Capa Tese (com foto)

Universidade Federal de Minas Gerais

Instituto de Ciências Biológicas

Programa de Pós-graduação em Ecologia, Conservação e Manejo

de Vida Silvestre

Comportamento de larvas de Phylloicus sp. (Trichoptera,

Calamoceratidae): um fragmentador característico

de córregos neotropicais

Tese apresentada à Universidade Federal de Minas Gerais, como pré-requisito do Programa de Pós-graduação em Ecologia, Conservação e Manejo de Vida Silvestre, para a obtenção do título de Doutor em Ecologia.

Marcelo da Silva Moretti

Orientador:

Prof. Dr. Marcos Callisto

(Departamento de Biologia Geral, ICB, UFMG)

Co-orientador:

Prof. Dr. Mark O. Gessner

(Departamento de Ecologia Aquática, Eawag)

Belo Horizonte, setembro de 2009.

Instituto de Ciências Biológicas

Programa de Pós-graduação em Ecologia, Conservação e Manejo

de Vida Silvestre

Tese de Doutorado

Comportamento de larvas de Phylloicus sp. (Trichoptera,

Calamoceratidae): um fragmentador característico

de córregos neotropicais

Marcelo da Silva Moretti

Orientador:

Prof. Dr. Marcos Callisto

(Departamento de Biologia Geral, ICB, UFMG)

Co-orientador:

Prof. Dr. Mark O. Gessner

(Departamento de Ecologia Aquática, Eawag)

Belo Horizonte, setembro de 2009.

Bolsas:

Financiamento:

Apoio:

Aos meus pais e à Mica por todo amor e

carinho durante esta caminhada.

“The major difference between a thing that might go wrong and a thing that cannot

possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it

usually turns out to be impossible to get at or repair.”

Douglas Adams

V

Agradecimentos

Após diversas idas ao campo, longos dias no laboratório, milhares de discos

de folhas cortados, centenas de fragmentadores “assados”, duas travessias de oceano,

alguns quilinhos a mais e outros fios de cabelo a menos, chegou o momento de

lembrar de todos aqueles que em algum ou em vários momentos estiveram presentes

na minha vida nos últimos 4 anos e me ajudaram a tornar esta tese em uma realidade.

Ao meu Orientador

Ao Prof. Marcos Callisto por ter me orientado também no doutorado e ter

permitido que eu me aventurasse por caminhos que ainda não haviam sido percorridos

dentro do nosso laboratório. Marcos, obrigado pelo apoio, orientação e ensinamentos

durante os últimos 10 anos!

To my Co-supervisor

To Prof. Mark Gessner who kindly accepted me in his research group during

my stay at Eawag. I will always be grateful for his teachings, conversations, advices

and opportunities. Mark, thank you for showing me that a good experiment starts with

a good planning and as much replication as possible!

Aos meus colegas de Laboratório

À Barbara Becker por ter acreditado sempre nas minhas idéias e ter

participado em todos os experimentos desta tese, e à Nayara Costa pela imensa ajuda

e apoio na reta final. Meninas, a disponibilidade de vocês foi fundamental para que

minhas idéias saíssem do papel!

Ao Prof. José Francisco Gonçalves (Júnior) por desde o início ter me

incentivado a trabalhar com decomposição, e à Profa. Adriana Medeiros que elaborou

e executou comigo vários experimentos, sempre acreditando nos hifomicetos do

córrego Garcia. A vocês dois sou imensamente grato pela amizade e conversas

científicas.

À Juliana França, Déborah Oliveira, Clarissa Dantas, Adriana Lessa, Ana

Paula Eller e Lurdemar Tavares pelo apoio nas idas ao campo, ajuda no laboratório,

análises de nutrientes, compras de material e fotografias das larvas. À velha guarda do

VI

laboratório (Raphael Ligeiro, Wander Ribeiro, Pablo Moreno, Augusto Mendes, Ivan

Monteiro e Aline Paz) pela amizade, conversas, trocas de idéias e ajuda no campo. Ao

Seu Jorge (COPASA) por facilitar as coletas, vigiar os experimentos de campo e pelas

boas conversas. Aos novos colegas de laboratório e ao pessoal do Nuvelhas (Cacá,

Diego e turma do Geoprocessamento) pelo ótimo convívio e boas risadas!

To my friends from Eawag

To Maria Alp, Vicenç Acunã, André Barbosa, Aline Frossard, Andreas

Bruder, Diego Tonolla, and Chris Robinson for their friendship. These “gorgeous”

friends were responsible for several special and unforgettable moments during my

time in Switzerland. Thank you guys for taking care of this “meat addicted,

Gammarus killer, hobo, crazy Brazilian” while he was far away from home!

Aos que fazem parte da minha vida

A Deus pela força e clareza nos pensamentos, principalmente nos momentos

mais difíceis. Aos meus colegas de graduação (Trilobitas) pela amizade, passeios,

festas e apoio durante todos esses anos. À Britney, por sempre me considerar o

melhor dono do mundo, mesmo estando a maior parte do tempo ausente. À minha

família pelo carinho e aos meus pais pelo amor, paciência e dedicação. À Michelle

que sempre esteve ao meu lado, sendo minha amiga, companheira e confidente

mesmo quando distante. Mica, obrigado pelo seu amor e por tudo o que você sempre

faz por mim! Sem vocês, eu não teria chegado até aqui!

A todos vocês e a alguém que eventualmente eu possa ter esquecido, MUITO

OBRIGADO!!!

VII

Sumário

Resumo 1

Abstract 4

Apresentação 6

ObjetivoseHipóteses 8

ÁreasdeEstudo 9

Capítulo1 10

Leaf abundance and phenolic concentrations codetermine theselection of casebuilding materials by Phylloicus sp.(Trichoptera,Calamoceratidae)

Capítulo2 25

LengthdrymassrelationshipsforatypicalshredderinBrazilianstreams(Trichoptera,Calamoceratidae)

Capítulo3 36

AreexoticEucalyptusleavesabetterfoodresourcetoPhylloicussp. (Trichoptera, Calamoceratidae) in Brazilian Cerradostreams?

Capítulo4 55

Influenceoflowqualityleavesonthefeedingactivityoftropicalandtemperateshredders

Conclusões 74

ConsideraçõesFinaisePerspectivasFuturas 75

Referências 77

1

Resumo

O objetivo geral desta tese foi avaliar o comportamento de Phylloicus sp.

(Trichoptera, Calamoceratidae) quando exposto a detritos foliares de diferentes

qualidades nutricionais e dureza. Três hipóteses direcionaram os experimentos aqui

apresentados: (1) as larvas de Phylloicus selecionam espécies de folhas para construir

um casulo mais resistente; (2) as folhas de Eucalyptus são preferidas e promovem um

maior crescimento e sobrevivência das larvas de Phylloicus do que as folhas do

Cerrado, com baixa qualidade nutricional; e (3) a atividade dos invertebrados

fragmentadores nos córregos do Cerrado é limitada pela qualidade das folhas

disponíveis. Esta tese é composta de 4 capítulos referentes a experimentos de

laboratório e campo realizados com invertebrados fragmentadores coletados em um

córrego tropical no Brasil e um de região temperada na Suíça.

O Capítulo 1 refere-se a um experimento elaborado para se avaliar o

comportamento de construção de casulos de larvas de Phylloicus quando expostas a

diferentes espécies de folhas. Os principais resultados foram: (a) Eucalyptus e Myrcia

foram mais utilizadas (36,4 e 35,7%) que Protium (20,0%) para a construção de

casulos, enquanto Miconia foi utilizada em uma proporção intermediária (28,6%); e

(b) a seleção de folhas para a construção de casulos foi recurso denso-dependente e as

larvas utilizaram preferencialmente folhas com maiores concentrações de fenóis,

independentemente da dureza e da razão biomassa : área. Foi possível concluir que as

larvas de Phylloicus apresentam ampla plasticidade no comportamento de construção

de casulos e sua preferência foi determinada pela composição química e

disponibilidade das folhas.

O Capítulo 2 refere-se a um trabalho desenvolvido para se determinar qual

medida corporal e qual função matemática poderiam ser utilizadas para melhor

descrever a relação comprimento-biomassa da população de Phylloicus estudada. O

comprimento corporal proporcionou as melhores equações para se estimar a biomassa

das larvas, seguido pela largura da cápsula cefálica e distância interocular, e os

maiores coeficientes de determinação foram encontrados utilizando-se os modelos

potencial e exponencial. Desta forma, concluiu-se que as relações comprimento-

biomassa descritas foram significativas e podem ser utilizadas em experimentos de

2

laboratório que avaliem, por exemplo, as taxas de crescimento e a produção

secundária de larvas de Phylloicus sp.

O Capítulo 3 refere-se a experimentos de laboratório e campo elaborados para

se determinar a preferência alimentar, taxas de crescimento e sobrevivência das larvas

de Phylloicus sp. alimentando-se de folhas de Eucalyptus camaldulensis e de 3

espécies nativas do Cerrado de Minas Gerais (Myrcia guyanensis, Miconia chartacea

e Protium brasiliense). Os principais resultados foram: (a) Eucalyptus foi mais

consumida que Myrcia (0,436 e 0,209 mg mg-1 dia-1) nos experimentos de preferência

alimentar, enquanto Miconia foi consumida em taxas intermediárias (0,230 mg mg-1

dia-1) e Protium não foi consumida; (b) as taxas de crescimento diário de Phylloicus

variaram de 0,08 a 0,14 mg dia-1 e não foram encontradas diferenças entre os

tratamentos alimentares; (c) a sobrevivência das larvas que se alimentaram de

Eucalyptus (68%) e Miconia (64%) foi significativamente maior das que se

alimentaram de Myrcia (50%) e Protium (32%). Foi possível concluir que o

comportamento alimentar de Phylloicus foi mais influenciado pela qualidade das

folhas disponíveis do que sua origem (nativas/exóticas) e as folhas de Eucalyptus não

constituíram um recurso alimentar de maior valor nutricional para as larvas de

Phylloicus do que as folhas de espécies do Cerrado.

No Capítulo 4 foram comparados a preferência alimentar, o consumo e a

sobrevivência de Phylloicus sp. e Gammarus pulex (Amphipoda, Gammaridae), um

fragmentador muito estudado em córregos temperados. As folhas deste experimento

foram escolhidas de forma que fossem constituídos dois pares compostos por espécies

tropicais e temperadas apresentando baixos teores de lignina e nitrogênio (Swietenia

macrophylla e Betula pubescens) e altos teores de lignina e moderados de nitrogênio

(Hymenaea courbaril e Fagus sylvatica). Os principais resultados deste experimento

foram: (a) Phylloicus e Gammarus apresentaram preferências alimentares

semelhantes: Betula foi a espécie mais consumida (Phylloicus: 0,11 a 0,17 mg mg-1

dia-1; Gammarus: 0,14 a 0,22 mg mg-1 dia-1), enquanto Hymenaea não foi consumida

e Fagus e Swietenia foram pouco consumidas pelos fragmentadores; (b) Phylloicus

consumiu mais Swietenia, enquanto Gammarus consumiu mais Swietenia e Betula do

que Hymenaea e Fagus; (c) a sobrevivência de Phylloicus foi extremamente baixa

(9%) e as larvas que se alimentaram de Betula sobreviveram mais do que aquelas que

se alimentaram de Hymenaea; (d) a sobrevivência de Gammarus foi bem maior (80%)

e não foram encontradas diferenças entre os tratamentos alimentares. Desta forma,

3

concluiu-se que Phylloicus e Gammarus apresentaram a mesma preferência alimentar

quando expostos a detritos de baixa qualidade nutricional e o comportamento

alimentar de ambos os fragmentadores foi influenciado principalmente pelas

concentrações de lignina dos detritos foliares.

Estes resultados evidenciaram que as baixas taxas de decomposição que vem

sendo observadas nos córregos do Cerrado estão mais relacionadas à qualidade das

folhas disponíveis do que a um diferente comportamento dos fragmentadores nestes

ecossistemas.

4

Abstract

The objective of this thesis was to evaluate the behavior of Phylloicus sp.

(Trichoptera, Calamoceratidae) when exposed to leaf species differing in nutritional

quality and toughness. The experiments were elaborated according to 3 hypotheses:

(1) Phylloicus larvae select leaf species to build a more resistant case; (2) Eucalyptus

leaves are preferred and promote higher growth and survivorship of Phylloicus larvae

than Cerrado native leaves, with low nutritional quality; and (3) shredders activity in

Cerrado streams is limited by the quality of available leaves. This thesis is composed

by 4 chapters concerning to laboratorial and field experiments done with invertebrate

shredders sampled in a tropical stream in Brazil and in a temperate one in

Switzerland.

The First Chapter refers to an experiment developed to evaluate the case-

building behavior of Phylloicus larvae exposed to different leaf species. The main

results were: (a) Eucalyptus and Myrcia were used more (36.4 and 35.7%) than

Protium (20.0%) for case-building while Miconia was used in an intermediate

proportion (28.6%); and (b) selection of material for case-building was resource

density-dependent and larvae preferentially used leaves with higher phenolic

concentrations independently of toughness and biomass : area ratio. It was concluded

that Phylloicus larvae present a wide plasticity in the case-building behavior and their

preference was determined by leaf chemistry and availability.

The Second Chapter refers to a procedure developed to determine which linear

body dimension was best suitable and which mathematical functions could be used to

describe length dry-mass relationships for the studied population of Phylloicus. Body

length provided the best fitted equations to estimate biomass, followed by head

capsule width and interocular distance, and the highest coefficients of determination

were found in power function and exponential models. It was concluded that the

described length-dry mass relationships were significant and can be used, for

example, in laboratorial experiments to assess the growth rates and secondary

production of these shredders.

The Third Chapter refers to laboratorial and field experiments developed to

determine Phylloicus sp. larvae food preferences, growth rates and survival when

feeding on leaves of Eucalyptus camaldulensis and 3 leaf species native from Cerrado

5

(Brazilian savannah) in Minas Gerais State (Myrcia guyanensis, Miconia chartacea,

and Protium brasiliense). The main results were: (a) Eucalyptus was more consumed

than Myrcia (0.436 and 0.209 mg mg-1 day-1) in the food preference experiment,

while Miconia was consumed on intermediate rates (0.230 mg mg-1 day-1) and

Protium was not consumed; (b) Phylloicus daily growth rates ranged from 0.08 to

0.14 mg day-1 and differences were not found among food treatments; and (c) survival

of larvae fed Eucalyptus (68%) and Miconia (64%) was significantly higher than

those fed Myrcia (50%) and Protium (32%). It was concluded that Phylloicus feeding

behavior was more influenced by the quality than origin (native/exotic) of available

leaves and Eucalyptus leaves did not consist in a resource of higher food quality to

Phylloicus larvae than Cerrado native species.

In the Forth Chapter it was compared the food preferences, consumption rates,

and survival of Phylloicus sp. and Gammarus pulex (Amphipoda, Gammaridae), a

well-studied shredder from temperate streams. Leaves were chosen such we had pairs

composed by tropical and temperate leaves presenting low lignin and nitrogen

contents (Swietenia macrophylla and Betula pubescens) and high lignin and moderate

nitrogen contents (Hymenaea courbaril and Fagus sylvatica). The main results were:

(a) Phylloicus and Gammarus exhibited similar food preferences: Betula was the most

consumed leaf species (Phylloicus: 0.11 to 0.17 mg mg-1 day-1; Gammarus: 0.14 to

0.22 mg mg-1 day-1) while Hymenaea was not consumed, and Fagus and Swietenia

were few consumed by both shredders; (b) Phylloicus consumed more Swietenia

while Gammarus consumed more Swietenia and Betula than Hymenaea and Fagus;

(c) Phylloicus survivorship was extremely low (9%) and larvae fed on Betula

survived more than those fed on Hymenaea; and (d) Gammarus survival was much

higher (80%) and no differences were found among food treatments. It was concluded

that Phylloicus and Gammarus had the same food preferences when exposed to leaves

of low nutritional quality, and the feeding behavior of both shredders was mainly

influenced by leaf lignin contents.

These results evidenced that the low breakdown rates that have been observed

in Cerrado streams are more related to the quality of available leaves than to a

different shredder behavior in these ecosystems.

6

Apresentação

Estudos realizados em diferentes córregos tropicais têm encontrado baixa

abundância de invertebrados fragmentadores (Dobson et al., 2002; Mathuriau &

Chauvet, 2002; Jacobsen et al., 2008; Li & Dudgeon, 2009). Além disso, poucos taxa

de invertebrados aquáticos são caracterizados como fragmentadores de detritos

foliares nestes ecossistemas. Neste sentido, as menores taxas de decomposição de

detritos foliares encontradas nos córregos tropicais, quando comparadas às de região

temperada, têm sido atribuídas principalmente a uma menor participação de

fragmentadores exclusivos neste processo (Irons et al., 1994). Entretanto, alguns

estudos mais recentes demonstraram que os invertebrados fragmentadores também

podem ser abundantes em córregos tropicais (Cheshire et al., 2005).

De acordo com Wantzen & Wagner (2006), duas das possíveis explicações

para a variável contribuição dos fragmentadores no processo de decomposição de

detritos foliares em ecossistemas tropicais são: (1) as características hidrológicas e os

padrões de distúrbios temporais/sazonais dos ecossistemas lóticos, pois as fortes

chuvas sazonais modificam o leito dos córregos e podem afetar a disponibilidade de

recursos alimentares, e (2) as características físicas e químicas das folhas tropicais,

que geralmente são mais duras e apresentam menores teores de nutrientes e maiores

teores de compostos secundários, que podem dificultar o uso destes recursos pelos

fragmentadores.

Os córregos de cabeceira localizados no Cerrado de Minas Gerais apresentam

comunidades de invertebrados bentônicos ricas e diversas. Sua vegetação ripária é

composta por várias espécies que apresentam folhas com características físicas e

químicas contrastantes. Porém, as taxas de decomposição de detritos foliares nestes

ecossistemas são extremamente baixas (Gonçalves et al., 2007; Moretti et al., 2007).

Dentre os invertebrados comumente encontrados associados aos detritos em

decomposição, as larvas do gênero Phylloicus (Trichoptera, Calamoceratidae),

embora pouco abundantes, contribuem com uma porção significativa da biomassa

total destes organismos. Por serem consideradas exclusivamente como

fragmentadoras de matéria orgânica e por também utilizarem os detritos foliares como

matéria prima para a construção de seus casulos, a atividade destas larvas pode

influenciar o processo de decomposição de detritos foliares nestes ambientes.

7

Os experimentos realizados nesta tese avaliaram o comportamento de larvas

de uma população de Phylloicus sp. do manacial Taboões, localizado no Parque

Estadual da Serra do Rola Moça (MG), e estão divididos em 4 capítulos. No Capítulo

1 foi avaliado o comportamento de construção de casulos das larvas de Phylloicus. No

Capítulo 2 foram descritas as equações comprimento-biomassa para 3 medidas

corporais das larvas da população estudada, utilizando-se três modelos matemáticos.

No Capítulo 3 foram avaliados a preferência alimentar, as taxas de crescimento e a

sobrevivência das larvas de Phylloicus quando expostas a folhas de Eucalyptus e de 3

espécies nativas do Cerrado em Minas Gerais. No Capítulo 4 o comportamento

alimentar das larvas de Phylloicus foi comparado ao de Gammarus pulex

(Amphipoda, Gammaridae), um fragmentador típico de regiões temperadas que tem

sido amplamente estudado em experimentos laboratoriais realizados na América do

Norte e Europa. Desta forma, os resultados aqui apresentados representam uma

contribuição ao maior entendimento da real participação dos invertebrados

fragmentadores no processamento de matéria orgânica nos córregos do Cerrado, além

de fornecer novas informações sobre os ecossistemas lóticos tropicais.

8

ObjetivoseHipóteses

Objetivo geral

Avaliar o comportamento de um fragmentador característico dos córregos do

Cerrado de Minas Gerais (Phylloicus sp.) quando exposto a detritos foliares de

diferentes qualidades nutricionais e dureza foliar.

Hipóteses

1. As larvas de Phylloicus selecionam espécies de folhas para construir um

casulo mais resistente.

2. As folhas de Eucalyptus são preferidas e promovem um maior crescimento e

sobrevivência das larvas de Phylloicus do que as folhas do Cerrado, com baixa

qualidade nutricional.

3. A atividade dos invertebrados fragmentadores nos córregos do Cerrado é

limitada pela qualidade das folhas disponíveis.

Objetivos específicos

• Testar se as larvas de Phylloicus selecionam espécies de folhas para a

construção de seus casulos.

• Determinar a melhor relação comprimento-biomassa de uma população de

Phylloicus sp. usando diferentes funções matemáticas e medidas corporais.

• Avaliar os efeitos das folhas de Eucalyptus no comportamento alimentar das

larvas de Phylloicus sp.

• Testar se a preferência alimentar e as taxas de consumo apresentadas pelas

larvas de Phylloicus sp. em experimentos de laboratório refletem as

observadas em campo.

• Testar se Phylloicus e Gammarus exibem a mesma preferência alimentar entre

folhas de diferentes qualidades e se dietas de baixas qualidades afetam o

consumo e a sobrevivência de ambos os fragmentadores de forma similar.

9

ÁreasdeEstudo

Os organismos dos dois taxa de fragmentadores estudados (Phylloicus sp. e

Gammarus pulex) foram coletados, respectivamente, no Manancial Taboões (Brasil) e

no Córrego Chriesbach (Suíça).

O Manancial Taboões (20º03’38" S – 44º03’03" O) localiza-se dentro do

Parque Estadual da Serra do Rola Moça, Minas Gerais. Este manancial está inserido

em um fragmento de mata e apresenta acúmulos de folhas em seu leito durante todo o

ano (Fig. 1). Suas águas apresentam elevadas concentrações de oxigênio (7,2 mg L-1),

pH alcalino (7,9), baixa condutividade elétrica (13,0 µS cm-1) e concentração de

nutrientes (N-total = 35.0 µg L-1 e P-total = 3.0 µg L-1). O Manancial Taboões é

circundado por uma floresta nativa (ca. 92 hectares) que apresenta uma diversa

quantidade de espécies arbóreas, composta principalmente por espécies do Cerrado.

No entanto, algumas espécies características da zona de transição entre o Cerrado e a

Mata Atlântica também estão presentes (COPASA-MG).

O Córrego Chriesbach (47°24’22” N - 08°35’49” L) localiza-se no município

de Dübendorf, a 7 km de Zürich. A microbacia no qual está inserido consiste de áreas

residenciais, industriais e de agricultura. Suas águas são ricas em nutrientes (N-NO3 =

7.0 mg L-1 e SRP = 50.8 µg L-1) e seu leito é esparsamente sombreado, sendo

composto principalmente por areia, cascalho fino e matéria orgânica particulada fina

associada aos bancos de macrófitas (Fig. 1). No verão, os bancos de macrófitas

cobrem quase 100% do leito e são formados principalmente por Ranunculus fluitans

Lam. e Myriophyllum spicatum L. (Kaenel et al., 2000).

1Leafabundanceandphenolic

concentrationscodeterminetheselectionofcasebuildingmaterials

byPhylloicussp.(Trichoptera,Calamoceratidae)*

MarceloS.Moretti1,RafaelD.Loyola2,BárbaraBecker1&MarcosCallisto1

1Lab.EcologiadeBentos,InstitutodeCiênciasBiológicas,UniversidadeFederaldeMinasGerais,Av.AntônioCarlos6627,

C.P.486,30161970,BeloHorizonte,MG,Brasil

2DepartamentodeBiologiaGeral,InstitutodeCiênciasBiológicas,UniversidadeFederaldeGoiás.

C.P.131,74001970,Goiânia,GO,Brasil

LarvaofPhylloicussp.(Trichoptera,Calamoceratidae), selecting leafmaterials for case‐building in alaboratoryexperiment.

*PublishedinHydrobiologia(2009),630,199‐206

Chapter1:Phylloicuscase‐buildingbehavior

11

Leaf abundance and phenolic concentrations codetermine the

selection of case-building materials by Phylloicus sp.

(Trichoptera, Calamoceratidae)

Marcelo S. Moretti, Rafael D. Loyola, Bárbara Becker & Marcos Callisto

Abstract Phylloicus sp. larvae live on leaf patches in slow flowing waters and build

dorso-ventrally flattened cases from leaf pieces. We hypothesized that Phylloicus

larvae are selective towards certain leaf species to build a more resistant case. We

exposed Phylloicus larvae to equal-area leaf discs of 3 plant species from the

Brazilian Cerrado (Myrcia guyanensis, Miconia chartacea and Protium brasiliense)

and one non-native species (Eucalyptus camaldulensis). Phylloicus larvae built cases

with discs of all plant species. However, discs of E. camaldulensis and M. guyanensis

were used more (36.4 and 35.7%, respectively) than those of P. brasiliense (20.0%).

Discs of M. chartacea were used in an intermediate proportion (28.6%). Selection

was resource density-dependent, i.e., when P. brasiliense was offered at higher

abundance, it was used more frequently by larvae (ANOVA, P < 0.001). Plant species

differed in leaf toughness, phenolic concentration, and biomass:area ratio (Kruskal-

Wallis, P < 0.05). Larvae preferentially used leaves with higher phenolic

concentrations (Rs = 0.907, P < 0.001) independently of toughness and biomass:area

ratio. We suggest that Phylloicus selects for case-building leaves that are chemically

protected against microbial degradation and shredder consumption, and this selection

depends on leaf abundance. Our results also reinforce the importance of riparian

resources and their diversity to the maintenance of aquatic consumers in tropical

shaded streams.

Keywords Shredder behavior, Brazilian Cerrado, Leaf traits, Tropical streams,

Phylloicus


12

Introduction

The Trichoptera (Caddisflies) are one of the most taxonomically rich orders of aquatic

insects, with larvae found in nearly all freshwaters (Resh & Rosenberg, 1984; Flint et

al., 1999) and occupying diverse microhabitats and trophic niches (de Moor &

Ivanov, 2008). Part of this diversity results from variable use of silk secreted by the

larvae during netspinning and case-building (Wiggins, 1996, 2004). Case-building is

energetically expensive but protects caddisflies against predators (Otto & Svensson,

1980; Stevens et al., 1999; Otto, 2000; Boyero et al., 2006).

The Calamoceratidae is a cosmopolitan family of eight extant genera, with

over 100 described species (Prather, 2003). The genus Phylloicus Müller, 1880,

includes 61 known species, and is the largest calamoceratid genus in the New World

(Huamantinco et al., 2005). Species are distributed throughout Latin America, but are

especially diverse in Brazil, Peru, and Venezuela (Prather, 2003). Phylloicus larvae

are normally found on submerged leaves in stream pools and/or lateral springs (Flint

et al., 1999; Wantzen & Wagner, 2006). These larvae are exclusive leaf feeders

(shredders), exerting an important role in the conversion of leaf litter into secondary

production and promoting the conversion of coarse particulate organic matter

(CPOM) into fine particulate organic matter (FPOM) in lotic ecosystems (Cummins et

al., 1989; Wallace & Webster, 1996; Flint et al., 1999; Rincón & Martínez, 2006).

Phylloicus larvae build dorso-ventrally flattened cases using several leaf

pieces (Wiggins, 1996). Because they cut leaf pieces themselves, it is possible that

abundance, quality, and toughness of available leaves in stream pools determine case-

building behavior. Rincón & Martínez (2006), studying the feeding preferences of

Phylloicus sp. in Venezuela, observed that these larvae preferred building their cases

from leaves with higher levels of phenolics and lignins.

The objective of this investigation was to test whether Phylloicus larvae select

leaves of particular plant species for case-building. We addressed three questions: (1)

Are Phylloicus larvae capable of building their cases with leaves of only one plant

species? (2) Whenever leaves of more than one species are available, is there a plant

species selected preferentially by the larvae for case-building? (3) Is the mechanism

of plant species selection (if it exists) dependent on resource density?


13

Methods

Larvae sampling site and collection

Larvae were collected at Taboões spring (20° 03' 38" S - 44° 03' 03" W), located

inside the Serra do Rola Moça State Park, Minas Gerais State, southeastern Brazil.

The spring has a canopy of riparian trees with fallen leaves accumulating in patches

within the spring all year. On the sampling date, abiotic parameters (temperature, pH,

electrical conductivity, redox, dissolved oxygen, total dissolved solids and turbidity)

were measured in situ using a Horiba multi-probe (Horiba, Irvine, CA, USA). Total

nitrogen and total phosphorus concentrations were determined according to APHA

(1992). Spring waters were well oxygenated, alkaline with low conductivity and

nutrient concentrations (Table 1).

Table 1 Water properties of Taboões spring (Serra do Rola Moça State Park, Minas Gerais, Brazil). Values from a single observation.

Parameters

Temperature (°C) 20.2 pH 7.9 Electrical conductivity (µS cm-1) at 25 °C 13.0 Redox (mV) 184.0 Dissolved oxygen (mg L-1) 7.2 Total N (µg L-1) 35.0 Total P (µg L-1) 3.0 Total dissolved solids (µg L-1) 10.0 Turbidity (NTU) 8.0

One hundred and twenty larvae of one undetermined species of Phylloicus

(Trichoptera: Calamoceratidae) were collected on June 23rd 2008, with a hand net.

They were taken to the laboratory in an isothermic box with stream water. In the

laboratory, they were acclimatized for 24 hours in an aquarium (80 cm long, 20 cm

wide, 40 cm high) with spring water and a bottom of fine gravel. The aquarium was

aerated continuously and some leaves, collected together with the insects in the same

pools, were offered as food.


14

Leaves

In our experiments, we used senescent leaves of three plant species native to the

Brazilian Cerrado (Myrcia guyanensis Aubl., Miconia chartacea Triana and Protium

brasiliense Engl.) and one non-native species (Eucalyptus camaldulensis Dehn.).

These leaves were collected from plastic nets (1 m2, 10 mm mesh size, approximately

1.5 m height) implanted in the riparian zone (native species) and in a monoculture of

E. camaldulensis located nearby. All leaves were air dried, sorted by species, and

stored at room temperature until needed.

We chose the native species because they are abundant in the riparian forests

and present different breakdown rates in Brazilian Cerrado streams. M. guyanensis

shows the highest breakdown rate (k = 0.0063 day–1) followed by M. chartacea (k =

0.0033 day–1) and P. brasiliense (k = 0.0020 day–1; Moretti et al., 2007). Eucalyptus

camaldulensis was used in the experiments because many Cerrado areas have been

replanted with eucalyptus monocultures (Klink & Machado, 2005). Eucalyptus trees

have invaded the riparian forests of some Cerrado streams and the consequences of

this change in the litter inputs on invertebrate assemblages are still poorly understood.

Case-building experiments

In the laboratory, the cases were carefully dismantled to remove the larvae. No larva

died during this procedure. The individuals were then placed individually in plastic

cups (12 cm diameter, 9 cm high) containing burnt fine gravel (4 h at 400 °C) and 400

ml of filtered spring water. Plastic cups were aerated and kept at 21 °C with a

photoperiod of 12 h light and 12 h dark. Leaf discs of 1.8 cm diameter cut with a cork

borer were offered according to the three experiments described below. All discs were

cut from unconditioned leaves to avoid them being eaten. No evidence of larvae

feeding was observed during the experiments. In each experiment, we observed the

cups over 24 h (different observers), measured the time spent to build a new case, and

counted the number of discs used by each larva for case-building. One leaf disc was

considered as “used” when the whole disc or part of it was incorporated in the new

case. The replicate (plastic cup) was discarded if the individual died after the

beginning of the experiments.

In the first experiment, we distributed 40 larvae in individual cups containing

16 discs from only one of the four plant species (M. guyanensis, M. chartacea, P.

brasiliense or E. camaldulensis) to evaluate the capability of larvae to use the leaf


15

substrates of the four tested plant species (10 replicates per species). In a second

experiment, 40 larvae were randomly assigned to individual cups along with 16 leaf

discs, 4 of each plant species, to evaluate whether the larvae would select discs of

particular species for case-building.

Finally, in a third experiment, we evaluated whether the abundance of leaf

discs could affect their selection. To pursue this question, we prepared two treatments

using the most used and the least used plant species observed in the second

experiment in different abundances. In the first treatment, we offered in each cup 12

discs of the most used plant species and 4 discs of the least used one. In the second

treatment, we offered 4 discs of the most used species and 12 of the least used plant

species. Larvae (12 replicates per treatment) were randomly distributed into each

treatment.

Characterization of plant species leaves

We estimated leaf toughness with a device that measures the force needed to pierce a

leaf disc with a piston of 0.79 mm diameter (see Graça & Zimmer, 2005). To

determine the toughness of each plant species, we cut leaf discs, avoiding leaf veins,

from four leaves using a cork borer of 1.8 cm diameter. Four leaf discs of each plant

species were weighed with a 0.1 mg precision balance to determine the biomass:area

ratio. We also ground some leaf material for the analyses of total phenolics according

to Bärlocher & Graça (2005). Four replicates of each plant species were analyzed.

Statistical analyses

In order to test for significant differences in the plant species use for case-building,

we compared the numbers of discs used in Experiment 1 with a one-way ANOVA

and in Experiment 2 with Friedman’s test, a non-parametric repeated measures

comparison test. This test was used because larvae choice was not independent when

all plant species were simultaneously offered in Experiment 2 (Roa, 1992). Multiple

pairwise comparisons were done by Wilcoxon signed ranks test, with the appropriate

Bonferroni correction. In Experiment 3, we compared the numbers of discs used with

a two-way ANOVA (log transformed data), using plant species and abundance as

factors. ANOVA models used in Experiments 1 and 3 were validated through residual

analyses. We used a Kruskal-Wallis test to look for differences in disc toughness


16

values (g), phenolic concentrations (% g-1 dry mass) and biomass:area ratios (mg cm-

2) among all plant species used in the experiments.

We also calculated Spearman's rank correlation coefficients to evaluate the

relationship between the total number of discs used by Phylloicus larvae in

Experiment 2 with leaf toughness, leaf phenolic concentration, and leaf biomass:area

ratio of each plant species. All statistical analyses were performed using SYSTAT

10.2 (SYSTAT Software Inc., 2002) and based on Zar (1999).

Results

Phylloicus larvae built cases with leaves of all studied species (Fig. 1) and mortality

rates were relatively low (l0.8%, across all experiments). When discs of only one

plant species were offered (Experiment 1), larvae used a mean number of discs

ranging between 5.7 and 7.3, regardless of which plant species was present (F3,32 =

1.661, P = 0.194; Fig. 2A).

Fig. 1 Cases built by Phylloicus larvae with monospecific (left) and multispecies (right) plant leaf discs.


17

Fig. 2 Number of discs used by Phylloicus larvae (mean ± SE) when plant species were offered individually (A) and combined (B). Values sharing superscript letters are not statistically different (P > 0.05).

When exposed to the four plant species together (Experiment 2), larvae were

selective (Friedman’s value = 9.439, df = 3, P = 0.024), using significantly fewer

discs of P. brasiliense (20.0%) than E. camaldulensis (36.4%) and M. guyanensis

(35.7%). Discs of M. chartacea were used in an intermediate proportion (28.6%), not

differing from the others (Fig. 2B).

The third experiment revealed that plant species selection was dependent on

leaf abundance, i.e., even though E. camaldulensis and M. guyanensis were more used

than P. brasiliense, whenever the least preferred species was more abundant in the

system, it was more used, independently if combined with discs of E. camaldulensis

or M. guyanensis (Table 2, Fig. 3).


18

Table 2 Results from a two-way ANOVA of general differences in number of discs used on the 2 plant

species combinations in experiment 3. Data are from Figure 3.

Leaf Combination Factor F df P

E. camaldulensis and Leaf 4.031 1 0.053 P. brasiliense Abundance 27.780 1 <0.001 Leaf X Abundance 0.000 1 1.000

M. guyanensis and Leaf 5.175 1 0.029 P. brasiliense Abundance 19.725 1 <0.001 Leaf X Abundance 0.106 1 0.747

Fig. 3 Number of discs used by Phylloicus larvae when Eucalyptus camaldulensis (A) and Myrcia guyanensis (B) were offered combined with Protium brasiliense in different leaf disc abundances. Values sharing superscript letters are not statistically different (P > 0.05).


19

Plant species differed in toughness (Table 3). M. guyanensis and P. brasiliense

were tougher, whereas E. camaldulensis and M. chartacea were the softest leaves.

Phenolic concentrations also differed among plant species and P. brasiliense showed

the lowest phenolic concentrations (Table 3). Finally, E. camaldulensis had thinner

leaves than M. guyanensis and M. chartacea. Leaf selection by Phylloicus was

correlated with phenolic concentrations (Rs = 0.907, t34 = 6.804, P < 0.001) but not

leaf toughness and thickness (toughness: Rs = - 0.294, t34 = - 1.795, P > 0.05;

biomass:area ratio: Rs = - 0.134, t34 = - 0.788, P > 0.05).

Table 3 Leaf toughness (g), phenolic concentration (% g-1 dry mass) and biomass:area ratio (mg cm-2) of the four plant species tested (mean ± SE; n = 4), Kruskal-Wallis H and probability P. Values sharing superscript letters are not statistically different (P > 0.05).

Toughness Phenolics Biomass:area

Myrcia guyanensis 276.95 ± 16.25a 10.70 ± 0.25a,b 39.32 ± 0.83a Miconia chartacea 100.35 ± 23.83b 9.38 ± 0.10a,b 47.83 ± 1.28a Protium brasiliense 210.85 ± 28.45a 7.55 ± 0.11a 30.01 ± 0.61a,b Eucalyptus camaldulensis 83.62 ± 10.26b 10.94 ± 0.06b 24.48 ± 0.61b H 27.83 10.46 27.44 P <0.001 0.015 <0.001

Discussion

Phylloicus larvae built cases with all plant species offered. After removal from their

original cases and placement in the plastic cups, each larva immediately took shelter

under the leaf discs and started to cut some of them. Larvae spent less than 24 h to

build new cases and in 11 cups new cases were observed only 3 h after the beginning

of the experiments. The behavior exhibited by the larvae and the time needed to build

their cases were similar to those observed by Norwood & Stewart (2002) for the

North American Phylloicus ornatus.

The preference of Phylloicus larvae for E. camaldulensis and M. guyanensis

was positively correlated with leaf phenolic concentrations, suggesting that Phylloicus

larvae preferred to build cases with less palatable leaf pieces (Graça, 2001; Bärlocher

& Graça, 2005). Given that Phylloicus cases present a relatively large surface area

(Wantzen & Wagner, 2006), the use of low quality plant species for case-building

may be a strategy to make their cases less attractive to other invertebrate shredders


20

(Bastian et al., 2007) and to microbial degradation since phenolics are known to retard

microbial colonization (Campbell & Fuchshuber, 1995; Salusso, 2000).

Phylloicus preference was independent of leaf toughness. This may indicate

that among all leaf characteristics that set the preferences for case-building, toughness

might be one of lower importance. Bastian et al. (2007) observed that two species of

caddisfly differed in their choice of leaves for case-building, with Anisocentropus

kirramus using relatively tough leaves and Lectrides varians using much softer

leaves. The effect of toughness should be more systematically tested in future studies

using a larger number of plant species.

In the multiple-choice experiment, Phylloicus larvae used more discs of

species with faster decomposition (E. camaldulensis and M. guyanensis). However,

even having differences in their breakdown rates, we must remark that breakdown

rates of Cerrado plant species are among the slowest observed in tropical

environments (see Moretti et al., 2007) and E. camaldulensis decomposes

approximately 2.5 times faster than M. guyanensis (J. F. Gonçalves, unpublished

data), the native species with the highest breakdown rate. Based on this, we believe

that Phylloicus cases built with E. camaldulensis are not as durable as the ones made

with native species. Then, in environments in which Eucalyptus leaves are available

alone or in larger quantities, larvae would tend to build and/or add more leaf pieces to

their cases, spending more energy than in natural conditions. And this fact could have

significant consequences to adult maturation of insects such as caddisflies, which

present minimal adult feeding (Stevens et al., 1999).

Despite preference for some plant species, Phylloicus case-building behavior

is sufficiently flexible to adapt to the abundance of leaves available in their habitats.

And this pattern was observed independently of the presence of E. camaldulensis, the

non-native species. According to some authors, the type of organic material used by

caddisflies for case-building may vary depending on its abundance (Hanna, 1961;

Otto & Svensson, 1980). Moreover, Moretti & Loyola (2005) demonstrated that

larvae of Barypenthus concolor, a tube-case-building trichopteran, use particle sizes

in the same proportions as they are available in the habitat to build their cases. All

these findings imply great selective pressure operating on this behavior.

Our results corroborate in part those of Rincón & Martínez (2006), who

observed Phylloicus preference for leaves of low palatability and higher lignin

contents for case-building. And this fact is probably related to the intrinsic traits of


21

plant species used in these two studies, which belong to different vegetation types.

For example, the lignin contents of the species used here are on average 1.6 times

higher than the ones from the species used by Rincón & Martínez (2006) in

Venezuela (M. S. Moretti, unpublished data). Due to harsh conditions (e.g., high solar

radiation, water stress, and herbivory), Brazilian Cerrado plant species have leaves

with thick cuticles and high contents of structural and inhibitory compounds (Marques

et al., 2000; Oliveira & Marquis, 2002; Wantzen & Wagner, 2006). On the other

hand, plant species used by Rincón & Martínez (2006) are native from a dry tropical

semideciduous forest located in Northwestern Venezuela (Rincón et al., 2005), a

region where annual precipitation is higher and plant species are subject to fewer

environmental stresses than in Cerrado. However, the results of both studies suggest

that Phylloicus larvae build cases with less palatable leaves when exposed to a group

of tough leaves (present study) or with the toughest leaves among the ones that were

available in the system (Rincón & Martinez, 2006).

Since E. camaldulensis is not present in the site where larvae were collected,

preferences here observed were measured in a non-natural environment where

animals responded to leaf traits they have not faced before, suggesting that our results

could be different if we had used only native species. In spite of this, we decided to

include E. camaldulensis in our experiments because this is one of the biggest threats

to the Cerrado biome in southeastern Brazil (Klink & Machado, 2005) and leaves of

this species will probably reach several streams in the near future.

According to our results, Phylloicus larvae select for case-building those

leaves that are chemically protected against microbial degradation and shredder

consumption, even if they are from a non-native species like E. camaldulensis.

Moreover, we found that this preference is also influenced by leaf abundance,

indicating that larvae are dependent on species composition in the riparian zone.

Therefore, changes in riparian vegetation providing leaf substrates (i.e., change of

species and/or substrate homogenization caused by reforestation with a single species)

can be important to Phylloicus larvae due to the impact on growth and reproduction.

In this way, our findings also reinforce the importance of riparian resources and their

diversity to the maintenance of aquatic consumers in tropical shaded streams.

To summarize, our results indicate that Phylloicus larvae present a high

plasticity on case-building behavior. Furthermore, it seems that their selection

depends on each situation, i.e., on which plant species are present and their relative


22

abundance. A combination of variables such as leaf chemistry, availability and

toughness will determine larvae preferences. Other factors, such as the breakdown

rates of different leaf types and the presence of predators may influence case

construction also, and should be evaluated.

Acknowledgments We thank COPASA-MG and IEF-MG for logistical facilities and

research licenses. This work was supported by FAPEMIG, CNPq, CAPES

Foundation, and US Fish & Wildlife Service. We also thank Manuel Graça, Robert

Hughes and 3 anonymous referees for valuable comments on the previous version of

the manuscript, Juliana França and Victor Gomes for assistance in the laboratory.

References

APHA, 1992. Standard methods for the examination of water and wastewater (18th

ed.). American Public Health Association, Washington DC.

Bärlocher, F. & M. A. S. Graça, 2005. Total phenolics. In Graça, M. A. S., F.

Bärlocher & M. O. Gessner (eds), Methods to study litter decomposition: a

practical guide. Springer, Dordrecht: 45-48.

Bastian, M., L. Boyero, B. R. Jackes & R. G. Pearson, 2007. Leaf litter diversity and

shredder preferences in an Australian tropical rain-forest stream. Journal of

Tropical Ecology 23: 219-229.

Boyero, L., P. A. Rincón & J. Bosch, 2006. Case selection by a limnephilid caddisfly

[Potamophylax latipennis (Curtis)] in response to different predators. Behavioral

Ecology and Sociobiology 59: 364-372.

Campbell, I. C. & L. Fuchshuber, 1995. Polyphenols, condensed tannins, and

processing rates of tropical and temperate leaves in an Australian stream. Journal

of the North American Benthological Society 14: 174-182.

Cummins, K. W., M. A. Wilzbach, D. M. Gates, J. B. Perry & W. B. Talaiferro, 1989.

Shredders and riparian vegetation. BioScience 39: 24-30.

de Moor, F. C. & V. D. Ivanov, 2008. Global diversity of caddisflies (Trichoptera:

Insecta) in freshwater. Hydrobiologia 595: 393-407.

Flint, O. S., R. W. Holzenthal & S. C. Harris, 1999. Catalog of the Neotropical

caddisflies (Insecta: Trichoptera). Special Publication, Ohio Biological Survey,

Columbus.


23

Graça, M. A. S., 2001. The role of invertebrates on leaf litter decomposition in

streams – a review. International Review of Hydrobiology 86: 383-393.

Graça, M. A. S. & M. Zimmer, 2005. Leaf toughness. In Graça, M. A. S., F.

Bärlocher & M. O. Gessner (eds), Methods to study litter decomposition: a

practical guide. Springer, Dordrecht: 109-113.

Hanna, H. M., 1961. Selection of materials for case-building larvae of caddis flies

(Trichoptera). Proceedings of the Royal Society of London B 36: 37-47.

Huamantinco, A. A., L. L. Dumas & J. L. Nessimian, 2005. Description of larva and

pupa of Phylloicus abdominalis Ulmer, 1905 (Trichoptera: Calamoceratidae).

Zootaxa 1039: 19-26.

Klink, C. A. & R. B. Machado, 2005. Conservation of the Brazilian Cerrado.

Conservation Biology 19: 707-713.

Marques, A. R., Q. S. Garcia, J. L. P. Resende & G. W. Fernandes, 2000. Variations

in leaf characteristics of two species of Miconia in the Brazilian cerrado under

different light intensities. Tropical Ecology 41: 47-60.

Moretti, M. S. & R. D. Loyola, 2005. Does Barypenthus concolor Burmeister

(Trichoptera: Odontoceridae) select particles of different sizes for case building?

Neotropical Entomology 34: 337-340.

Moretti, M. S., J. F. Gonçalves, R. Ligeiro & M. Callisto, 2007. Invertebrates

colonization on native tree leaves in a neotropical stream (Brazil). International

Review of Hydrobiology 92: 199-210.

Norwood, J. C. & K. W. Stewart, 2002. Life history and case-building behavior of

Phylloicus ornatus (Trichoptera: Calamoceratidae) in two spring-fed streams in

Texas. Annals of the Entomological Society of America 95: 44-56.

Oliveira, P. S. & R. J. Marquis, 2002. The cerrados of Brazil: Ecology and natural

history of netropical Savanna. Columbia University Press, New York.

Otto, C., 2000. Cost and benefit from shield cases in caddis larvae. Hydrobiologia

436: 35-40.

Otto, C. & B. S. Svensson, 1980. The significance of case material selection for the

survival of caddis larvae. Journal of Animal Ecology 49: 855-865.

Prather, A. L., 2003. Revision of the Neotropical caddisfly genus Phylloicus

(Trichoptera: Calamoceratidae). Zootaxa 275: 1-214.

Resh, V. H. & D. M. Rosenberg (eds), 1984. The ecology of aquatic insects. Praeger,

New York.


24

Rincón, J. & I. Martínez, 2006. Food quality and feeding preferences of Phylloicus sp.

(Trichoptera: Calamoceratidae). Journal of the North American Benthological

Society 25: 209-215.

Rincón, J., I. Martínez, E. León & N. Ávila, 2005. Procesamiento de la hojarasca de

Anacardium excelsum en una corriente intermitente tropical del noroeste de

Venezuela. Interciencia 30: 228-234.

Roa, R., 1992. Design and analysis of multiple-choice feeding preference

experiments. Oecologia 89: 509-515.

Salusso, M. M., 2000. Biodegradation of subtropical forest woods from north-west

Argentina by Pleurotus laciniatocrenatus. New Zealand Journal of Botany 38:

721-724.

Stevens, D. J., M. H. Hansell, J. A. Freel & P. Monaghan, 1999. Developmental

trade-offs in caddis flies: increased investment in larval defense alters adult

resource allocation. Proceedings of the Royal Society of London B 266: 1049-

1054.

Wallace, J. B. & J. R. Webster, 1996. The role of macroinvertebrates in stream

ecosystem function. Annual Review of Entomology 41: 115-139.

Wantzen, K. M. & R. Wagner, 2006. Detritus processing by invertebrate shredders: a

neotropical–temperate comparison. Journal of the North American Benthological

Society 25: 216-232.

Wiggins, G. B., 1996. Larvae of North American caddisfly genera (Trichoptera), 2nd

edn. University of Toronto Press, Ontario.

Wiggins, G. B., 2004. Caddisflies, the underwater architects. University of Toronto

Press, Toronto, Buffalo, London.

Zar, J. H., 1999. Biostatistical analysis. Prentice Hall, New Jersey.

2LengthdrymassrelationshipsforatypicalshredderinBrazilianstreams

(Trichoptera,Calamoceratidae)*

BárbaraBecker,MarceloS.Moretti&MarcosCallisto

Lab.EcologiadeBentos,InstitutodeCiênciasBiológicas,UniversidadeFederaldeMinasGerais,Av.AntônioCarlos6627,

CP486,30161970,BeloHorizonte,MG,Brasil

Preparation ofPhylloicus sp. larvae(Trichoptera, Calamoceratidae) forthe measurement of body length,headcapsulewidth,andinteroculardistance.

*PublishedinAquaticInsects(2009),31,227‐234

Chapter2:Length‐drymassrelationshipsforPhylloicus

26

Length-dry mass relationships for a typical shredder

in Brazilian streams (Trichoptera, Calamoceratidae)

Bárbara Becker, Marcelo S. Moretti & Marcos Callisto

Abstract The aims of this study were to determine which linear body dimensions are

best suitable and which mathematical functions can be used to describe length dry-

mass relationships for a population of Phylloicus sp. (Trichoptera, Calamoceratidae)

larvae. We measured 3 linear body dimensions (body length, head capsule width and

interocular distance) of 54 larvae to use as dry mass predictors. For the description of

length-dry mass relationships we used linear, exponential and power function models.

Body length provided the best fitted equations to estimate biomass, followed by head

capsule width and interocular distance. The highest coefficients of determination were

found in power function and exponential models. These relationships can be useful to

determine the growth rate and/or secondary production of Phylloicus larvae in future

laboratory experiments, as well as to understand the importance of these shredders in

the energy flux of shaded tropical streams.

Keywords Size-mass equations, biomass estimation, linear body dimensions,

Phylloicus, tropical shredders.


27

Introduction

Biomass of aquatic macroinvertebrates is important to determine growth rates and/or

secondary production, as well as to understand life histories, seasonal patterns and

trophic relationships between functional feeding groups (Benke, 1996; Burgherr and

Meyer, 1997). Data on macroinvertebrate biomass can also be useful in colonization

studies or quantifying the role of detritivores on leaf decomposition (Cressa, 1999).

Among the different approaches to biomass determination, the most common

is the direct weighing of individual specimens (Dermott and Paterson, 1974; Smock,

1980; Meyer, 1989). However, this approach is often very time consuming, and prone

to error if the insects have been previously stored in chemical preservatives (e.g.

formaline or alcohol), which can cause alterations in their dry mass (Donald and

Paterson, 1977; Downing and Rigler, 1984; Kato and Miyasaka, 2007). Direct

determination of dry mass has the added disadvantage of rendering the specimen

useless for further examination as a result of the drying process (Towers et al., 1994).

An alternative to avoid such disadvantages is to estimate the biomass

indirectly, using length-dry mass conversions (Gould, 1966; Peters, 1983; Burgherr

and Meyer, 1997, Benke et al., 1999). Estimating dry-mass indirectly from linear

body dimensions (e.g. body length, head capsule width) is more rapid than direct mass

determination, particularly for small invertebrates. Moreover, in laboratory

experiments assessing invertebrates feeding behavior, this approach allows the

estimation of initial biomass without stressing and/or killing the organisms.

Length-dry mass relationships have been used to estimate the biomass of

invertebrates from different geographical locations and of taxa with similar body

shapes (Johnston and Cunjak, 1999). Most of the length dry-mass relationships for

stream invertebrates were estimated for North American and European taxa (Smock,

1980; Meyer, 1989) and, until now, only a few data were proposed for the Tropical

region. Furthermore, previous studies suggested the need to use taxa-specific

relationships because they are more precise, once different taxa may differ in body

shape and volume (Schoener, 1980; Smock, 1980; Gowing and Recher, 1985; Cressa,

1986).

Only few invertebrate taxa have been mentioned as shredders in neotropical

streams. Among them, larvae of the genus Phylloicus Müller, 1880 (Trichoptera,

Calamoceratidae) are well distributed throughout Latin America and, in some streams,


28

can be found easily on leaf patches with low water current (Prather, 2003). Because

these larvae are also easy to manipulate and keep alive in laboratory conditions, they

have been used in many experiments (e.g. feeding preference, growth, survival and

case building) that aimed to better understand the behavior of shredders and their

influence on leaf decomposition in tropical streams (Graça et al., 2001, Rincón and

Martínez, 2006).

In this study, we analyzed the length-dry mass relationships for a population of

Phylloicus sp. by using three different regression functions (linear, power and

exponential) and three body dimensions in order to determine the best relationship.

Material and Methods

Phylloicus sp. larvae were collected on July 2007 in Taboões spring (20° 03' 38" S -

44° 03' 03" W), located in the Serra do Rola Moça State Park, Minas Gerais State,

southeastern Brazil. The Taboões spring is inside a forest fragment, presenting a well

developed riparian area, which forms a closed canopy. Leaves fall throughout the year

and accumulate in the streambed.

Larvae were found visually, collected with a hand net, and taken to the

laboratory in an isothermic box with stream water. In the laboratory, undamaged

individuals of the same morphospecies were carefully removed from their cases and

placed individually in Petri dishes. Three linear body dimensions were chosen among

the most common used as biomass predictors: body length, head capsule width, and

interocular distance (Meyer, 1989). Body length (BL) was measured as the distance

from the anterior of the head to the posterior of the last abdominal segment. Head

capsule width (HW) was measured across the widest section of the head. Interocular

distance (ID) was measured as the minimum distance between eyes, paralel to head

width. Body dimensions were measured to the nearest 0.1 mm with a Zeiss dissecting

microscope fitted with an ocular micrometer (magnification: 8x for BL measurements,

and 50x for HW and ID measurements). Animals were then placed individually in

pre-weighed aluminum foils, dried at 60°C for 48 h (Meyer, 1989), left to cool in a

desiccator, and their dry-mass (DM) was measured to the nearest 0.1 mg.

Three regression models were calculated for the three Phylloicus body

dimensions, using the method of least squares. The fit of regression equations was

judged by the determination coefficient (r2), the significance level (p, obtained from


29

regression ANOVA) and residual analysis. All statistical analyses were performed

based on Zar (1999).

Results

Body dimensions measures and dry weights of 54 larvae were used for statistical

analyses. Phylloicus dry mass presented the highest variation coefficient, with values

ranging from 1.3 to 26.6 mg (Table 1). Among body dimensions, body length

presented higher range (10.4 - 28.9 mm) and variation coefficient (Table 1).

Table 1 Ranges, mean, standard deviation (SD) and coefficient of variation (CV, in percentage) for body length, head capsule width, interocular distance (mm) and dry mass (mg) of Phylloicus sp. larvae; n = 54. CV = (SD/Mean) x 100.

Range Mean SD CV Body length 10.4 – 28.9 16.7 2.7 16.2 Head capsule width 0.8 – 1.5 1.3 0.2 14.0 Interocular distance 0.6 – 1.1 1.0 0.1 14.3 Dry mass 1.3 – 26.6 12.1 6.7 54.2

The following regression models were chosen because they provided the best

fits. Conversion of Phylloicus body dimensions to dry mass was determined by linear

(1), exponential (2) and power function (3) models or its logarithmic equivalents:

DM = a + b . L (1)

DM = a . ebL (in linear form: ln DM = ln a + b . L) (2)

DM = a . Lb (in linear form: ln DM = ln a + b . ln L) (3)

where a/b are regression constants, DM is dry mass, L is the linear body dimension

(BL, HW, ID) and e is a mathematical constant (Euler’s number: 2.718).

The parameters of equations (1), (2) and (3) are listed in Table 2. All body

dimensions showed a very high level of significance in the three models (p < 0.01).

Body length provided the best relationships to estimate biomass (Table 2), followed

by interocular distance and head capsule width. These relationships were best fitted by

power function and exponential models that presented very similar coefficients of


30

determination to each body dimension. Figure 1 shows the relations of dry mass as a

function of body length, head capsule width and interocular distance for Phylloicus

larvae. The regression lines and curves were given by power function.

Table 2 Parameters (with the 95% confidence intervals) of the linear, exponential and power function models for the relationship between a linear body dimension (L = body length [BL], head capsule width [HW] or interocular distance [ID], in mm) and dry mass (DM, in mg) of Phylloicus sp. larvae; a, b = regression constants, r2 = coefficient of determination (* p < 0.01, ** p < 0.001). n = 54.

Fig. 1 Scatter diagrams of dry mass versus body length (A), head capsule width (B) and interocular distance (C) on normal coordinates (□) as well as on logarithmic coordinates (●) for Phylloicus sp. larvae. The regression equations (power function) are DM = a . Lb and ln DM = ln a + b . ln L.

A B

C


31

Discussion

Even though all relationships between body dimensions and biomass were highly

significant, body length was the best predictor explaining 75-76% of the variation in

mass. This linear body dimension is widely used for determining length-dry mass

relationships of aquatic invertebrates (e.g., Smock, 1980; Towers et al., 1994;

Burgherr and Meyer, 1997) mainly because it has a broader measuring range. Body

length also provides slightly higher determination coefficients than head capsule

width and interocular distance (González et al., 2002).

Although body length usually gives the best relationships, some authors (see

Cressa, 1999; Marchant and Hehir, 1999; González et al., 2002) prefer to use other

linear body dimensions, like head capsule width, case width, pronotum length or

tarsus length. This probably owes to the fact that, among other reasons, these

structures are sclerotized and less subject to distortion or breakage under manipulation

than body length. In addition, Becker (2005) found that pronotum length is the best

measurement to distinct larval instars of Agapetus fuscipes (Trichoptera) in a German

first-order stream. In the present study, larvae were measured in the same day they

have been sampled. So, all measurements were done on fresh, undamaged and

completely stretched animals, which allowed a precise and reliable determination of

the three studied body dimensions.

The exponential and power function models did not differ between the body

dimensions determined. Most authors found the highest fit between body length and

dry-mass when they use the power function model (e.g., Smock, 1980; Meyer, 1989;

Burgherr and Meyer, 1997) but exponential regressions have been also used by

Dudgeon (1995) and Perán et al. (1999) for length-dry mass relationships of

Hydrocyphon (Coleoptera) and Caenis luctuosa (Ephemeroptera), respectively.

Wenzel et al. (1990) pointed out that differences between the results obtained using

different regression models are low and they decrease when higher number of animals

is used. Although power function is more used, the exponential model should not be

discarded when looking for the best fit of length-dry mass relationships.

In practice, when interpreting a length-dry mass regression equation, “b”

values represent the rate of increase (i.e., slope) of dry weight against length in a

linear relationship, whereas the constant “a” only represents the dry mass of an

organism at a unit length (i.e., 1 mm). It’s known that for tropical aquatic insects the


32

constant b falls short of the expected value of 3, which means that body mass of

insects is more influenced by surface than by volume (Engelmann, 1961). Our results

support those from Cressa (1999) who found that Phylloicus sp. is one of the few taxa

of tropical invertebrates whose slope is higher than 3, so it is possible that in this

genus volume could influence body mass more than surface.

Some variations in length-dry mass relationships for the populations of the

same species, but from different locations, can be caused by physical-chemical

differences of the environment, trophic conditions or genetics. In this way, it is

recommended to determine the relationships for populations under study or use

relationships that were determined for populations from the same streams and/or

regions. For example, Rincón & Martinez (2006), studying the growth rates of

Phylloicus in laboratorial experiments, used the empirical relationship described by

Cressa (1999) who had studied populations from a similar region of Venezuela. On

the other hand, length-dry mass relationships are not much affected by seasons, as

shown by Kato and Miyasaka (2007). These authors suggested that it is not necessary

to measure larvae in dry and wet seasons to have a consistent relationship.

When sampling organisms to determine length-dry mass relationships, one

must be sure that organisms from different sizes (cohorts) have been collected. If not,

only part of the logistic curve of population growth is quantified and the resulting

relationships may not represent the whole population (Begon et al., 1996). In this

study, if we consider the Dyar’s law, an empirical law that suggests an increase of 1.5

in growth at each instar (Wigglesworth, 1972), and the ranges of each body dimension

measured, we can infer that only larvae from the last two instars were sampled. Based

on this, our equations were determined with data from the right side of the curvilinear

relationships between dry mass and body dimensions of this population of Phylloicus

(see Majecki et al., 1997). On the other hand, as we have been monitoring this

population for several months, larvae used in this study presented the same range of

size of the ones that are found visually in most part of the year, suggesting that our

equations were adequate to determine the dry mass of larvae destined to laboratory

experiments.

In conclusion, the length-dry mass relationships here presented can be useful

to determine the growth rate and/or secondary production of Phylloicus. Besides, our

results also reinforce the necessity of more studies focusing on the life cycles of

aquatic insects in the tropical region. We do hope that the present study encourage


33

future research assessing the population dynamics of tropical shredders, as well as to

understand the importance of these individuals on leaf processing, trophic

relationships, colonization rates, and even to compare populations within and between

habitats.

Acknowledgments This study was supported by FAPEMIG, CNPq, CAPES, Eawag,

US Fish and Wildlife Service. We appreciated the help of our laboratory colleagues

Lurdemar Tavares and Juliana França during field and laboratory activities. We are

also thankful to João José Leal, Leandro Oliveira, Vicenç Acuña and two anonymous

reviewers who provided useful comments on the manuscript.

References

Becker, G. (2005), “Life cycle of Agapetus fuscipes (Trichoptera, Glossosomatidae)

in a first-order upland stream in central Germany”, Limnologica, 35, 52-60.

Begon, M., Mortimer, M., and Thompson, D.J. (1996), Population ecology: a unified

study of animals and plants. Blackwell Science, Oxford. 3rd edition.

Benke, A. (1996), “Secondary production of macroinvertebrates”, in Methods in

stream ecology, ed. F.R. Hauer and G.A. Lamberti, New York: Academic, 557-

578.

Benke, A., Huryn, A., Smock, L., and Wallace, J. (1999), “Length-mass relationships

for freshwater macroinvertebrates in North America with particular reference to

the southeastern United States”, Journal of the North American Benthological

Society, 18, 308-343.

Burgherr, P., and Meyer, E.I. (1997), “Regression analysis of linear body dimensions

vs. dry mass in stream macroinvertebrates”, Archiv für Hydrobiologie, 139,

101-112.

Cressa, C. (1986), “Estimaciones de peso seco en función de la longitud cefálica y

clases de tamaño en Campsurus sp. (Ephemeroptera, Polymitarcidae)”, Acta

Científica Venezolana, 37, 170-173.

Cressa, C. (1999), “Dry mass estimates of some tropical aquatic insects”, Revista de

Biología Tropical, 47, 133-141.


34

Dermott, R.M., Paterson, C.G. (1974), “Determining dry weight and percentage dry

matter of chironomid larvae”. Canadian Journal of Zoology 52, 1243-1250.

Donald, G.L., and Paterson, C.G. (1977), “Effects of preservation on wet weight

biomass of chironomid larvae”, Hydrobiologia, 53, 75-80.

Downing, J.A., and Rigler, F.H. (1984), A manual on methods for the assessment of

secondary productivity in fresh waters. Blackwell Scientific Publications.

Oxford.

Dudgeon, D. (1995), “Life history, secondary production and microdistribution of

Hydrocyphon (Coleoptera: Scirtidae) in a tropical forest stream”, Archiv für

Hydrobiologie, 133, 261-271.

Engelmann, M.D. (1961), “The role of soil arthropods in the energetics of an old field

community”, Ecological Monographs, 31, 221-238.

González, J.M., Basaguren, A., and Pozo, J. (2002), “Size-mass relationships of stream

invertebrates in a northern Spain stream”, Hydrobiologia, 489, 131-137.

Gould, S. (1966), “Allometry and size in ontogeny and phylogeny”, Biological

Research, 41, 587-640.

Gowing, G., and Recher, H.F. (1985), “Length-weight relationships for invertebrates

from forest in south-eastern New South Wales”, Australian Journal of Ecology,

9, 5-8.

Graça, M.A.S., Cressa, C., Gessner, M.O., Feio, M.J., Callies, K.A., and Barrios, C.

(2001), “Food quality, feeding preferences, survival and growth of shredders

from temperate and tropical streams”, Freshwater Biology, 46, 947-957.

Johnston, T., and Cunjak, R., (1999), “Dry mass-length relationships for benthic

insects: a review with new data from Catamaran Brook, New Brunswick,

Canada”, Freshwater Biology, 41, 653-674.

Kato, M.G., and Miyasaka, H. (2007), “Length–weight relationships of four predatory

stonefly species in Japan”, Limnology, 8, 171-174.

Majecki, J., Grzybkowska, M., and Reddy, R. (1997), “Density, production and life

cycle of Brachycentrus subnubilus Curtis (Trichoptera: Brachycentridae) in a

lowland river, Central Poland”, Hydrobiologia, 354, 51-56.


35

Marchant, R., and Hehir, G. (1999), “Growth, production and mortality of two species

of Agapetus (Trichoptera: Glossosomatide) in the Acheron River, south-east

Australia”, Freshwater Biology, 42, 655-671.

Meyer, E. (1989), “The relationship between body length parameters and dry mass in

running water invertebrates”, Archiv fiir Hydrobiologie, 117, 191-203.

Perán, A., Velasco, J., and Millán, A. (1999), “Life cycle and secondary production of

Caenis luctuosa (Ephemeroptera) in a semiarid stream (Southeast Spain)”,

Hydrobiologia, 400, 187-194.

Peters, R.H. (1983), The ecological implications of body size, Cambridge University

Press, Cambridge.

Prather, A.L. (2003), “Revision of the Neotropical caddisfly genus Phylloicus

(Trichoptera: Calamoceratidae)”, Zootaxa, 275, 1-214.

Rincón, J., and Martínez, I. (2006), “Food quality and feeding preferences of

Phylloicus sp. (Trichoptera: Calamoceratidae)”, Journal of the North American

Benthological Society, 25, 209-215.

Schoener, T.W. (1980), “Length-weight regressions in tropical and temperate forest-

understorey insects”, Annals of the Entomological Society of America, 73, 106-

109.

Smock, L.A. (1980), “Relationships between body size and biomass of aquatic

insects”, Freshwater Biology, 10, 375-383.

Towers, D.J., Henderson, I.M., Veltman, C.J. (1994), “Predicting dry weight of New

Zealand aquatic macroinvertebrates from linear dimensions”, New Zealand

Journal of Marine and Freshwater Research, 28, 159-166.

Wenzel, F., Meyer, E., Schwoerbel, J. (1990), “Morphometry and biomass

determination of dominant mayfly larvae (Ephemeroptera) in running waters”,

Archiv für Hydrobiologie,118, 31-46.

Wigglesworth, V. B. (1972), The Principles of Insect Physiology. Butler and Tanner,

United Kingdom.

Zar, J. H. (1999), Biostatistical Analysis. Prentice Hall, New Jersey. 4th edition.

3AreexoticEucalyptusleavesabetter

foodresourcetoPhylloicussp.(Trichoptera,Calamoceratidae)in

BrazilianCerradostreams?

MarceloS.Moretti,BárbaraBecker&MarcosCallisto

Lab.EcologiadeBentos,InstitutodeCiênciasBiológicas,UniversidadeFederaldeMinasGerais,Av.AntônioCarlos6627,


LarvaofPhylloicussp.(Trichoptera,Calamoceratidae), one of the fewinvertebrate taxa considered asshredderinNeotropicalstreams.

Chapter3:EffectsofEucalyptusonPhylloicusfeeding

37

Are exotic Eucalyptus leaves a better food resource to Phylloicus sp.

(Trichoptera, Calamoceratidae) in Brazilian Cerrado streams?

Marcelo S. Moretti, Bárbara Becker & Marcos Callisto

SUMMARY

1. Through laboratory and field settings, we investigated food preferences, growth

and survival of Phylloicus sp. (Trichoptera, Calamoceratidae) larvae feeding on

leaves of the exotic Eucalyptus camaldulensis and three Cerrado native species

(Myrcia guyanensis, Miconia chartacea, and Protium brasiliense).

2. The N : P ratio for Miconia was the highest (83 : 1), followed by Myrcia (37 : 1),

Protium (26 : 1) and Eucalyptus (22 : 1). Eucalyptus, Myrcia, and Miconia had

similar amount of tannins (6.00 - 8.16%). The lowest amounts of lignin and cellulose

were presented by Eucalyptus (12.87 and 15.93%).

3. With the exception of Protium, all leaf types were consumed in the food preference

experiments. In laboratory, Eucalyptus was more consumed than Myrcia (0.436 and

0.209 mg mg-1 day-1), while Miconia was consumed on intermediate rates (0.230 mg

mg-1 day-1). In the field, no significant differences were found among leaf types.

4. After 21 days, Phylloicus daily growth rates ranged from 0.08 to 0.14 mg day-1 and

differences were not found among food treatments. Survival of larvae fed on

Eucalyptus (68%) and Miconia (64%) was significantly higher than those fed on

Myrcia (50%) and Protium (32%).

5. Our results suggest that the low quality of food resources available in Cerrado

streams might impose a constraint to shredders consumption that probably reflects

their lower abundance and the slow breakdown rates of native species in these

ecosystems.

Keywords Exotic leaf litter, food preferences, shredders behavior, leaf chemistry,

tropical streams


38

Introduction

Allochthonous inputs of leaf litter are the main energy source to aquatic communities

living in shaded headwater streams (Webster & Benfield, 1986; Cummins et al., 1989,

Abelho, 2001). After entering streams, leaves are retained on the streambed and used

as food resource by many invertebrates (Vannote et al., 1980; Wallace & Webster,

1996; Mathuriau et al., 2008). Through their feeding activity, invertebrate shredders

convert part of leaf litter and other coarse particulate organic matter (CPOM) into fine

particulate organic matter (FPOM), making it available to other consumers further

downstream (Cummins & Klug, 1979; Wallace & Webster, 1996).

Previous studies concerning the feeding behavior of shredders revealed that

they prefer to feed on leaves that promote higher growth (Arsuffi & Suberkropp,

1986) and survival (Canhoto & Graça, 1995), rejecting the ones that are tough, poor

in nutrients and present high contents of secondary compounds (Graça, 2001; Rincón

& Martínez, 2006; Li & Dudgeon, 2008). It is also known that invertebrate shredders

prefer conditioned leaves, i.e., leaves colonized by aquatic microorganisms, over

unconditioned (Graça et al., 2001). Moreover, Bastian et al. (2008), manipulating the

diversity of shredders and leaf litter in a detritus food web in Australia, suggested that

lowered leaf diversity promotes competitive interactions among shredders.

Riparian zones present a high susceptibility to invasion by exotic species

(Naiman et al., 2005). The invasion of exotic plant species in the riparian vegetation

can alter stream metabolism and invertebrate assemblages by changing the nutritional

resource base into streams (Vitousek, 1990; Going & Dudley, 2008; Moline & Poff,

2008) and habitat structure because exotic species may present different timing of leaf

abscission and decomposition rates (Wallace et al., 1999; Graça et al., 2002;

Thompson & Townsend, 2003). Canhoto & Graça (1995) observed lower

consumption and no growth of the stream detritivore Tipula lateralis (Diptera:

Tipulidae) when feeding on the exotic Eucalyptus globulus in central Portugal.

Reinhart & VandeVoort (2006) suggested that the replacement of the native riparian

species Populus trichocarpa by the exotic Acer platanoides might affect the most

common families of detritivore invertebrates in lotic ecosystems from the United

States.

In Brazil, Eucalyptus trees were first introduced in 1868 (Miritz et al., 2008).

Nowadays, Eucalyptus plantations occupy an area of 6 million of hectares and are one


39

of the greatest threats to Cerrado Biome, a biodiversity hotspot in South America

(Myers et al., 2000). Many native areas, including riparian zones, are being deforested

and replaced with Eucalyptus monocultures (Klink & Machado, 2005; Maquere et al.,

2008) to provide wood for a variety of purposes (e.g., charcoal, paper industry,

furniture). However, the consequences of this replacement on terrestrial and aquatic

ecosystems are still poorly understood.

In the present study, we aimed to evaluate the effects of Eucalyptus leaves on

the feeding behavior of a typical shredding caddisfly from Brazilian Cerrado

headwater streams. Through laboratory and field settings, we investigated food

preferences, growth and survival of Phylloicus sp. larvae feeding on leaves of the

exotic Eucalyptus camaldulensis Dehn. and three Cerrado native species (Myrcia

guyanensis Aubl., Miconia chartacea Triana, and Protium brasiliense Engl.). Due to

the low quality of Brazilian Cerrado leaf litter, we hypothesized that Eucalyptus

leaves would be preferred and promote higher growth and survival of Phylloicus

larvae than native species.

Methods

Shredders

The caddisfly used in our experiments, Phylloicus sp. (Trichoptera, Calamoceratidae),

is one of the few invertebrate taxa considered as shredder in Neotropical streams.

Individuals of this genus can be found throughout Americas (Prather, 2003) and have

been used in laboratory experiments evaluating the behavior of tropical shredders (see

Graça et al., 2001; Rincón & Martínez, 2006; Becker et al., 2009; Moretti et al.,

2009). Larvae were collected by hand on several occasions between August and

December 2006 in Taboões (20° 03' 38" S - 44° 03' 03" W), a well-preserved spring

located inside the Serra do Rola Moça State Park, Minas Gerais State, southeastern

Brazil. This spring is surrounded by a native forest (ca. 92 hectares), presenting

diverse woody riparian vegetation that is mainly composed by Cerrado species. Some

tree species characteristic from the transition zone between Cerrado and Atlantic

Forest are also present. Fallen leaves form patches along spring margins all year,

which support a large population of one undetermined species of Phylloicus. More

information and the water properties of Taboões spring can be found in Moretti et al.

(2009).


40

After collection, larvae were placed in an isothermic box with spring water

and brought to the laboratory, where they starved for 2 days in an aquarium (80 cm

long, 20 cm wide, 40 cm high) with spring water and a bottom of burnt fine gravel (4h

at 500º C). The aquarium was aerated continuously and maintained under 21 ºC and

12 L : 12 D photoperiod.

Leaves

In all experiments, we used senescent leaves of the four studied species. Leaves of

Myrcia, Miconia, and Protium were chosen because they are abundant in the riparian

zones of Cerrado streams in Minas Gerais State and decompose differently (Moretti et

al., 2007). All leaves were collected at once from plastic nets (1 m2, 10 mm mesh size,

approximately 1.5 m height) implanted in the riparian zone (native species) and in a

monoculture of Eucalyptus located nearby. Leaves were air dried and stored at room

temperature until needed. Before use in the experiments, leaves were conditioned by

incubating in fine mesh bags (0.5 mm mesh size) for 2 weeks in the studied spring.

Initial total nitrogen content of leaves was determined using the Kjeldahl method, and

total phosphorus was measured by the ascorbic acid method (Flindt & Lillebo, 2005).

Nitrogen to phosphorus (N : P) ratios were calculated from dry mass by dividing the

nitrogen fraction (%N) by the phosphorus fraction (%P). Tannins were determined as

tannic acid equivalents by the radial diffusion assay (Graça & Bärlocher, 2005) and

proximate lignin and cellulose contents were determined gravimetrically according to

Gessner (2005). All analyses were done with 4 replicates and differences in leaf

chemical properties were tested using one-way ANOVA and Tukey’s HSD test, after

arcsine square root transformation (Zar, 1999).

Food preference experiments

We assessed the food preference of Phylloicus larvae using the multiple-choice

approach described by Canhoto et al. (2005). We also tested whether larvae had

consumed similar amounts of each leaf type under laboratory and field conditions. In

laboratory, feeding arenas consisted of plastic cups (12 cm diameter, 9 cm high)

containing burnt fine gravel and 400 ml of filtered spring water. All cups were

maintained under temperature and photoperiod conditions described above and

aerated through pipette tips connected to an aquarium pump. In Taboões, feeding

arenas consisted of PVC pipes (5 cm diameter, 20 cm long) closed with a fine mesh


41

(0.5 mm mesh size) on both sides. Pipes were fixed on the spring bottom with a nylon

rope and iron bars.

Phylloicus larvae of similar size (early instars) were visually selected from the

pool of organisms and placed individually in each arena. Food was offered in the form

of leaf discs (1.4 cm diameter). Paired discs were cut from contiguous areas of leaves

with a cork borer, avoiding the major leaf veins. For each arena (replicate), we cut

one pair of each one of the four leaf types. One disc of each pair was used as control

and the other was offered to the shredder. The control discs were placed inside a small

0.5 mm mesh bag and suspended from the rim of the cups or tied close to the pipe

internal surface, such that they were fully submerged but inaccessible to the larvae.

Leaf discs exposed to the larvae were marked with colored pins (one color for each

leaf type). The number of replicates was 80 in laboratory and 80 in the field.

Larvae were allowed to feed until one of the leaf discs was reduced to about

two-thirds of its initial size. This was achieved within 1-4 days, both in laboratory and

field conditions. The leaf material remaining after the feeding period (control and

exposed discs) was dried at 60º C for 72 hours and weighed to the nearest 0.01 mg.

The animals were also dried and weighed. Individual consumption of each leaf type

was expressed in terms of mg of leaf dry mass ingested (difference between the

weight of the control and exposed discs) per individual biomass (mg) over the feeding

period (days).

The first procedure to analyze food preference data was to determine whether

animals had consumed significant amount of each leaf type. For this, we used paired

t-tests to compare the mean weight of exposed leaf discs and the corresponding

control discs (Friberg & Jacobsen, 1994). If differences were not found, larvae had

not consumed significantly that leaf type and it was then removed from further

analyses.

Consumption rates were evaluated by permutation tests according to Bärlocher

(2005). Permutation tests were used because consumers choice is not independent in

multiple-choice experiments (Roa, 1992). These tests are more flexible than standard

tests and do not require normal distribution of data and errors (Bärlocher, 1999). For

each possible pair of leaf types, we calculated differences between the original

consumption values. These differences were shuffled and randomly assigned among

pairs. For each pair of leaf type, the average of shuffled differences was compared

with the average of original differences 10,000 times. The P-value was determined by


42

dividing the number of times that the average shuffled differences was larger than the

average of original differences by the number of permutations used. By convention, P

< 0.05 would lead to the rejection of H0 which shows that there is no significant

difference in the consumption of the two leaf types that constitute the pair. We also

used t-tests (independent samples) to determine if there were significant differences in

larvae consumption rates of each leaf type under laboratory and field conditions (Zar,

1999).

Growth and survival experiments

To assess the food value of each studied leaf type, we determined in laboratory

growth and survival of Phylloicus larvae fed on Eucalyptus, Myrcia, Miconia, and

Protium. Larvae selected for these experiments presented initial interocular distance

ranging between 0.54 mm and 1.11 mm. The initial dry mass of each larva was

estimated by relating the interocular distance (mm) to animal dry mass (mg). The

established relationship for this population of Phylloicus was (Becker et al., 2009):

ln DM = 2.50 + 3.84 (ln ID) r2 = 0.71

where ID = interocular distance and DM = animal dry mass. Interocular distance was

measured as the minimum distance between eyes, parallel to head width, with a Zeiss

dissecting microscope fitted with an ocular micrometer (50x magnification).

After acclimatization, larvae were allocated individually to the aerated feeding

arenas described above. Each arena contained 5 discs (1.4 cm diameter) from one of

the four leaf types (treatments). Food, gravel, and filtered spring water were replaced

every week and arenas were maintained under the same temperature (21° C) and

photoperiod conditions (12 L : 12 D). Growth trials were to be maintained until pupal

development was first observed or until other factors made termination necessary.

The final dry mass of each larva was measured directly. Due to limitation in number

of organisms collected per sampling day, we used a randomized block design and

trials were run in 3 blocks across 3 months (one block per month). The same number

of replicates of each food treatment was assigned to each block and, in total, 56

replicates per food treatment were set up.

Larvae daily growth rates (DGR) and specific growth rates (SDGR) were

obtained according to Feio & Graça (2000):


43

DGR = (DMf - DMi)/t

SDGR = (DGR/DMi) × 100

where DMi is the initial dry mass, DMf is the final dry mass, and t is the feeding

period (days). The effect of leaf species on larvae growth was analyzed by ANOVA

for Randomized Block Design (Zar, 1999), after checking for normality and

homogeneity of variances (log transformed data). Only larvae surviving until the end

of the experiment were included in analyses.

The feeding arenas were checked every day and survival (% of larvae alive on

each day) was calculated. The logrank test (Hutchings et al., 1991) was used to

compare survivorship of larvae in the four food treatments.

Results

Leaf litter of Eucalyptus, Myrcia, Miconia, and Protium differed in their initial

chemical composition (Table 1). Nitrogen and phosphorus contents varied among leaf

species. Eucalyptus had more than double the nutrient content of Protium, the native

species with the highest nitrogen and phosphorus contents. The N : P ratio for

Miconia was the highest (83 : 1) and differed from the ones presented by Myrcia (37 :

1), Protium (26 : 1), and Eucalyptus (22 : 1). Protium had the lowest amount of

tannins (2.44%) while the other three species presented similar amounts (6.00 -

8.16%). Miconia had the highest amount of lignin (36.23%) and the highest amounts

of cellulose were presented by Protium, Miconia, and Myrcia (26.93 - 23.07%). The

lowest amounts of lignin and cellulose were presented by Eucalyptus (12.87 and

15.93%, respectively; Table 1).

Table 1 Leaf chemistry values for Eucalyptus, Myrcia, Miconia, and Protium (mean ± SE; n = 4), and ANOVA F values.

Litter chemistry Eucalyptus Myrcia Miconia Protium F Nitrogen (%) 1.65 ± 0.03a 0.66 ± 0.02b 0.59 ± 0.01c 0.74 ± 0.01d 656.4* Phosphorus (%) 0.076 ± 0.002a 0.018 ± 0.001b 0.007 ± 0.001c 0.029 ± 0.001d 678.0* Nitrogen : Phosphorus 21.87 ± 0.14a 37.13 ± 0.30a 82.89 ± 2.96b 25.98 ± 0.25a 24.8* Tannins (%) 8.16 ± 1.06a 8.11 ± 0.39a 6.00 ± 0.52a 2.44 ± 0.28b 25.2* Lignin (%) 12.87 ± 0.74a 30.68 ± 1.07b 36.23 ± 0.20c 28.85 ± 1.07b 148.4* Cellulose (%) 15.93 ± 0.59a 23.07 ± 1.16b 23.69 ± 0.49b 26.93 ± 0.97b 31.8*

*P < 0.001. Values sharing superscript letters are not statistically different (P > 0.05).


44

In the food preference experiments, Phylloicus mortality rates were low

(approximately 1% in both laboratory and field) and none of the larvae used leaf discs

for case-building. Larvae consumed a significant amount of all studied leaf types with

the exception of Protium (paired t-test, P > 0.05), both under laboratory and field

conditions. In laboratory, discs of Eucalyptus were more consumed than the ones of

Myrcia (Figure 1A; Table 2). Discs of Miconia were consumed on intermediate rates,

not differing from the other two species. In the field, no significant differences were

found in the consumption of Eucalyptus, Myrcia, and Miconia discs (Figure 1B;

Table 2). Consumption rates observed under laboratory and field conditions did not

differ in any leaf type (Table 3).

Fig. 1 Food preferences of Phylloicus larvae exposed to Eucalyptus, Myrcia, Miconia, and Protium in multiple-choice feeding experiments done in laboratory (A) and field (B). Mean ± SE. NC = no consumption. Values sharing superscript letters are not statistically different (P > 0.05).


45

Table 2 Analysis of leaf consumption data. Differences between averages of consumption of each possible pair of leaf types and P values of permutation tests.

Difference P Laboratory Field Laboratory Field Eucalyptus - Myrcia +0.227 +0.156 0.047 0.088 Eucalyptus - Miconia +0.206 +0.126 0.117 0.287 Myrcia - Miconia −0.021 −0.029 0.998 0.968

Table 3 Consumption rates (mg mg-1 day-1) of Eucalyptus, Myrcia, and Miconia (mean ± SE) and t-tests results.

Consumption Laboratory Field t df P Eucalyptus 0.436 ± 0.041 0.322 ± 0.042 1.909 156 0.058 Myrcia 0.209 ± 0.036 0.166 ± 0.047 0.702 156 0.484 Miconia 0.230 ± 0.038 0.195 ± 0.030 0.715 156 0.476

The growth experiment was terminated after 21 days when larval mortality in

the Protium food treatment became excessive (68%). Only a few larvae in all food

treatments used available leaf discs for case-building. Phylloicus larvae grew

positively in the four food treatments, with daily growth rates ranging from 0.08 to

0.14 mg day-1 and specific growth rates ranging from 4.76 to 6.83% (Figure 2).

However, none of these growth rates had significant differences among food

treatments (Table 4). At the end of 21 days, 55% of the 224 Phylloicus larvae that

started the experiment died over its course. Survival of larvae fed Eucalyptus and

Miconia was significantly higher than those fed Myrcia and Protium (logrank statistic

> -2.88, P < 0.05; Figure 3). No significant differences were found in larvae survival

within these two groups (logrank statistic < 1.05, P > 0.05).

Table 4 Results from ANOVA for Randomized Block Design of general differences in Phylloicus daily growth rates (DGR) and specific growth rates (SDGR).

Factor SS MS F df P DGR Food treatment 0.03 0.01 1.12 3 0.346 Block 0.58 0.29 31.71 2 < 0.001 Error 0.86 0.01 94

SDGR Food treatment 0.99 0.33 0.50 3 0.684 Block 34.55 17.27 26.16 2 < 0.001 Error 62.06 0.66 94

Data are from Fig. 2


46

Fig. 2 Daily growth rates (DGR, A) and specific growth rates (SDGR, B) of Phylloicus larvae fed on Eucalyptus, Myrcia, Miconia, and Protium for 21 days. Mean ± SE.

Fig. 3 Survival (%) of Phylloicus larvae during growth experiments (21 days). , Eucalyptus; , Myrcia; , Miconia; , Protium.


47

Discussion

Eucalyptus had the highest nutrient content and was the softest leaf type among those

used. These leaf characteristics were probably responsible for the attractiveness of

this exotic species to Phylloicus. Contrary to our expectations, Eucalyptus discs were

not more consumed than all native leaf types. This suggests that the high levels of

tannins, chemical compounds that decrease leaf palatability to shredders (Canhoto &

Graça, 1999; Graça & Bärlocher, 2005), might have diminished nutrient effects.

Nonetheless, the consumption rates of Eucalyptus observed here contrast with the

ones from other studies that also evaluated the food preferences of shredders in the

presence of Eucalyptus leaves (see Canhoto & Graça, 1995; Yeates & Barmuta, 1999)

and found lower consumption of this leaf species.

The lack of consumption of Protium in the presence of other species with

lower nutrient content and higher tannins concentrations suggests that leaves of this

species may present other defensive compounds that difficult shredders consumption.

In fact, this plant species and other Neotropical species from the family Burseraceae

have leaves with an epicuticular wax and many oil ducts (Watson & Dallwitz, 1992)

that may act as a barrier to microbial colonization and shredder feeding. Recently,

studies done by Rüdiger et al. (2007) demonstrated that some of these leaf oils present

antimicrobial properties. The slowest decomposition rate observed for Protium leaves

in Cerrado streams (k = -0.0020; Moretti et al., 2007) and the low survivorship of

Phylloicus larvae feeding on this species (32% after 21 days) also support this

hypothesis.

Several studies have reported the effects of leaf chemistry and toughness on

shredders feeding (see Graça, 2001). Rincón & Martínez (2006) observed that

Phylloicus sp. larvae from Venezuela preferred to feed on leaves with high nutrient

content and low lignin and polyphenol concentrations. In Hong Kong, Li & Dudgeon

(2008) found that leaf toughness was the primary determinant of feeding and fitness

of Anisocentropus maculatus, a caddisfly from the same family Calamoceratidae. In

the present study, Phylloicus food preferences could not be totally explained by litter

initial chemistry and toughness values (evaluated indirectly as lignin contents). And

this may be partially explained by the high lignin contents and low nutrient quality of

native species, which presented N : P ratios ranging among the highest reported for

leaf litter (see Enriquez et al., 1993; Ostrofsky, 1997). These characteristics prevented


48

us to evaluate Phylloicus food preferences using a wide ranking of leaf toughness and

nutritional quality. On the other hand, the majority of Cerrado leaves present similar

characteristics and stream shredders are naturally exposed to this type of leaf litter. As

an adaptation to the low nutrient availability of Savannah-like ecosystems such as the

Brazilian Cerrado, tree species absorb high quantities of nutrients from leaves before

abscission, producing leaf litter of low quality dominated by structural and inhibitory

compounds (Marques et al., 2000; Oliveira & Marquis, 2002).

Phylloicus larvae exhibited similar food preferences in laboratory and field

conditions. The absence of differences between consumption of Eucalyptus and

Myrcia observed in the field trials might have resulted from the entrance of fine

particulate organic matter (FPOM) in the pipes, which served as an alternate food

source to larvae and diminished leaf discs consumption. However, our decision to run

the food preference experiment both in laboratory and field was to confirm if shredder

preferences and consumption rates measured in laboratory were reliable. Given that

no differences were found between consumption rates observed in both conditions for

each leaf type, the applicability of laboratory approach to evaluate the food

preferences of shredders was corroborated in our study.

The stoichiometric theory implies that food quality is relative based on the

nutritional requirements of individual consumers (Cross et al., 2003). Although

having higher nutrient contents, the growth experiments revealed that Eucalyptus was

not a food source of better quality to Phylloicus. Growth rates were highly variable

and no differences were found among food treatments. Similar patterns were observed

by Rincón & Martinez (2006) assessing the growth of Phylloicus sp. when exposed to

the tropical Ficus sp. and Tabebuia rosea leaves. Comparing to the present study,

these authors also found higher specific daily growth rates of Phylloicus larvae,

corroborating the low food value of the leaf species used here. In addition, daily

growth rates measured in our study were almost 3 times lower than the one measured

in a similar experiment, when Phylloicus larvae from this same population fed on

Alnus glutinosa, a fast-decomposing and nitrogen-rich leaf species from temperate

latitudes (M.S. Moretti, unpublished data). Therefore, we believe that larvae growth

in this study was probably limited by nutrients and presence of secondary compounds.

Phylloicus survival percentages were low if compared to shredders from

temperate streams (see Canhoto & Graça, 1995; Going & Dudley, 2008).

Furthermore, survival of larvae fed on Eucalyptus and Miconia was higher than larvae


49

fed on Myrcia and Protium, indicating that, among the three preferred leaf types,

Myrcia was the one of worse food quality to larvae. The same leaves used in this

study were also used to assess Phylloicus case-building behavior (Moretti et al.,

2009). Surprisingly, the species that were more used to build cases (Eucalyptus and

Myrcia) were also consumed by larvae. However, Phylloicus selection in that

experiment was related to leaf abundance and phenolic concentrations. Taken

together, the results of both studies confirm the low attractiveness of Protium leaves

in multiple-choice trials, which are not consumed and less used for case-building.

To summarize, our results indicate that Phylloicus feeding is more influenced

by leaf quality than the origin (native/exotic) of the leaves. These findings are in

accordance with other studies that also evaluated the feeding preferences of

invertebrate shredders (Canhoto & Graça, 1995; Going & Dudley, 2008; Li &

Dudgeon, 2008; Moline & Poff, 2008). Additionally, the low quality of food

resources available in Cerrado streams might impose a constraint to shredder

consumption that probably reflects their lower abundance and the slow breakdown

rates of native species in these ecosystems (Gonçalves et al., 2007; Moretti et al

2007). Although our results suggest that Eucalyptus supports similar shredders growth

as native species, the ecological implications of riparian invasion by exotic trees on

natural streams will depend not only on leaf chemistry and invertebrates assimilation

but also on the availability of leaf litter on streams (Moline & Poff, 2008). It is known

that afforestation with Eucalyptus produces changes in stream litter budgets (Graça et

al., 2002; Bañuelos et al., 2004) and, especially in the case of Cerrado streams,

provides a more ephemeral food resource to aquatic communities (J.F. Gonçalves,

unpublished data). Thus, we hypothesize that the replacement of native riparian

forests with Eucalyptus trees would alter food availability to consumers in Cerrado

streams. The assimilation efficiency and survival of Phylloicus larvae during longer

periods feeding on Eucalyptus leaves should also be evaluated to completely

understand the effects of this exotic species on shredders fitness and, consequently,

stream food webs.

Acknowledgements We are grateful to Juliana França and Lurdemar Tavares for

field and laboratory assistance. This research was supported by FAPEMIG, CNPq,

Capes Foundation, Eawag, and US Fish & Wildlife Services. We also thank

COPASA-MG and IEF-MG for logistical facilities and research licenses.


50

References

Abelho M. (2001) From litterfall to breakdown in streams: a review. The Scientific

World, 1, 656-680.

Arsuffi T.L. & Suberkropp K. (1986) Growth of two stream caddisflies (Trichoptera)

on leaves colonized by different fungal species. Journal of the North American


Bañuelos R., Larrañaga S., Elosegi A., Pozo J. (2004) Effects of eucalyptus

plantations on CPOM dynamics in headwater streams: a manipulative approach.

Archiv für Hydrobiologie, 159, 211-228.

Bärlocher F. (1999) Biostatistik. Georg Thieme-Verlag, Stuttgart.

Bärlocher F. (2005) A primer for statistical analysis. In: Methods to study litter

decomposition (Eds. M.A.S. Graça, F. Bärlocher & M.O. Gessner), pp. 313-329.

Springer, Dordrecht, Berlin, Heidelberg, New York.

Bastian M., Pearson R.G. & Boyero L. (2008) Effects of diversity loss on ecosystem

function across trophic levels and ecosystems: A test in a detritus-based tropical

food web. Austral Ecology, 33, 301-306.

Becker B., Moretti M.S. & Callisto M. (2009) Length dry-mass relationships for a

typical shredder in Brazilian streams (Trichoptera: Calamoceratidae). Aquatic

Insects, 31, in press.

Canhoto C. & Graça M.A.S. (1995) Food value of introduced eucalypt leaves for a

Mediterranean stream detritivore: Tipula lateralis. Freshwater Biology, 34, 209-

214.

Canhoto C. & Graça M.A.S. (1999) Leaf barriers to fungal colonization and shredders

(Tipula lateralis) consumption of decomposing Eucalyptus globulus. Microbial

Ecology, 37, 163-172.

Canhoto C., Graça M.A.S. & Bärlocher F. (2005). Feeding preferences. In: Methods

to study litter decomposition (Eds. M.A.S. Graça, F. Bärlocher & M.O. Gessner),

pp. 297-302. Springer, Dordrecht, Berlin, Heidelberg, New York.

Cross W.F., Benstead J.P., Rosemond A.D. & Wallace J.B. (2003) Consumer-

resource stoichiometry in detritus-based streams. Ecology Letters, 6, 721-732.

Cummins K.W. & Klug M.J. (1979) Feeding ecology of stream invertebrates. Annual

Review of Ecology and Systematics, 10, 147-172.


51

Cummins K.W., Wilzbach M.A., Gates D.M., Perry J.B. & Taliaferro W.B. (1989)

Shredders and riparian vegetation. Bioscience, 39, 24-30.

Enriquez S., Duarte C.M. & Sandjensen K. (1993) Patterns in decomposition rates

among photosynthetic organisms – the importance of detritus C-N-P content.

Oecologia, 94, 457-471.

Feio M.J. & Graça M.A.S. (2000) Food consumption by the larvae of Sericostoma

vittatum (Trichoptera), an endemic species from the Iberian Peninsula.

Hydrobiologia, 439, 7-11.

Flindt M.R. & Lillebo A.I. (2005) Determination of total nitrogen and phosphorus in

leaf litter. In: Methods to study litter decomposition (Eds. M.A.S. Graça, F.

Bärlocher & M.O. Gessner), pp. 53-59. Springer, Dordrecht, Berlin, Heidelberg,

New York.

Friberg N. & Jacobsen D. (1994) Feeding plasticity of two detritivore-shredders.

Freshwater Biology, 32, 133-142.

Gessner M.O. (2005) Proximate lignin and cellulose. In: Methods to study litter

decomposition (Eds. M.A.S. Graça, F. Bärlocher & M.O. Gessner), pp. 115-120.

Springer, Dordrecht, Berlin, Heidelberg, New York.

Going B.M. & Dudley T.L. (2008) Invasive riparian plant litter alters aquatic insect

growth. Biological Invasions, 10, 1041-1051.

Gonçalves, J.F., Graça M.A.S. & Callisto M. (2007) Litter decomposition in a

Cerrado savannah stream is retarded by leaf toughness, low dissolved nutrients and

a low density of shredders. Freshwater Biology, 52, 1440-1451.

Graça M.A.S. (2001) The role of invertebrates on leaf litter decomposition in streams

– a review. International Review of Hydrobiology, 86, 383-393.

Graça M.A.S. & Bärlocher F. (2005) Radial diffusion assay for tannins. In: Methods

to study litter decomposition (Eds. M.A.S. Graça, F. Bärlocher & M.O. Gessner),

pp. 101-105. Springer, Dordrecht, Berlin, Heidelberg, New York.

Graça M.A.S., Cressa C., Gessner M.O., Feio M.J., Callies K.A. & Barrios C. (2001)

Food quality, feeding preferences, survival and growth of shredders from

temperate and tropical streams. Freshwater Biology, 46, 947-957.

Graça M.A.S., Pozo J. Canhoto C. & Elosegi A. (2002) Effects of Eucalyptus

plantations on detritus, decomposers, and detritivores in streams. The Scientific

World, 2, 1173-1185.


52

Hutchings M.J., Booth K.D. & Waite S. (1991) Comparison of survivorship by the

logrank test: criticisms and alternatives. Ecology, 72, 2290-2293.

Klink C.A. & Machado R.B. (2005) Conservation of the Brazilian Cerrado.

Conservation Biology, 19, 707-713.

Li A.O.Y. & Dudgeon D. (2008) The effects of leaf litter characteristics on feeding

and fitness of a tropical stream shredder, Anisocentropus maculatus (Trichoptera:

Calamoceratidae). Marine and Freshwater Research, 59, 897-901.

Maquere V., Laclau J.P., Bernoux M., Saint-Andre L., Gonçalves J.L.M., Cerri C.C.,

Piccolo M.C. & Ranger J. (2008) Influence of land use (savanna, pasture,

Eucalyptus plantations) on soil carbon and nitrogen stocks in Brazil. European

Journal of Soil Science, 59, 863-877.

Marques A.R., Garcia Q.S., Resende J.L.P. & Fernandes G.W. (2000) Variations in

leaf characteristics of two species of Miconia in the Brazilian cerrado under

different light intensities. Tropical Ecology, 41, 47-60.

Mathuriau C., Thomas A.G.B. & Chauvet E. (2008) Seasonal dynamics of benthic

detritus and associated macroinvertebrate communities in a neotropical stream.

Fundamental and Applied Limnology, 171, 323-333.

Miritz L.D., Gomes L.S., Noguez M.B. & Palmeira E.M. (2008) Reflorestamento:

mercado interno e externo. Observatorio de la Economía Latinoamericana, 105, 1-

8.

Moline A.B. & Poff N.L. (2008) Growth of an invertebrate shredder on native

(Populus) and non-native (Tamarix, Elaeagnus) leaf litter. Freshwater Biology, 53,

1012-1020.

Moretti M.S., Gonçalves J.F., Ligeiro R. & Callisto M. (2007) Invertebrates


Review of Hydrobiology, 92, 199-210.

Moretti M.S., Loyola R.D., Becker B. & Callisto M. (2009) Leaf abundance and

phenolic concentrations codetermine the selection of case-building materials by

Phylloicus sp. (Trichoptera, Calamoceratidae). Hydrobiologia, 630, 199-206.

Myers N., Mittermeier R.A., Mittermeier C.G., Fonseca G.A.B. & Kent J. (2000)

Biodiversity hotspots for conservation priorities. Nature, 403, 853-858.

Naiman R.J., Décamps H. & McClain M.E. (2005) Riparia: Ecology, Conservation,

and Management of Streamside Communities. Elsevier Academic Press,

Burlington, San Diego, London.


53

Oliveira P.S. & Marquis R.J. (2002) The cerrados of Brazil: Ecology and Natural

History of Neotropical Savanna. Columbia University Press, New York.

Ostrofsky M.L. (1997) Relationship between chemical characteristics of autumn-shed

leaves and aquatic processing rates. Journal of the North American Benthological

Society, 16, 750-759.

Prather A.L. (2003) Revision of the Neotropical caddisfly genus Phylloicus

(Trichoptera: Calamoceratidae). Zootaxa, 275, 1-214.

Reinhart K.O. & VandeVoort R. (2006) Effect of native and exotic leaf litter on

macroinvertebrate communities and decomposition in a western Montana stream.

Diversity and Distributions, 12, 776-781.

Rincón J. & Martínez I. (2006) Food quality and feeding preferences of Phylloicus sp.

(Trichoptera: Calamoceratidae). Journal of the North American Benthological

Society, 25, 209-215.

Roa R. (1992) Design and analysis of multiple-choice feeding-preference

experiments. Oecologia, 89, 509-515.

Rüdiger A.L., Siani A.C. & Veiga, V.F. (2007) The chemistry and pharmacology of

the South America genus Protium Burm. f. (Burseraceae). Pharmacognosy

Reviews, 1, 93-104.

Thompson R.M. & Townsend C.R. (2003) Impacts on stream food webs of native and

exotic forest: an intercontinental comparison. Ecology, 84, 145-161.

Vannote R.L., Minshall G.W., Cummins K.W., Sedell J.R. & Cushing C.E. (1980)

The river continuum concept. Canadian Journal of Fisheries and Aquatic Science,

37, 130-137.

Vitousek P.M. (1990) Biological invasions and ecosystem processes: towards an

integration of population biology and ecosystem studies. Oikos, 57, 7-13.

Wallace J.B. & Webster J.R. (1996) The role of macroinvertebrates in stream

ecosystem function. Annual Review of Entomology, 41, 115-139.

Wallace J.B., Eggert S.L., Meyer J.L., Webster J.R. (1999) Effects of resource

limitation in a detrital-based ecosystem. Ecological Monographs, 69, 409-442.

Watson L. & Dallwitz M.J. (1992) The families of flowering plants: descriptions,

illustrations, identification, and information retrieval. Version: November 2008.

http://delta-intkey.com

Webster J.R. & Benfield E.F. (1986) Vascular plant breakdown in freshwater

ecosystems. Annual Review of Ecology and Systematics, 17, 567-594.


54

Yeates L.V. & Barmuta L.A. (1999) The effects of willow and eucalypt leaves on

feeding preference and growth of some Australian aquatic macroinvertebrates.

Australian Journal of Ecology, 24, 593-598.

Zar J.H. (1999) Biostatistical Analysis. Prentice-Hall, New Jersey.

4In%luenceoflowqualityleavesonthefeedingactivityoftropicaland

temperateshredders

MarceloS.Moretti1,2,MarcosCallisto1&MarkO.Gessner2,3

1Lab.EcologiadeBentos,InstitutodeCiênciasBiológicas,UniversidadeFederaldeMinasGerais,Av.AntônioCarlos6627,


2DepartmentofAquaticEcology,Eawag:SwissFederalInstituteofAquaticScienceandTechnology,Ueberlandstrasse133,

8600Dübendorf,Switzerland

3InstituteofIntegrativeBiology,ETHZurich,CHNH68,8092Zürich,Switzerland

Specimens of Phylloicus sp.(Trichoptera, Calamoceratidae) andGammarus pulex (Amphipoda,Gammaridae) sampled in Taboõesspring (Brazil) and Chriesbachstream(Switzerland),respectively.

Chapter4:FeedingactivityofPhylloicusandGammarus

56

Influence of low quality leaves on the feeding activity of

tropical and temperate shredders

Marcelo S. Moretti, Marcos Callisto & Mark O. Gessner

Abstract We hypothesized that shredders activity in Brazilian Cerrado streams is

restricted by the quality of available leaves. Thus, we tested whether shredders from

tropical (Phylloicus sp.) and temperate (Gammarus pulex) streams would exhibit

similar food preferences, and whether low quality leaf diets would affect shredders

consumption and survivorship similarly. Leaves were chosen such that we had two

pairs composed by one tropical and one temperate leaf species each, presenting low

lignin and N contents (Swietenia macrophylla and Betula pubescens) and high lignin

and moderate N contents (Hymenaea courbaril and Fagus sylvatica). Phylloicus and

Gammarus exhibited similar food preferences. Betula was the most consumed leaf

species (Phylloicus: 0.11 to 0.17 mg mg-1 day-1; Gammarus: 0.14 to 0.22 mg mg-1

day-1) while Hymenaea was not consumed. Fagus and Swietenia were little consumed

by Gammarus (0.02 and 0.01 mg mg-1 day-1, respectively) and Phylloicus preferred

none of these leaf species. According to shredders preference, leaf species were

ranked as follow: Betula > Swietenia = Fagus > Hymenaea. Consumption rates of

both shredders were affected by leaf species. Phylloicus consumed more Swietenia

than the other 3 leaf species while Gammarus consumed more Swietenia and Betula

than Hymenaea and Fagus. After 4 weeks, Phylloicus survivorship was extremely low

(9% across all treatments) and larvae fed on Betula survived more than those fed on

Hymenaea. Gammarus survival was much higher than Phylloicus (80%) and no

differences were found among food treatments. Our results suggest that the low

abundance of effective shredders in Brazilian Cerrado streams is probably more

linked to the low quality of leaves than a different behavior exhibited by these

invertebrates.

Keywords Leaf chemistry, food preferences, Phylloicus sp., Gammarus pulex,

Brazilian Cerrado streams.


57

Introduction

Shredders are the functional feeding group of invertebrates whose mouthparts allow

them to consume leaf litter effectively (Cummins & Klug 1979). Therefore,

individuals of this group perform a key role in the energetics of shaded streams,

where leaf allochthonous inputs constitute the main source of organic matter (Petersen

& Cummins 1974, Webster & Benfield 1986). Shredders activity produces leaf

fragments and fecal pellets, which can be subsequently used as food by other

functional feeding groups (Heard & Richardson 1995, Wallace and Webster 1996,

Hieber & Gessner 2002, Jonsson & Malmqvist 2005).

Studies done in temperate streams demonstrated that shredders abundance or

biomass is positively related to leaf litter breakdown rates (Sponseller & Benfield

2001, Hagen et al. 2006). Additionally, consumption of leaf litter by shredders might

vary in response to litter availability (Cummins et al. 1989), leaf chemistry (Irons et al.

1988, Campbell & Fuchshuber 1995) and microbial conditioning (Graça et al. 1993,

Rong et al. 1995). In this sense, the chemical attributes of leaf species (e.g., C:N or

lignin:N) may influence shredders feeding and resource use in these ecosystems

(Ostrofsky 1997, Swan & Palmer 2006). Contrasting with all existing knowledge

about the feeding behavior of shredders at higher latitudes, few studies concerning

this issue were done in the tropics. Nonetheless, there is evidence that tropical

shredders exhibit the same basic patterns of food exploitation as their temperate

counterparts (Graça et al. 2001, Rincón & Martinez 2006, Li & Dudgeon 2008).

Brazilian Cerrado leaf species are known for presenting high toughness and

contents of inhibitory compounds (Oliveira & Marquis 2002, Moretti et al. 2009).

Moreover, these leaves also present lower nutrient content if compared to leaves from

temperate zones (Gonçalves et al. 2007, Moretti et al. 2007). All these characteristics,

associated with a low seasonality on leaf fall (Gonçalves et al. 2006), result in a

diversity of litter types varying in food quality that may affect shredders consumption

(Irons et al. 1988, Bastian et al. 2007, Jacobsen et al. 2008, Li & Dudgeon 2008).

We hypothesized that the activity of shredders in Brazilian Cerrado streams is

restricted by the quality of available leaves. In order to evaluate how low quality

leaves affect the activity of shredders from tropical and temperate streams, we

determined food preferences, consumption rates, FPOM production and survival of

Phylloicus sp. Müller and Gammarus pulex L. exposed to four leaf species commonly


58

found in Brazilian Cerrado and European deciduous forests. These leaf species were

chosen in such a way that we had two pairs, each composed by one tropical and one

temperate species presenting identical lignin and similar N contents. Specifically, we

tested whether Phylloicus and Gammarus would exhibit similar food preferences

among leaves of different quality, and whether low quality leaf diets would affect the

consumption and survivorship of both shredders similarly.

Methods

Shredders

The tropical Phylloicus sp. (Trichoptera: Calamoceratidae) and the temperate

Gammarus pulex (Amphipoda: Gammaridae) are common leaf-shredding detritivores

that inhabit small streams in south east Brazil and north east Switzerland, respectively

(Kelly et al. 2002, Dangles et al. 2004, Becker et al. 2009). Phylloicus live in pools,

and can be found where fallen leaves form patches that can be used as food source

and material to build their cases (Prather 2003). Gammarus are normally found in

high abundances associated to macrophytes in rapid and pool reaches. Individuals of

Phylloicus larvae were collected in Taboões spring, located inside the Rola Moça

State Park, Minas Gerais State, Brazil (20° 03’ 38” S - 44° 03’ 03” W) and

Gammarus were collected in Chriesbach stream at Dübendorf, Switzerland (47° 24’

22” N - 08° 35’ 49” O). Taboões spring present leaf patches on the streambed during

all year and waters are well oxygenated, alkaline with low conductivity and nutrient

concentrations (Moretti et al. 2009). Chriesbach is a macrophyte-rich stream of the

Swiss Plateau that presents high levels of major nutrients (N, P). The catchment of the

Chriesbach mainly consists of agricultural, industrial and residential areas (Kaenel et

al. 2000). After collection, individuals were immediately brought to the laboratory

and acclimatized in an aerated aquarium for 2 days (Phylloicus: 21º C, 12L : 12D

photoperiod; Gammarus: 17º C, 16L : 8D photoperiod).

Leaves

Senescent leaves of the tropical Swietenia macrophylla King and Hymenaea courbaril

L. were collected at once in the campus of Federal University of Minas Gerais (Belo

Horizonte, Brazil) on August 2007 while leaves of the temperate Betula pubescens

Ehrh. and Fagus sylvatica L. were collected on November 2007 from single trees


59

located near Chriesbach stream. These leaf species were chosen in such a way that we

had a pair composed by one tropical and one temperate species with high lignin and

low nitrogen contents. A second pair of tropical and temperate species had moderately

low lignin content and similar nitrogen content. All leaves were conditioned for 7

days in fine mesh bags (mesh size: 0.5 mm) in the respective streams before offering

to shredders.

Some leaves of the four studied species were ground for leaf initial chemistry

determination. Total N content of ~5-mg subsamples was determined with a CHNS-

Analyzer (EuroEA 3000; EuroVector S.p.A., Milan, Italy). P was determined on ~5-

mg subsamples as soluble reactive P after digestion with K2S2O8 in an autoclave at

121º C (Ebina et al. 1983). Lignin was measured gravimetrically (~250-mg

subsamples) as acid detergent lignin (Gessner 2005) while tannins were determined

on ~100-mg subsamples as tannic acid equivalents by the radial diffusion assay

(Graça & Bärlocher 2005).

Fungal biomass was analyzed from conditioned leaves. Ten leaf discs (1.4 cm

diameter) were cut from 5 different leaves of each species (2 discs per leaf) using a

cork borer. One set of 5 discs was placed in a small plastic bag and frozen at -20 ºC

for ergosterol analysis to provide an estimate of fungal biomass. Leaf ergosterol

content was quantified according to Gessner & Schmitt (1996) and converted to

fungal biomass based on an average ergosterol content of 5.5 mg per g fungal dry

mass (Gessner & Chauvet 1993). The second set of leaf discs was placed in a separate

aluminum pan, and dried to constant mass at 60 ºC for 72 h, before weighing to the

nearest 0.01 mg. Leaf chemical characteristics and fungal biomass were compared

among leaf species by one-way ANOVA followed by Tukey’s HSD test (after arcsine

square root transformation).

Food preference experiments

A total of 120 individuals of similar size of each shredder (Phylloicus: 6.53 ± 0.42 mg;

Gammarus: 3.28 ± 0.17 mg; body weight ± SE) were individually placed in plastic

containers (Phylloicus: 12 cm diameter × 9 cm high; Gammarus: 10 × 10 × 12 cm

high), containing 400 ml of filtered spring water, which were aerated continuously

and kept under temperature and light conditions described above. To evaluate food

preferences, we adopted an experimental design similar to the one used by Graça et al.


60

(2001). Shredders (containers) were randomly distributed among 6 treatments, each

composed by pairs of the four studied leaf species: Swietenia, Hymenaea, Betula and

Fagus.

Food was offered in the form of leaf discs (1.4 cm diameter). Pairs of leaf

discs were cut from the same leaf using a cork borer (one from each side of the main

vein). The control discs were placed inside a small 0.5 mm mesh bag and suspended

from the rim of the containers, while exposed discs were pierced with colored pins

(one color for each species) and offered to shredders. The number of replicates for

each shredder in each pairwise comparison was 20. Shredders were allowed to feed

until one of the exposed leaf discs was reduced to about one-third of its initial size.

After the feeding period, shredders, control and exposed discs were placed in separate

aluminum pans, dried to constant mass at 60 °C for 72 h and weighed to the nearest

0.01 mg. Food consumption was expressed in terms of mg of leaf dry mass ingested

(difference between control and exposed discs) per individual biomass (mg) over the

feeding period (days).

Data analysis was done according to Petersen & Renauld (1989) and Friberg

& Jacobsen (1994). Firstly, we determined whether shredders had consumed a

significant amount of leaves during the experiment. Paired t-tests were used to

compare the average weight of exposed and control discs of each species.

Consumption was considered to have occurred when at least one of the two leaf discs

exposed to a shredder was significantly lower than that of the corresponding control

disc. Once consumption had occurred, the next procedure was to test, again by paired

t-tests, whether one of the two offered food items was significantly more consumed,

that is preferred.

Consumption and survival experiments

We evaluated consumption rates, FPOM production and survival of Phylloicus and

Gammarus fed on Swietenia, Hymenaea, Betula and Fagus. Eighty animals of each

shredder were used in these experiments (20 per food treatment). The average initial

body weight (± SE) for Phylloicus was 2.19 ± 0.10 mg and for Gammarus was 2.56 ±

0.12 mg. After the acclimatization period, animals were placed individually in the

aerated plastic containers described above. Two conditioned leaf discs (1.4 cm

diameter) of one food treatment were offered in each container and animals were


61

allowed to feed ad libitum for 4 weeks. Leaf discs and water were replaced every

week.

Remaining leaf discs were placed individually in aluminum pans and dried to

constant mass at 60 ºC for 72 h, before weighing to the nearest 0.01 mg. Food

consumption was determined weekly by the difference between initial and final dry

mass of offered discs. Leaf discs initial dry mass was estimated by a conversion factor

wet mass/oven dry mass calculated previously for each studied leaf species. Control

replicates were set to determine discs mass loses in the absence of shredders and

correct leaf consumption values. Consumption rates were determined by the amount

of food ingested (mg) per individual biomass (mg) over the feeding period (days).

The total amount of FPOM produced by each shredder was filtered onto preweighed

0.45-µm glass microfiber filters (Grade GF/F, Whatman Ltd.). Filters were dried at 60

ºC for 72 h and reweighed to the nearest 0.01 mg to determine the FPOM production

(mg mg-1 day-1). The effect of leaf species on shredders consumption and FPOM

production was analyzed by one-way ANOVA followed by Tukey’s HSD test, after

checking for normality and homogeneity of variances (square-root transformed data).

Replicates were checked daily and the percentage of animals alive on each day

was calculated. Dead animals and respective containers were removed from the

experiments. The logrank test (Hutchings et al. 1991) was used to compare

survivorship of shredders in the four food treatments. The number of Phylloicus

larvae that had used leaf discs as material for case-building was also recorded.

Because Phylloicus presented high mortality in all food treatments, the relationship

between leaf consumption, and FPOM production (independent variables) to insect

survivorship (dependent variable) was examined with simple linear regressions

followed by residual analysis. All statistical analyses were done with SPSS (version

16.0 for Macintosh, SPSS, Chicago, Illinois) and based on Zar (1999).

Results

Studied leaf species differed in their chemistry (Table 1). Hymenaea presented the

highest N content (0.86%) while Betula and Swietenia had the lowest contents (0.56

and 0.52%, respectively). The P content of Betula was more than 10 times higher than

the one of Hymenaea (0.037%), the species that had the second high content of this

nutrient. Fagus and Hymenaea had higher lignin contents (24.48 and 25.92%) than


62

Betula and Swietenia (15.52 and 15.29%). Tannin contents were higher in Swietenia

(6.63%) than in Betula and Hymenaea (2.72 and 3.33%, Table 1). After conditioning

for one week, Betula leaves were more colonized by aquatic fungi (F = 8.190, P =

0.015, df = 3) than Swietenia and Fagus, while Hymenaea presented intermediate

levels of fungal biomass (Fig. 1).

Table 1 Leaf chemistry values for Swietenia, Hymenaea, Betula and Fagus (mean ± SE; n = 4), and ANOVA F values.

Litter chemistry Swietenia Hymenaea Betula Fagus F Nitrogen (%) 0.52 ± 0.01a 0.86 ± 0.01b 0.56 ± 0.02a 0.62 ± 0.01c 133.99* Phosphorus (%) 0.014 ± 0.001a 0.037 ± 0.001b 0.383 ± 0.020c 0.017 ± 0.001a 821.65* Lignin (%) 15.29 ± 0.21a 25.92 ± 1.16b 15.52 ± 0.53a 24.48 ± 0.24b 84.95* Tannins (%) 6.63 ± 0.76a 3.33 ± 0.32b 2.72 ± 0.15b 4.51 ± 0.52a,b 14.02*

*P < 0.001. Values sharing superscript letters are not statistically different (P > 0.05).

Fig. 1 Fungal biomass of each leaf species after 7 days of incubation in Chriesbach stream (mean ± SE, n = 4).

Food preference experiments lasted 3 weeks for Phylloicus and 5 weeks for

Gammarus. Mortality rates were relatively low (Phylloicus: 12%; Gammarus: 18%)

and none of Phylloicus larvae used exposed discs for case-building. Consumption of

Betula discs was fast by both shredders and all replicates containing this leaf species

were stopped in the first week. Phylloicus and Gammarus exhibited similar food

preferences (Fig. 2). Betula was the most consumed leaf species (Phylloicus: 0.11 to

0.17 mg mg-1 day-1; Gammarus: 0.14 to 0.22 mg mg-1 day-1) and none of the other 3

leaf species were consumed in its presence. Hymenaea was not consumed in any

treatment (t-test < 0.155, P > 0.616). Fagus and Swietenia were little consumed by


63

Gammarus (0.02 and 0.01 mg mg-1 day-1, respectively) and Phylloicus did not differ

between these leaf species (t-test = -1.361, P = 0.190, Fig. 2). The pairwise treatment

containing Swietenia and Fagus was the only one that differed qualitatively between

Phylloicus and Gammarus. According to shredders preference, leaf species were

ranked as follow: Betula > Swietenia = Fagus > Hymenaea.

Fig. 2 Food preferences of Phylloicus (empty bars) and Gammarus (filled bars) when exposed to four leaf species as tested pairwise. Mean ± SE. NC = no consumption. Values sharing superscript letters are not statistically different (P > 0.05).

During the consumption and survival experiments, 25% of Phylloicus larvae

used leaf discs for case-building. Hymenaea and Fagus were the most used species

for this purpose (10 and 8 replicates, respectively). These experiments revealed that,

except for replicates containing Betula, Phylloicus presented higher consumption

rates than Gammarus (Fig. 3A). However, the consumption rates of both shredders

were affected by leaf species (Table 2). Phylloicus consumed more Swietenia than the

other 3 leaf species while Gammarus consumed more Swietenia and Betula than


64

Hymenaea and Fagus. Shredders FPOM production presented similar patterns (Fig.

3B; Table 2). Values of FPOM production were higher in Swietenia and Betula than

in Hymenaea treatments. When feeding on Fagus, Phylloicus produced intermediate

amounts of FPOM, not differing from other food treatments, and Gammarus

produced the lowest amounts.

Fig. 3 Consumption rates (A) and FPOM production (B) of Phylloicus and Gammarus fed on Swietenia, Hymenaea, Betula and Fagus during 4 weeks. Mean ± SE. Values sharing superscript letters are not statistically different (P > 0.05).

Table 2 Results from one-way ANOVAs of general differences in shredders consumption rates and FPOM production among food treatments. Shredder species SS MS F df P Consumption Phylloicus 2.317 0.772 14.884 3 <0.001 Gammarus 2.084 0.695 67.452 3 <0.001 FPOM Phylloicus 0.995 0.332 8.801 3 <0.001 Gammarus 0.892 0.297 40.251 3 <0.001

Data are from Fig. 3


65

Phylloicus survivorship was extremely low with only 9% of larvae remaining

alive after 4 weeks (across all food treatments). Survival percentages of larvae fed on

Betula were higher but differences were only found between Betula and Hymenaea

food treatments (logrank statistic = -2.695, P = 0.007; Fig 4A). Phylloicus

survivorship was positively related to leaf consumption and FPOM production

(consumption: F = 79.253, R2 = 0.504, P < 0.001; FPOM: F = 43.979, R2 = 0.361, P <

0.001). Gammarus survival was much higher than Phylloicus (80% across all food

treatments) and no differences were found among food treatments (logrank statistic <

0.412, P > 0.112; Fig 4B).

Fig. 4 Survival (%) of Phylloicus (A) and Gammarus (B) during 4 weeks. , Swietenia; , Hymenaea; , Betula; , Fagus.


66

Discussion

Our experiments revealed that Phylloicus and Gammarus have similar preferences

when selecting among food resources. When given a choice, both shredders preferred

Betula, the leaf species that presented the highest P content, and rejected Hymenaea

that, together with Fagus, had the highest lignin contents. The low consumption of

Swietenia is probably related to its tannins content, which was higher than in Betula.

These results are in agreement with what was previously observed by other studies

that assessed the feeding preferences of shredders (e.g., Canhoto & Graça 1995,

Motomori et al. 2001, Rincón & Martínez 2006) and give more evidences that tropical

shredders are also able to distinguish among different leaf species as their temperate

counterparts.

Excepting Hymenaea, shredders feeding preferences were in accordance with

the values of fungal colonization. The low biomass of fungi associated to Swietenia

and Fagus after conditioning might be due to the chemistry of these species (Gessner

& Chauvet 1994, Royer & Minshall 2001), which presented lower P contents.

Because fungal colonization affects the nutritional value and palatability of decaying

leaves to shredders (Arsuffi & Suberkropp 1986, Graça et al. 1993, Graça et al. 2001),

these results suggest that Betula was the leaf type of better quality in these

experiments. In addition, the lack of consumption of Hymenaea, which also presented

high levels of N and fungal biomass, by both shredders in any pairwise treatment, is

probably related to its leaf anatomy and chemistry. As a result of the scleromorphic

aspect of Cerrado vegetation, Hymenaea leaves have a high density of veins and also

present a thick epicuticular wax layer and many secondary compounds (Sugayama &

Salatino 1995) that difficult consumption by herbivores (Welker et al. 2007). And

these characteristics might have diminished the attractiveness of this species to

shredders.

The food preference experiments showed that leaf consumption of Phylloicus

and Gammarus was not related with leaf N contents, indicating that possible N effects

on shredders preference were minimized by other factors, such as lignin contents and

presence of secondary compounds. These findings corroborate the studies of Albariño

& Balseiro (2001) and Li & Dudgeon (2008, 2009). Furthermore, because both

shredders had similar food preferences, our results suggest that leaf quality was more

important than the origin of leaf species in determining shredders choices. The lack of


67

association between shredders and native leaves, i.e., leaves they are normally

exposed to in their natural habitats, was also documented by previews studies (see

Parkyn & Winterbourn 1997, Yeates & Barmuta 1999, Li & Dudgeon 2008).

The observed consumption rates demonstrated that the feeding activity of both

shredders was deterred by the high lignin content present in Hymenaea and Fagus

leaves. Despite being considered as a highly opportunistic consumer (Cummins &

Klug 1979, Kelly et al. 2002), our results suggest that Gammarus was more affected

by this leaf structural compound, while Phylloicus consumption rates did not differ

among Hymenaea, Betula and Fagus food treatments. However, the results of FPOM

production, which was mainly composed by fecal pellets, indicate that Hymenaea was

the leaf species that most affected the feeding activity of this tropical shredder. On the

other hand, the higher consumption rates observed for Phylloicus feeding on

Swietenia suggest that these larvae exhibited compensatory feeding (Cruz-Rivera &

Hay 2000). According to Simpson & Simpson (1990), when environmental

constraints confine consumers to lower quality diets, they may still obtain sufficient

nutrients, if they can compensate by increasing their consumption rate. Moreover,

compensatory feeding is likely to be an advantage in environments where food

sources vary seasonally or in an unpredictable way (Graça et al. 2001) and lowers

susceptibility to predation by reducing consumers movement among different food

resources (Stachowicz & Hay 1996).

Although presenting higher consumption rates in all food treatments but

Betula, Phylloicus survivorship was much lower than Gammarus in our experiments.

The effect of leaf species on shredders survivorship was only observed in Phylloicus

trials, where larvae fed on Betula survived more than those fed on Hymenaea. As far

as we know, Phylloicus survival percentages observed in the present study are among

the lowest recorded for shredders in laboratory experiments (see Canhoto & Graça

1995, Cruz-Rivera & Hay 2000, Graça et al. 2001, Li & Dudgeon 2008). We have

two possible reasons for the low survivorship of Phylloicus. First, the low quality of

leaves used in this experiment, which might have constituted resources of low food

value for this shredder. Second, the small size of larvae used, which could have had

more difficult to break down tough leaf discs (Nolen & Pearson 1993, Wantzen &

Wagner 2006). Analysis of gut contents revealed that larvae from all food treatments

were not starving when they died (M.S. Moretti, personal observation). Additionally,

Phylloicus survivorship was positively related with consumption rates and FPOM


68

production, indicating that these two variables partially explained larvae survivorship

(R2 = 0.504 and 0.361, respectively). Thus, we have evidences to believe that leaf

quality is the most likely reason. Given that waters in Taboões are not so rich in

nutrients as in Chriesbach stream, leaves conditioned in this spring were probably less

colonized by fungi (Chung & Suberkropp 2008) and presented lower food quality

when offered to Phylloicus. Furthermore, the high survivorship of Gammarus in all

food treatments suggests that this shredder species might have a better strategy than

Phylloicus to deal with low quality food resources. In fact, Graça et al. (1993)

demonstrated that G. pulex has the ability to compensate for lower food quality by

adjusting its respiration rate.

In conclusion, our results suggest that tropical and temperate shredders have

very similar food preferences when exposed to leaves of different quality, and that

their activity is mainly affected by the presence of structural compounds (e.g., lignin)

when exposed to low quality leaf litter. These findings corroborate our hypothesis,

and indicate that the low abundance of effective shredders in Brazilian Cerrado

streams is probably more linked to the low quality of leaves than a different behavior

exhibited by these invertebrates. Taking into account that tropical forests present a

high diversity of leaf species, and shredders activity is influenced by leaf availability

and quality, we believe that the abundance and importance of invertebrate shredders

on ecological processes in tropical streams, such as leaf litter breakdown, is more

related to local characteristics that determine plant species composition in the riparian

forests (e.g., altitude, vegetation type, and climate) than generalized latitudinal

patterns.

Acknowledgements We thank Richard Illi and the AuA Laboratory at Eawag for

providing nutrient analysis data. We also thank Barbara Becker, Nayara Costa,

Deborah Oliveira, and Clarissa Dantas for laboratory and field assistance. This study

was supported by CAPES Foundation, Eawag, CNPq, FAPEMIG, and US Fish &

Wildlife Services. We also thank COPASA-MG and IEF-MG for logistical facilities

and research licenses.


69

References

Albariño, R. J. & Balseiro, E. G., 2001: Food quality, larval consumption, and growth

of Klapopteryx kuschell (Plecoptera: Austroperlidae) from a south Andes stream. –

Journal of Freshwater Ecology 16: 517-526.

Arsuffi, T. L. & Suberkropp, K., 1986: Growth of two stream caddisflies (Trichoptera)

on leaves colonized by different fungal species. – Journal of the North American

Bethological Society 5: 297-305.

Bastian, M., Boyero, L., Jackes, B. R. & Pearson, R. G., 2007: Leaf litter diversity

and shredder preferences in an Australian tropical rain-forest stream. – Journal of

Tropical Ecology 23: 219-229.

Becker, B., Moretti, M. S. & Callisto, M., 2009: Length dry-mass relationships for a

typical shredder in Brazilian streams (Trichoptera: Calamoceratidae). – Aquatic

Insects 31: in press.

Campbell, I. C. & Fuchshuber, L., 1995: Polyphenols, condensed tannins, and

processing rates of tropical and temperate leaves in an Australian stream. – Journal

of the North American Bethological Society 14: 174-182.

Canhoto, C. & Graça M. A. S., 1995: Food value of introduced eucalypt leaves for a

Mediterranean stream detritivore: Tipula lateralis. – Freshwater Biology 34: 209-

214.

Chung, N. & Suberkropp, K., 2008: Influence of shredder feeding and nutrients on

fungal activity and community structure in headwater streams. – Fundamental and

Applied Limnology 173: 35-46.

Cruz-Rivera, E. & Hay, M. E., 2000: Can quantity replace quality? Food choice,

compensatory feeding, and fitness of marine mesograzers. – Ecology 81: 201-219.

Cummins, K. W. & Klug, M. J., 1979: Feeding ecology of stream invertebrates. –

Annual Review of Ecology and Systematics 10: 147-172.

Cummins, K. W., Wilzbach, M. A., Gates, D. M., Perry, J. B. & Taliaferro, W. B.,

1989: Shredders and riparian vegetation. – Bioscience 39: 24-30.

Dangles, O., Gessner, M. O., Guérold, F. & Chauvet, E., 2004: Impacts of stream

acidification on litter breakdown: implications for assessing ecosystem functioning.

– Journal of Applied Ecology 41: 365-375.


70

Ebina, L., Tsutsui, T. & Shirai, T., 1983: Simultaneous determination of total nitrogen

and total phosphorus in water using peroxodisulfate oxidation. – Water Research

17: 1721-1726.

Friberg, N. & Jacobsen, D., 1994: Feeding plasticity of two detritivore-shredders. –

Freshwater Biology 32: 133-142.

Gessner, M. O., 2005: Proximate lignin and cellulose. – In: Graça, M. A. S.,

Bärlocher, F. & Gessner, M. O. (eds): Methods to study litter decomposition. –

Springer, Dordrecht, pp. 115-120.

Gessner, M. O. & Chauvet, E., 1993: Ergosterol-to-biomass conversion factors for

aquatic hyphomycetes. – Applied and Environmental Microbiology 59: 502-507.

Gessner, M. O. & Chauvet, E., 1994: Importance of stream microfungi in controlling

breakdown rates of leaf litter. – Ecology 75: 1807-1817.

Gessner, M. O. & Schmitt, A. L., 1996: Use of solid-phase extraction to determine

ergosterol concentrations in plant tissue colonized by fungi. – Applied and

Environmental Microbiology 62: 415-419.

Gonçalves, J. F., França, J. S. & Callisto, M., 2006: Dynamics of allochthonous

organic matter in a tropical Brazilian headstream. – Brazilian Archives of Biology

and Biotechnology 49: 967-973.

Gonçalves, J. F., Graça, M. A. S. & Callisto, M., 2007: Litter decomposition in a


a low density of shredders. – Freshwater Biology 52: 1440-1451.

Graça, M. A. S. & Bärlocher, F., 2005: Radial diffusion assay for tannins. – In: Graça,

M. A. S., Bärlocher, F. & Gessner, M. O. (eds): Methods to study litter

decomposition. – Springer, Dordrecht, pp. 101-105.

Graça, M. A. S., Maltby, S. L. & Calow, P., 1993: Importance of fungi in the diet of

Gammarus pulex and Asellus aquaticus: II. Effects on growth, reproduction and

physiology. – Oecologia 96: 304-309.

Graça, M. A. S., Cressa, C., Gessner, M. O., Feio, M. J., Callies, K. A. & Barrios, C.,

2001: Food quality, feeding preferences, survival and growth of shredders from

temperate and tropical streams. – Freshwater Biology 46: 947-957.

Hagen, E. M., Webster, J. R. & Benfield, E. F., 2006: Are leaf breakdown rates a

useful measure of stream integrity along an agricultural landuse gradient? –

Journal of the North American Benthological Society 25: 330-343.


71

Heard, S. B. & Richardson, J. S., 1995: Shredder-collector facilitation in stream

detrital food webs: is there enough evidence? – Oikos 72: 359-366.

Hieber, M. & Gessner, M. O., 2002: Contribution of stream detritivores, fungi, and

bacteria to leaf breakdown based on biomass estimates. – Ecology 83: 1026-1038.

Hutchings, M. J., Booth, K. D. & Waite, S., 1991: Comparison of survivorship by the

logrank test: criticisms and alternatives. – Ecology 72: 2290-2293.

Irons, J. G., Oswood, M. W. & Bryant, J. P., 1988: Consumption of leaf detritus by a

stream shredder: influence of tree species and nutrient status. – Hydrobiologia 160:

53-61.

Jacobsen, D., Cressa, C., Mathooko, J. M. & Dudgeon, D., 2008: Macroinvertebrates:

composition, life histories and production. – In: Dudgeon, D. (ed): Tropical stream

ecology. – Academic Press, Burlington, pp. 65-105.

Jonsson, M. & Malmqvist, B., 2005: Species richness and composition effects in a

detrital processing chain. – Journal of the North American Benthological Society

24: 798-806.

Kaenel, B. R., Buehrer, H. & Uehlinger, U., 2000: Effects of aquatic plant

management on stream metabolism and oxygen balance in streams. – Freshwater

Biology 45: 85-95.

Kelly, D. W., Dick, J. T. A. & Montgomery, W. I., 2002: The functional role of

Gammarus (Crustacea, Amphipoda): shredders, predators, or both? –

Hydrobiologia 485: 199-203.

Li, A. O. Y. & Dudgeon, D., 2008: The effects of leaf litter characteristics on feeding

and fitness of a tropical stream shredder, Anisocentropus maculatus (Trichoptera:

Calamoceratidae). – Marine and Freshwater Research 59: 897-901.

Li, A. O. Y. & Dudgeon, D., 2009: Shredders: species richness, abundance, and role

in litter breakdown in tropical Hong Kong streams. – Journal of the North

American Benthological Society 28: 167-180.

Moretti, M. S., Gonçalves, J. F., Ligeiro, R. & Callisto, M., 2007: Invertebrates

colonization on native tree leaves in a neotropical stream (Brazil). – International

Review of Hydrobiology 92: 199-210.

Moretti, M. S., Loyola, R. D., Becker, B. & Callisto, M., 2009: Leaf abundance and

phenolic concentrations codetermine the selection of case-building materials by

Phylloicus sp. (Trichoptera, Calamoceratidae). – Hydrobiologia 630: 199-206.


72

Motomori, K., Mitsuhashi, H. & Nakano, S., 2001: Influence of leaf litter quality on

the colonization and consumption of stream invertebrate shredders. – Ecological

Research 16: 173-182.

Nolen, J. & Pearson, R. G., 1993: Factors affecting litter processing by

Anysocentropus kirrhamus (Trichoptera: Calamoceratidae) from an Australian

tropical forest stream. – Freshwater Biology 19: 469-479.

Oliveira, P. S. & Marquis, R. J., 2002: The cerrados of Brazil: Ecology and Natural

History of Neotropical Savanna. – Columbia University Press, New York.

Ostrofsky, M. L., 1997: Relationship between chemical characteristics of autumm-

shed leaves and aquatic processing rates. – Journal of the North American

Benthological Society 16: 750-759.

Parkyn, S. M. & Winterbourn, M. J., 1997: Leaf breakdown and colonization by

invertebrates in a headwater stream: comparisons of native and introduced tree

species. – New Zealand Journal of Marine and Freshwater Research 31: 301-312.

Petersen, R. C. & Cummins, K. W., 1974: Leaf processing in a woodland stream. –

Freshwater Biology 4: 343-368.

Petersen, C. H. & Renauld, P. E., 1989: Analysis of feeding preference experiments. –

Oecologia 80: 82-86.

Prather, A. L., 2003: Revision of the Neotropical caddisfly genus Phylloicus

(Trichoptera: Calamoceratidae). – Zootaxa 275: 1-214.

Rincón, J. & Martínez, I., 2006: Food quality and feeding preferences of Phylloicus

sp. (Trichoptera: Calamoceratidae). – Journal of the North American Benthological

Society 25: 209-215.

Rong. Q., Sridhar, K. R. & Bärlocher, F., 1995: Food selection in three leaf-shredding

stream invertebrates. – Hydrobiologia 316: 173-181.

Royer, T. V. & Minshall, G. W., 2001: Effects of nutrient enrichment and leaf quality

on the breakdown of leaves in a hardwater stream. – Freshwater Biology 46: 603-

610.

Simpson, S. J. & Simpson, C. L., 1990: The mechanisms of nutritional compensation

by phytophagous insects. – In: Bernays, E. A. (ed): Insect-plant interactions.

Volume 2. – CRC Press, Boca Raton, pp. 111-160.

Sponseller, R. A. & Benfield E. F., 2001: Influences of land-use on leaf breakdown in

southern Appalachian headwater streams: a multiple-scale analysis. – Journal of

the North American Benthological Society 20: 44-59.


73

Sugayama, R. L. & Salatino, A., 1995: Influence of leaf epicuticular waxes from

cerrado species on substrate selection by Atta sexdens rubropilosa. – Entomologia

Experimentalis et Applicata 74: 63-69.

Stachowicz, J. J. & Hay, M. E., 1996: Facultative mutualism between an herbivorous

crab and a coralline alga: advantages of noxious seaweeds. – Oecologia 105: 377-

387.

Swan, C. M. & Palmer, M. A., 2006: Composition of speciose leaf litter alters stream

detritivore growth, feeding activity and leaf breakdown. – Oecologia 147: 469-478.

Wallace, J. B. & Webster, J. R., 1996: The role of macroinvertebrates in stream

ecosystem function. – Annual Review of Entomology 41: 115-139.

Wantzen, K. M. & Wagner, R., 2006: Detritus processing by invertebrate shredders: a

neotropical-temperate comparison. – Journal of the North American Benthological

Society 25: 216-232.

Webster, J. R. & Benfield, E. F., 1986: Vascular plant breakdown in freshwater

ecosystems. – Annual Review of Ecology and Systematics 17: 567-594.

Welker, B. J., König, W., Pietsch, M. & Adams, R. P., 2007: Feeding selectivity by

mantled howler monkeys (Alouatta palliata) in relation to leaf secondary

chemistry in Hymenaea courbaril. – Journal of Chemical Ecology 33: 1186-1196.

Yeates, L. V. & Barmuta, L. A., 1999: The effects of willow and eucalypt leaves on

feeding preference and growth of some Australian aquatic macroinvertebrates. –

Australian Journal of Ecology 24: 593-598.

Zar, J. H., 1999: Biostatistical Analysis. – Prentice-Hall, New Jersey.

74

Conclusões

1. As larvas de Phylloicus apresentaram ampla plasticidade no comportamento

de construção de casulos e sua preferência foi determinada pela composição

química e disponibilidade das folhas.

2. As relações comprimento-biomassa descritas para esta população de

Phylloicus foram significativas e podem ser utilizadas em experimentos de

laboratório que avaliem, por exemplo, as taxas de crescimento e a produção

secundária destes fragmentadores.

3. O comportamento alimentar de Phylloicus foi mais influenciado pela

qualidade das folhas disponíveis do que sua origem (nativas/exóticas).

4. As folhas de Eucalyptus não constituíram um recurso alimentar de maior

valor nutricional para as larvas de Phylloicus do que as folhas de espécies do

Cerrado.

5. Phylloicus e Gammarus apresentaram a mesma preferência alimentar quando

expostos a detritos de baixa qualidade nutricional e o comportamento

alimentar de ambos fragmentadores foi influenciado principalmente pela

concentração de lignina dos detritos foliares.

75

ConsideraçõesFinaisePerspectivasFuturas

Os resultados aqui apresentados revelaram que o comportamento das larvas de

Phylloicus é determinado principalmente pelas características químicas, dureza e

disponibilidade dos detritos foliares. Além disso, a baixa qualidade nutricional e

palatabilidade das folhas normalmente encontradas nos córregos do Cerrado brasileiro

podem ser responsáveis pela menor abundância de organismos fragmentadores nestes

ambientes. No entanto, o comportamento alimentar exibido por Phylloicus foi

semelhante ao exibido por Gammarus, corroborando estudos anteriores e reforçando a

idéia de que os padrões observados para o comportamento alimentar de

fragmentadores de região temperada também podem ser aplicados aos equivalentes

tropicais.

Dentre as características das folhas do Cerrado, a dureza foliar tem um papel

fundamental na atividade dos fragmentadores e, conseqüentemente, nas taxas de

decomposição de detritos foliares. Os resultados obtidos mostraram que os teores de

lignina foram a principal característica que afetou as taxas de consumo exibidas pelos

fragmentadores. Desta forma, para o maior entendimento dos padrões que

determinam o processamento de detritos foliares em córregos do Cerrado brasileiro e

de outras regiões tropicais com características semelhantes, recomenda-se que os

seguintes tópicos sejam avaliados em experimentos futuros:

1. As taxas de decomposição e o comportamento alimentar de invertebrados

fragmentadores expostos a detritos foliares que apresentem um amplo

gradiente de dureza.

2. As taxas de assimilação e respiração das larvas de Phylloicus, bem como seu o

crescimento e sobrevivência durante maiores períodos de exposição.

3. As possíveis alterações no comportamento de Phylloicus na presença de outros

fragmentadores tropicais e organismos de outros grupos trópicos funcionais

como, por exemplo, predadores.

4. A dinâmica populacional e o ciclo de vida das larvas de Phylloicus.

5. A influência das assembléias de decompositores, microrganismos e

invertebrados fragmentadores, no processamento de detritos foliares em

córregos tropicais de diferentes regiões e tipos vegetacionais.

76

Por fim, as informações levantadas pelos experimentos realizados com

Phylloicus e a elevada riqueza de espécies presentes na vegetação ripária de córregos

de cabeceira tropicais sugerem que a abundância e a importância dos organismos

fragmentadores no processamento de detritos foliares nestes ambientes estão mais

relacionadas às características locais que determinam a composição de espécies

arbóreas das zonas ripárias (p.ex., altitude, tipo de vegetação e clima) do que a

padrões latitudinais generalizados, contrariando o que vem sendo sugerido nas últimas

décadas pela literatura internacional.

77

Referências

Cheshire K., Boyero B. & Pearson, R.G. (2005) Food webs in tropical Australian

streams: shredders are not scarce. Freshwater Biology, 50, 748-769.

Copasa-MG, Companhia de Saneamento de Minas Gerais. Website:

www.copasa.com.br

Dobson M., Magana A., Mathooko J.M. & Ndegwa F.K. (2002) Detritivores in

Kenyan highland streams: more evidence for the paucity of shredders in the

tropics? Freshwater Biology, 47, 909-919.

Gonçalves, J.F., Graça M.A.S. & Callisto M. (2007) Litter decomposition in a


a low density of shredders. Freshwater Biology, 52, 1440-1451.

Irons J. G., Oswood M. W., Stout J. R. & Pringle C. M. (1994) Latitudinal patterns in

leaf litter breakdown: is temperature really important? Freshwater Biology, 32,

401-411.

Jacobsen, D., Cressa C., Mathooko J. M. & Dudgeon, D. (2008) Macroinvertebrates:

composition, life histories and production. In: Dudgeon D. (ed): Tropical stream

ecology, pp. 65-105. Academic Press, Burlington.

Kaenel B. R., Buehrer H. & Uehlinger U. (2000) Effects of aquatic plant management

on stream metabolism and oxygen balance in streams. Freshwater Biology, 45, 85-

95.

Li A. O. Y. & Dudgeon D. (2009) Shredders: species richness, abundance, and role in

litter breakdown in tropical Hong Kong streams. Journal of the North American


Mathuriau C. & Chauvet E. (2002) Breakdown of leaf litter in a neotropical stream.

Journal of the North American Benthological Society, 21, 384-396.

Moretti M.S., Gonçalves J.F., Ligeiro R. & Callisto M. (2007) Invertebrates


Review of Hydrobiology, 92, 199-210.

Wantzen K. M. & Wagner R. (2006) Detritus processing by invertebrate shredders: a

neotropical-temperate comparison. Journal of the North American Benthological

Society, 25, 216-232.

Documents

Capa Tese (com foto)