This is a deconstructive reading of scientific discourse, undertaken through a particular scientific publication titled, "Arthropod 'Rain' into Tropical Streams: the Importance of Intact Riparian Forest and Influences on Fish Diet" (2008, 59, 653-660), which exhibits all the standard features of the genre. One of these features we call, IMRAD, acronym for introduction, methods, results, and discussion. Arguably, IMRAD is a narrative device which inscribes phenomena as a reading. My paper, therefore, may be described as a theorization of this most iconic of scientific genres, the research article. I demonstrate how scientific writing, notwithstanding its outward imperviousness to notions such as textuality, closure, and différance, is still susceptible to them. This is because, in the end, scientific discourse is still writing, that spatio-temporally differed structure Derrida speaks of. Scientific writing cannot escape the constitutive metaphoricity of writing. One particular trope I examine thematically is scope. I argue further that in the research article what we encounter is not facts but their representations. Writing functions as a phenomenology. The theoretical tools I mobilize for my reading are organized broadly around Derrida, Knorr-Cetina, Greg Meyers, Latour and Woolgar and so forth. We have an imperative to understand scientific writing, which is a particular form of that much larger formation, academic writing. Most of our attempts are narrowly pragmatic, not 'literary.' The inter-disciplinary nature of the paper embraces a number of key areas of the conference: literature, literary studies; language, linguistics; teaching and learning; science, environment and the humanities. [ABSTRACT FROM AUTHOR]