No abstract

Two cautionary points, one perhaps encouraging and another less so, emerge from our study. First, it is evident that a high mutation rate is not always bene®cial. The adaptive neighbourhood of an adaptive peak can constrain evolvability, despite a high mutation rate. A high rate can lead to a high mutational load, as seen during the evolution by clone B. Thus, evolutionary changes in a population of RNA viruses, whether under clinical treatment or otherwise, should not automatically be interpreted as adaptive to the virus. Second, as had already been noted 12 , if multiple peaks exist, a shift to a new adaptive peak can be triggered by a costly resistance mutation and the ensuing compensatory mutations. If some peaks are higher, drugs should be used with caution not only because resistance can evolve but also because drug treatment can lead to the evolution of viruses that are more ®t rather than less. M Methods Strains and culture conditionsThe RNA bacteriophage f6 used in this study is a laboratory clone descended from the original isolate 15 . Populations A and B were generated in a previous study 7 . They are the end points of populations propagated using bottlenecks of 100 and 33 phage, respectively, to allow ®tness recovery after acquisition of a common deleterious mutation. Pseudomonas syringae sv. phaseolicola, the standard host of f6, was obtained from the American Type Culture Collection (ATCC# 21781), and Pseudomonas pseudocaligenes ERA, an alternate host, was obtained from L. Mindich. Details of diluting, ®ltering, culture and storage of phage and bacteria are published 16,17 . All phage and bacteria were grown in LC medium at 25 8C.

ABSTRACT. COMP()~rrJON ANil IlYNAMIC~OI' Til!' HIRIl COMMllNflY OI' MATA DF SANTA GENEBRA. CAMPINA~. SÃO PAULO. BRAZII .. A quali-quantitatiw survey 01' avi fauna was earried out at the Sanla Genehra forest reserve (MSG) in order to evaluate lhe <:urrent stah~oI' çonservation ar this ilnportant conservation arca in lhe interior 01' São Paulo State. The qualitative survey reeorueu 134 species and lhe quantilative survey through unlimited uistanee point counls reeonled lhe ahundanee 01' 97 speeies. The low numhcr oI' speeies reeordeu in bolh surveys was due to the small size (151 ha) anu lhe high degree oI' isolation oI' the MSG. These faelors plus as resulting from the density compensalion phenomenon. The struclure oI' the bird communily 01' MSG resembles Ihat from an island with low number 01' speeies hul overall uensity similar lO thal ofthe mainland. Icauing to a higher uensily per speeies. The proeess responsible fi", these patterns seems to be relateu to the habilat approprialeness. mainly beeause lhe MSG is eomplctcly isolatcu from olher foresl ar~as .

No abstract

How information transmission processes between individuals are shaped by natural selection is a key question for the understanding of the evolution of acoustic communication systems. Environmental acoustics predict that signal structure will differ depending on general features of the habitat. Social features, like individual spacing and mating behavior, may also be important for the design of communication. Here we present the first experimental study investigating how a tropical rainforest bird, the white-browed warbler Basileuterus leucoblepharus, extracts various information from a received song: species-specific identity, individual identity and location of the sender. Species-specific information is encoded in a resistant acoustic feature and is thus a public signal helping males to reach a wide audience. Conversely, individual identity is supported by song features susceptible to propagation: this private signal is reserved for neighbors. Finally, the receivers can locate the singers by using propagation-induced song modifications. Thus, this communication system is well matched to the acoustic constraints of the rain forest and to the ecological requirements of the species. Our results emphasize that, in a constraining acoustic environment, the efficiency of a sound communication system results from a coding/decoding process particularly well tuned to the acoustic properties of this environment.