).

My observation that tadpoles moved lessin the presence of snakes indicates that thetrade-offs tadpoles make between mortality

and growth, which in turn determine thesize at and time to metamorphosis (Werner,1986), are manifest in activity level ratherthan habitat use per se. Tadpoles spent lesstime eating high quality algal foods, notbecause they were inherently risky places,but because feeding activity was risky.For negatively buoyant tadpoles, whichH. regilla at the Eel River appear to be(S. Kupferberg, personal observation) de-creased activity results in sinking awayfrom floating algal resources. It is interest-ing to note that Anholt and Werner (1995)have found that food level and predatorpresence interact. At high food levels tad-poles adapt their behavior by decreasing ac-tivity and hence decreasing mortality, but atlow food levels tadpoles must be more ac-tive to get food and thus suffer higher mor-tality. These predator effects on tadpoles re-late to issues of hormonal control of meta-morphosis via two pathways. Predators caninfluence food consumption, which in turnaffects TH function. Predators could also di-rectly increase the levels of corticoid stresshormones, which also influence TH.

Tadpole effects on algae: Basis forunderstanding effects of competitionon metamorphosis

Because food quantity and quality havesignificant effects on tadpole growth andmetamorphosis, the impact of tadpoles onthe biomass and species composition of al-gal assemblages can determine the outcomeof tadpole competition. Tadpole effects onstanding stocks of algal biomass includedepletion by consumption (Dickman, 1968;Seale, 1980; Bronmark et al., 1991), en-hancement by nutrient regeneration (Os-borne and McLachlan, 1985), and facilita-tion of macroalgae via epiphyte removal(Kupferberg, 1997). Because tadpole mouthand feeding morphologies vary among taxa,interspecific differences in exploitative abil-ity (Wassersug, 1972; Diaz-Paniagua, 1985;Kupferberg, 1997) lead to competitive inter-actions among tadpoles that are more com-plex than reduction of total food availabil-ity.

For example, the impact of non-indigenous bullfrog tadpoles, R. catesbei-


Dow

nloaded from



° 00is1 o

1.6-,

1.4-

1.2-

with bullfrogswithout bullfrogs

0.8 h

0.64 r-0.0 0.1 0.2 0.3 0.4 0.0

floating algaelog(g AFDM/enclosure)

1.4-

1.2-

1.0-

0.8-

0.6-

O

*

8

•

oo

••o

• a

0.5 1.0 1.5 2.0

attached algaelog (mg AFDM / sq cm)

FIG. 4. Relationship between the production of native frogs from pens enclosing or excluding bullfrog tadpolesand (a) high food quality floating algae, Cladophora with epiphytic diatoms and Nostoc (r2 = 0.79, P = 0.007);and (b) low food quality attached algae, Cladophora with few epiphytes (r2 = 0.15, P = 0.67).

ana, as novel grazers and competitors in theEel River, was due to their effect on highprotein algae rather than on total algal bio-mass. In 2 m2 enclosures of natural riversubstrates and ambient densities of nativetadpoles, bullfrog tadpoles caused a 48% re-duction in survivorship of/?, boylii to meta-morphosis, and a 24% decline in mass atmetamorphosis, relative to native only con-trol enclosures. Bullfrog larvae had smallerimpacts on H. regilla, causing 16% reduc-tion in metamorph size, and no significanteffect on survivorship. Four variables weretracked in this experiment. Production ofeach native frog was calculated as the sumof the masses of each species of nativemetamorph leaving an enclosure. An esti-mate of high food quality algal productionwas based on rank transformations of float-ing algal biomass consisting of senescing,heavily epiphytized Cladophora, and Nos-toc. Production of lower food quality at-tached algae was based on biomass datafrom the midpoint of the experiment, indi-cating the bullfrog modified conditions thatnative grazers were experiencing.

Bullfrog impacts on algal food resourcescan explain the effects on metamorphosis ofnatives (Fig. 4), but differences in foodquality among algae must be considered inorder to understand the interactions. Therewas no significant correlation between na-tive frog production and total algal biomass

(Table 2), but lumping all potential foodinto a pooled biomass category, althoughcommon, is not well justified (Mittelbachand Osenberg, 1994). When resources werepartitioned into attached and floating algae,diminished food quality in the bullfrog en-closures is suggested by the decreased abun-dance of floating algae harvested in bullfrogenclosures (0.046 ± 0.026 g AFDM, mean± SD) than in enclosures without bullfrogs(0.325 ± 0.182). Although the treatment ef-fect on floating algae was not significant(Mann-Whitney U = 2.56, n = 12 enclo-sures, P = 0.109), there was a significantcorrelation between ranked floating algalmass and the total mass of R. boylii meta-morphosing from the enclosures (Table 2). Asmall impact of bullfrogs on algal quality,thus resulted in a large impact on Rana boyliibecause there is such a strong correlation be-tween algal quality and the biomass of R.boylii metamorphosing from the enclosures.Only a small portion of the variation in Hylawas explained. Hyla may be able to extractlow quality or low availability resources notmeasured with my sampling regimes. Forexample, I observed Hyla tadpoles swim-ming on their backs at the surface grazingepineustic films of diatoms.

CONCLUSIONS

My research at the Eel River shows howfood resources can influence anuran meta-


Dow

nloaded from



TABLE 2. Observed correlations and descriptive statistics of variables in the bullfrog manipulations (n = 12enclosures).

C = Rana catesbeianapresence vs. absence (1 vs. 0)

H = Hyla regillatotal metamorph biomass (g)

B = Rana boyliitotal metamorph biomass (g)

A = Total algal standing stockash free dry mass (g)

Q = Quality of algae as tadpole food(rank)

c1

-0.15

-0.63

0.18

-0.48

H

1

0.53

-0.12

0.51

B

1

0.07

0.75

A

1

-0.07

Q

1

Mean

0.5

2.58

11.7

24.7

6.5

S D .

0.52

2.14

7.9

16.7

3.6

s.s.

2.8

50.3

688.6

3050.2

142.0

morphosis. In addition to abundance, thequalitative differences among food types,such as the amount of protein in differentalgae, can affect size at and time to meta-morphosis. Tadpoles can select the foodsthey grow best on, and the presence ofpredators can induce changes in activity thatalter selectivity. As grazers, tadpoles influ-ence algal assemblages, thus changing foodquality for competitors. Through my reviewof the literature I speculate that these foodeffects may be mediated through diet in-duced changes in thyroid function.

For future research I suggest that diet ex-periments be conducted in which tadpolesare fed different types of food with varyingamounts of protein, carbohydrate, and lipidand are collected at similar stages for thy-roid function assays. Specifically, does theamount of TH produced by tadpoles differamong treatments or does a tadpole requirea specific amount of TH to reach a particu-lar stage?

ACKNOWLEDGMENTS

I thank T. Hayes for organizing the sym-posium, inviting my participation, and dis-cussing ideas presented here; R. J. Wasser-sug for valuable comments made on an ear-lier draft; and two anonymous reviewers.The following organizations provided sup-port for the research reported here: NSFgrants BSR91-00123 and DEB 93-17980 toM. E. Power; Sigma Xi, Theodore Roos-evelt Fund of the American Museum ofNatural History, Phi Beta Kappa, Universityof California Natural Reserve System, andNASA Global Change Fellowship NGT-30165 to S. J. K.

REFERENCES

Alford, R. A. and R. N. Harris. 1988. Effects of lar-val growth history on anuran metamorphosis. Am.Nat. 131:91-106.

Ahlgren, M. O. and S. H. Bowen. 1991. Growth andsurvival of tadpoles (Bufo americanus) fed amor-phous detritus derived from natural waters. Hydro-biologia 218:49-51.

Anholt, B. R. and E. E. Werner. 1995. Interaction be-tween food availability and predation mortalitymediated by adaptive behavior. Ecology 76:2230—2234.

Atterwill, C. K., C. G. Brown, C. Conybeare, C. W.Holland, and C. A. Jones. 1989. Relation be-tween dopaminergic control of pituitary lactotrophfunction and deceleration of age-related changesin serum prolactin of diet-restricted rats. FoodChem. Toxicol. 27:97-104.

Audo, M. C, T. M. Mann, T. L. Polk, C. M. Louden-slager, W. J. Diehl, and R. Altig. 1995. Food dep-rivation during different periods of tadpole (Hylachrysoscelis) ontogeny affects metamorphic per-formance differently. Oecologia 103:518-522.

Berven, K. A. and B. G. Chadra. 1988. The relation-ship among egg size, density and food level on lar-val development in the wood frog (Rana syl-vatica). Oecologia 75:67—72.

Blaustein, L. and B. P. Kotler. 1993. Oviposition habi-tat selection by the mosquito, Culiseta longiareo-lata: effects of conspecifics, food, and green toadtadpoles. Ecol. Ent. 18:104-108.

Bold, H. C. and M. J. Wynne. 1985. Introduction tothe algae. Prentice-Hall Inc., Englewood Cliffs,New Jersey.

Bragg, A. N. 1956. Further observations on spadefoottoads. Herpetologica 12:201-204.

Brockelman, W. Y. 1969. An analysis of density de-pendent effects and predation in Bufo americanustadpoles. Ecology 50:632-644.

Bronmark, C, S. D. Rundle, and A. Erlandsson. 1991.Interactions between freshwater snails and tad-poles: Competition and facilitation. Oecologia87:8-18.

Buscaglia, M., J. Leloup, and A. De Luze. 1985. Therole and regulation of monodeiodination of thy-roxine to 3,5,3'-triiodothyronine during amphibianmetamorphosis. In M. Balls and M. Brownes


Dow

nloaded from



(eds.), Metamorphosis, The Eighth Symposium ofthe British Soc. for Dev. Biology, pp. 273-293.Clarendon Press, Oxford.

Calef, G. W. 1973. Natural mortality of tadpoles in apopulation of Rana aurora. Ecology 54:741—758.

Carlson, H. E. 1989. Prolactin stimulation by proteinis mediated by amino acids in humans. J. Clin. En-docrinol. Meta. 69:7-14.

Chapin, F. S., Ill, E.-D. Schulze, and H. A. Mooney.1990. The ecology and economics of storage inplants. Ann. Rev. Ecol. Syst. 21:423-447.

Crump, M. L. 1983. Opportunistic cannibalism byamphibian larvae in temporary aquatic environ-ments. Am. Nat. 121:281-289.

Crump, M. L. 1986. Cannibalism by younger tad-poles: Another hazard of metamorphosis. Copeia1986; 1007-1009.

Crump, M. L. 1990. Possible enhancement of growthin tadpoles through cannibalism. Copeia 1990:560-564.

DeBenedictis, P. A. 1974. Interspecific competitionbetween tadpoles of Rana pipiens and Rana syl-vatica: An experimental field study. Ecol. Mono.44:129-151.

DePew, C. L., D. L. Thompson, Jr., J. M. Fernandez,L. S. Sticker, and D. W. Burleigh. 1994. Changesin concentrations of hormones, metabolites, andamino acids in plasma of adult horses relative toovernight feed deprivation and followed by apellet-hay meal fed at noon. J. Anim. Sci.72:1530-1539.

Delidow, B. C, 1989. Reevaluation of the effects ofgrowth hormone and prolactin on anuran tadpolegrowth and development. J. Exper. Zool. 249:279-283.

Denver, R. J. 1997. Proximate mechanisms of phe-notypic plasticity in amphibian metamorphosis.Amer. Zool. 37:172-184.

Diaz-Paniagua, C. 1985. Larval diets related to mor-phological characters of five anuran species in thebiological reserve of Donana (Huelva, Spain).Amphibia-Reptilia 6:307-322.

Dickman, M. 1968. The effect of grazing by tadpoleson the structure of a periphyton community. Ecol-ogy 49:1188-1190.

Eales, J. G. 1988. The influence of nutritional stateon thyroid function in various vertebrates. Amer.Zool. 28:351-362.

Eales, J. G., D. A. Higgs, L. M. Uin, D. L. MacLatchy,O. Bres, J. R. McBride, and B. S. Dosanjh. 1990.Influence of dietary lipid and carbohydrate levelsand chronic 3,5,3'-triiodo-L-thyronine treatmenton thyroid function in immature rainbow trout,Oncorhynchus mykiss. Gen. Comp. Endo. 80:146—154.

Eales, J. G., D. L. MacLatchy, D. A. Higgs, and B. S.Dosanjh. 1992. The influence of dietary proteinand caloric content on thyroid function and hepaticthyroxine 5'-monodeiodinase activity in rainbowtrout, Oncorhynchus mykiss. Can. J. Zool. 70:1526-1535.

Farlowe, V. 1928. Algae of ponds as determined byan examination of the intestinal contents of tad-poles. Biol. Bull. 55:443^148.

Galton, V. A. 1988. The role of thyroid hormone inamphibian development. Amer. Zool. 28:309-318.

Goldman, J. C , J. J. McCarthy, and D. G. Peavey.1979. Growth rate influence on the chemical com-position of phytoplankton in oceanic waters. Na-ture 279:210-215.

Hay, M. E. 1986. Associational plant defenses and themaintenance of species diversity: Turning com-petitors into accomplices. Am. Nat. 128:617-641.

Hensley, F. R. 1993. Ontogenetic loss of phenotypicplasticity of age at metamorphosis in tadpoles.Ecology 74:2405-2412.

Heyer, W. R. 1972. Food item analysis of mixed tad-pole populations from Thailand. Herpetol. Rev.4:127.

Heyer, W. R. 1976. Studies in larval amphibian habi-tat partitioning. Smithson. Contrib. Zool. 242:1-27.

Hill W. R. and A. W. Knight. 1988. Nutrient and lightlimitation of algae in two northern Californiastreams. J. Phycol. 24:125-132.

Himick, B. A., D. A. Higgs, and J. G. Eales 1991.The acute effects of alteration in the dietary con-centrations of carbohydrate, protein, and lipid onplasma T4, T3, and glucose levels in rainbow trout,Oncorhynchus mykiss. Gen. Comp. Endo. 82:451—458.

Jenssen, T. A. 1967. Food habits of the green frog,Rana clamitans, before and during metamorpho-sis. Copeia 1967:214-218.

Jungfer, K. H., and L. C. Schiesari. 1995. Descrip-tion of a central Amazonian and Guianan tree frog,genus Osteocephaius (Anura, Hylidae), withoophagous tadpoles. Alytes 13:1 — 13.

Kloren, W. R. L., B. W. Norton, and M. J. Waters.1993. Fleece growth in Australian cashmeregoats: I. The effects of nutrition and age on fleecegrowth, prolactin and thyroxine concentration.Australian J. Agric. Research 44:1003-1021.

Kupferberg, S. J. 1996. The ecology of native tad-poles {Rana boylii and Hyla regiila) and the im-pact of invading bullfrogs (Rana catesbeiana) ina northern California river. PhD. Diss. Universityof California Berkeley.

Kupferberg, S. J. 1997. Facilitation of periphyton pro-duction by tadpole grazing: Functional differencesbetween species. Freshwater Biol. (In press)

Kupferberg, S. J., J. C. Marks, and M. E. Power. 1994.Effects of variation in natural algal and detrital di-ets on larval anuran (Hyla regiila) life-historytraits. Copeia 1994:446-457.

Landefeld, T. D., F. J. P. Ebling, J. M. Suttie, L. A.Vannerson, V. Padmanabhan, I. Z. Beitins, andD. L. Foster. 1989. Metabolic interfaces betweengrowth and reproduction: II. Characterization ofchanges in messenger RNA concentrations of go-nadotropin subunits, growth hormone, and prolac-tin in nutritionally growth-limited lambs and thedifferential effects of increased nutrition. Endocri-nology 125:351-356.

Lannoo, M. J., D. S. Townsend, and R. J. Wassersug.1987. Larval life in the leaves: Arboreal tadpoletypes, with special attention to the morphology,ecology, and behavior of the oophagous Osteopi-


Dow

nloaded from



lus brunneus (Hylidae) larva. Fieldiana Zool.1987:I-IV, 1-32.

Lawler, S. P. 1989. Behavioural responses and pre-dation risk in four species of larval anurans. Anim.Behav. 38:1039-1047.

Lee, C. W., B. Bitter-Suermann, L. A. Bourque, andR. J. Wassersug. 1993. Epidermal growth factor-like immunoreactivity in the buccopharyngeal mu-cous glands of Xenopus laevis tadpoles. Gen.Comp. Endocrinol. 89:82-90.

Loschenkohl, A. 1986. Niche partitioning and com-petition in tadpoles. In Z. Rocek (ed.), Studies inherpetology, Proc. 3rd Ann. Gen. Mtg. SocietasEuropean Herpetologica, pp. 399-402. CharlesUniversity, Prague.

Marine, N., J. M. Hershman, M. H. Maxwell, L. P.Dornfeld, and P. Schroth. 1991. Dietary restric-tion on serum thyroid hormone levels. Am. J. Med.Sci. 301:310-313.

McAttee, J. W. and A. Trenkle. 1971. Effects of feed-ing, fasting, glucose or arginine on plasma prolac-tin levels in bovine. Endocrinology 89:730—734.

McGuire, M. K., H. Pachon, W. R. Butler, and K. M.Rasmussen. 1995. Food restriction, gonado-tropins, and behavior in the lactating rat. Physiol.Behav. 58:1243-1249.

Mittelbach, G. G. and C. W. Osenberg. 1994. Usingforaging theory to study trophic interactions. InD. J. Stouder, K. L. Fresh, and R. J. Feller (eds.).Theory and application in fish feeding ecology.Belle W. Baruch Library in Marine Sciences, no.18, pp. 45—59. University of South Carolina Press,Columbia, South Carolina.

Nagai, Y., S. Nagai, and T. Nishikawa. 1971. The nu-tritional efficiency of cannibalism and an artificialfeed for the growth of tadpoles of Japanese toad(Bufo vulgaris) Agr. Biol. Chem. 35:697-703.

Nathan, J. M. and V. G. James. 1972. The role of pro-tozoa in the nutrition of tadpoles. Copeia1972:669-679.

Osborne, P. L. and A. J. McLachlan. 1985. The ef-fect of tadpoles on algal growth in temporary rain-filled rock pools. Freshwater Biol. 15:77—87.

Pandian, T. J. and M. P. Marian. 1985a. Predictinganuran metamorphosis and energetics. Physiol.Zool. 58:538-552.

Pandian, T. J. and M. P. Marian. 1985ft. Time and en-ergy costs of metamorphosis in the Indian bull-frog, Rana tigrina. Copeia 1985:653—662.

Pfennig, D. W. 1990. The adaptive significance of anenvironmentally-cued developmental switch in ananuran tadpole. Oecologia 85:101-107.

Pfennig, D. W. 1992. Proximate and functional causesof polyphenism in an anuran tadpole. Func. Ecol.6:167-174.

Pfennig, D. W., A. Mabry, and D. Orange. 1991. En-vironmental causes of correlation between age andsize at metamorphosis in Scaphiopus multiplied-tus. Ecology 72:2240-2248.

Pimm, S. L. and R. L. Kitching. 1987. The determi-nants of food chain length. Oikos 50:302-307.

Polis, G. A. and C. A. Meyers. 1985. A survey of in-traspecific predation among reptiles and amphib-ians. J. Herp. 19:99-107.

Power, M. E. 1992. Hydrologic and trophic controlsof seasonal algal blooms in northen California riv-ers. Archivs fur Hydrobiologie 125:385-410.

Riley, W. W. Jr., D. A. Higgs, B. S. Dosanjh, and J. G.Eales. 1993. Influence of dietary amino acidcomposition on thyroid function of juvenile rain-bow trout, Oncorhyncus mykiss. Aquaculture112:253-269.

Rodgers, B. D., L. M. H. Helms, and E. G. Grau. 1992.Effects of fasting, medium glucose, and aminoacid concentrations on prolactin and growth hor-mone release, in vitro, from the pituitary of the ti-lapia Oreochromis mossambicus. Gen. Comp. En-docrinol. 86:344-351.

Seale, D. B. 1980. Influence of amphibian larvae onprimary production, nutrient flux, and competitionin a pond ecosystem. Ecology 61:1531-1550.

Seale, D. B. and N. Beckvar. 1980. The comparativeability of anuran larvae (genera: Hyla, Bufo, andRana) to ingest suspended blue green algae. Co-peia 1980:495-503.

Shi, Y. 1994. Molecular biology of amphibian meta-morphosis: A new approach to an old problem.Trends Endo. Met. 5:14-20.

Shi, Z.-X., A. Levy, and S. L. Lightman. 1993. Theeffect of dietary protein on thyrotropin-releasinghormone and thyrotropin gene expression. BrainRes. 606:1-4.

Skelly, D. K. 1994. Activity level and the suscepti-bility of anuran larvae to predation. Anim. Behav.48:465-468.

Steinwachser, K. and J. Travis. 1983. Influence offood quality and quantity on early growth of twoanurans. Copeia 1983:238-242.

Sterner, R. W. 1993. Daphnia growth on varyingquality of Scenedesmus: Mineral limitation ofzooplankton. Ecology 74:2351-2360.

Sticker, L. S., D. L. Thompson, Jr., J. M. Fernendez,L. D. Bunting, and C. L. DePew. 1995. Dietaryprotein and(or) energy restriction in mares: Plasmagrowth hormone, IGF-I, prolactin, cortisol, andthyroid hormone responses to feeding, glucose,and epinephrine. J. Anim. Sci. 73:1424—1432.

Storre, R. 1994. Predation of cattail (Typha latifolia)seedlings by tadpoles. Hortus Northwest 5:1.

Tata, J. R., A. Kawahara, and B. S. Baker. 1991. Pro-lactin inhibits both thyroid induced morphogenesisand cell death in cultured amphibian larval tissues.Develop. Biol. 146:72-80.

Tejedo, M. and R. Reques. 1994. Does larval growthhistory determine timing of metamorphosis in an-urans? A field experiment. Herpetologica 50:113—118.

Thomas, G. B., J. E. Mercer, T. Karalis, A. Rao, J. T.Cummins, and I. J. Clarke. 1990. Effect of re-stricted feeding on the concentrations of growthhormone (GH), gonadotropins, and prolactin(PRL) in plasma, and on the amounts of messen-ger RNA for GH, gonadotropin subunits, and PRLin the pituitary glands of adult ovarectomizedewes. Endocrinology 126:1361-1367.

Travis, J. 1984. Anuran size at metamorphosis: Ex-perimental test of a model based on intraspecificcompetition. Ecology 65:1155—1160.


Dow

nloaded from



Voss, S. R. 1995. Genetic basis of paedomorphosisin the axolotl, Ambystoma mexicanum: A test ofthe single gene hypothesis. J. Heredity 86:441-447.

Wagner, W. E., Jr. 1986. Tadpoles and pollen: Obser-vations on the feeding behavior of Hyla regilla lar-vae. Copeia 1986:802-304.

Warkentin, K. M. 1992a. Effects of temperatureand illumination on feeding rates of green frogtadpoles (Rana clamitans). Copeia 1992:725-730.

Warkentin, K. M. \992b. Microhabitat use and feed-ing rate variation in green frog tadpoles (Ranaclamitans). Copeia 1992:731-740.

Wassersug, R. J. 1972. The mechanism of ultra-planktonic entrapment in anuran larvae. J. Morph.137:279-288.

Wassersug, R. J. 1975. The adaptive significance ofthe tadpole stage with comments on the mainte-nance of complex life cycles in anurans. Amer,Zool. 15:405^17.

Wassersug, R. J. 1997. Where the tadpole meets theworld—Observations and speculations on biome-chanical and biochemical factors that influencemetamorphosis in anurans. Amer. Zool. 37:124—136.

Werner, E. E. 1986. Amphibian metamorphosis:Growth rate, predation risk and the optimal sizeat transformation. Am. Nat. 128:319-341.

Werner, E. E. 1991. Nonlethal effects of a predatoron competitive interactions between two anuranlarvae. Ecology 72:1709-1720.

Werner, E. E., and J. F. Gilliam. 1984. The ontoge-netic niche and species interactions in size-structured populations. Ann. Rev. Ecol. Syst.15:393^125.

White, B. A., and C. S. Nicoll. 1981. Hormonal con-trol of amphibian metamorphosis. In L. I. Gilbertand E. Frieden (eds.), Metamorphosis: A problemin developmental biology pp. 363-396. PlenumPress, New York.

Wilbur, H. M. 1977a. Density-dependent aspects ofgrowth and metamorphosis in Bufo americanus.Ecology 48:196-200.

Wilbur, H. M. 19776. Interaction of food level andpopulation density in Rana sylvatica. Ecology58:206-209.

Wilbur, H. M. 1980. Complex life cycles. Annu. Rev.Ecol. Syst. 11:67-93.

Wilbur, H. M., and J. P. Collins. 1973. Ecological as-pects of amphibian metamorphosis. Science182:1305-1314.


Dow

nloaded from


Documents

by guest on July 11, 2016 ...angelo.berkeley.edu/wp-content/uploads/Kupferberg... · 1988, Berven and Chadra, 1988) can deter-mine tadpole growth and timing of meta-morphosis. These