Research conducted in remote settings away from the formalized experimental settings found in universities is often a mix of science research and survival and is comprised of complexities that extend far beyond the actual conduct of the scientific research. For instance, an ecologist conducting field research needs to realize that encounters with bears may occur, during which they are within only a few meters. Yet, presentations of the research in the formal texts of lectures, textbooks, and journal articles used to enculturate newcomers to ecology rarely detail the experiences of the field researchers in an explicit fashion that can prepare newcomers; nor do they detail the methods used to collect the data in the study to a degree that they could be applied in any simple manner (Roth & Bowen, 2001b). This raises the question, "Where do beginning field ecologists (undergraduates or new graduate students) learn about the conduct of field research and how researchers feel during that work?"

Actual or vicarious field experience also affects ones perception of field settings. For instance, how a videotape of a bear walking in a forest is perceived depends on whether the viewer has ever seen a bear move through underbrush before. The videotape segment underlying the field note above showed a bear moving slowly, seemingly with studied deliberation, as it walked along the fallen, moss-covered trees in a British Columbian rainforest. Yet, how one "sees" that video depends on the understanding one has both of bears and of the setting in which the bear was seen; professional vision is not given but develops in communities of practice (Goodwin, 1994). Before seeing a black bear for the first time (while we were conducting ethnographic research with ecologists) we assumed a human could outrun a small black bear. However, after working as field assistants with ecologists in mountain underbrush, and after having observed bears moving through that underbrush, the videotape had different meaning--we now know that if the bear in that video had begun an attack against the researcher making the videotape, who was less than six meters away from the bear, then there was little the researcher could have been done to avoid that attack. Our framework for interpreting the video of the bear shifted as a result of our experiences in the field with bears.

It is clearly not possible for all field researchers to be able to experience all such possible encounters for a field sense to develop which would allow successful field research to be conducted--the various settings from which one could learn about successfully conducting field research (either methodology or survival) must also play a significant role in communicating this information. For us, this raises the question, "By what means and where is critical information about field practices relayed in the community of ecologists?"

From our research among ecologists, we present evidence that the formal settings of classrooms--lectures, journal articles, textbook writings--convey little of the information that is necessary for ecologists (especially undergraduate and graduate students learning to become ecologists) to effectively do field research and that other resources are critical to enable a future ecologist to do this type of work. This paper is about the appropriation of knowledge, information, in formal and informal settings by ecology students (i.e., individuals in formation). We address the questions, "How are formal texts (both written and spoken) and informal texts (such as in bars, over coffee, and during informal public talks) in which ecologists discuss their work structured?" and "What are the implications of any similarities and differences for the enculturation of newcomers into the concerns, practices, and discourse(s) of ecology?"

There are few ethnographic studies that examine the enculturation of new researchers into scientific research practices (Roth & Bowen, 1999b, 2001a). Some of the existing work comes from the study of physics students. One study of the formal texts in undergraduate physics detailed how students engaged in group problem solving activities working on theoretical and abstract problems (Nespor, 1994). During these activities, students develop an identity as part of a "physics actors network" within which they learned to explain physical phenomena primarily in mathematical terms. The textbooks are "useful in holding networks together in a stable configuration and keeping activity ‘on-track’" (p. 59) so that students become enculturated to the standardized practices and interpretations of solving the mathematical problems. These mathematical problems are presented as if they were physics, not as mere representations of phenomena (which mathematics approximates). In addition, the lectures students attend differ little from the textbooks. Overall, the descriptions of the programs suggest that undergraduate physics students learn little about doing physics research but learn to use mathematical formulae. Even at the graduate student level, formal physics texts differ little from those of the undergraduate physics students (Traweek, 1988). Both graduate and undergraduate students study the heroes of the discipline and learn how to succeed by examining how those heroes were successful. Graduate students they learn a little about style of "doing" physics by listening to stories of success and failure. Both undergraduate and graduate physics students are enculturated into thinking of physicists (and then themselves, if successful) as people who are specially gifted and above the common milieu.

An examination of an undergraduate program in environmental biology (EB) led to the conclusion that the "EB view of a scientist is a kind of challenge to the hegemony of the theoretical, laboratory, or research scientist who is widely celebrated in the physical sciences" (Eisenhart, 1996, p. 175). Employers of graduates from this program are dissatisfied with their preparation to deal with "real-world issues" (p. 180). Students were well schooled in the concerns of environmental biology but not skilled in balancing the concerns of different communities (i.e., business, political, etc.) or in applying them at their future job site.

In summary, research on enculturation therefore suggests that undergraduate science students are enculturated into the concerns, practices, and claims of their respective disciplines but little into the respective research practices. An important aspect of scientific research is the construction and interpretation of inscriptions (Latour, 1993)--particularly tables and graphs--which are central to the claims derived from research.

We each participated in extended ethnography studies among ecologists over several years. During this time we attended most settings common to ecologists in both their field and laboratory settings, interacting with them both in their work related settings and at their social gatherings. To address our research questions we analyze texts from a multitude of different settings that were both formal and informal. We considered formal texts such as those that were structured for a particular professional audience (in the discipline). Thus, we conducted a fourteen-week ethnographic study in a second year ecology class, attended university seminars, participating in university fieldtrips (for both graduate and undergraduate students), and attended conferences and symposia. We also considered textbooks and journal articles as formal texts. In addition, we conducted over fifty interviews with various members of science communities (ranging from undergraduates to senior professors) around and about their interpretations of graphical inscriptions. In most of these settings we documented practices on videotape (occasionally only audio recordings were possible), collected samples of written artifacts, and recorded impressionistic ethnographic fieldnotes.

We also participated in countless "informal" gatherings of ecologists where they discussed their work practices. We considered informal settings or texts as those which included the characteristics of being unstructured, included or were directed towards non-professional participants (i.e., those not a formal trajectory to become a scientists), were held in non-academic settings, and which may have been spontaneous or serendipitous (such as accidentally meeting others in the community of ecologists in a coffee shop). These informal settings included after-work gatherings at someone’s house, in coffee shops, bars, pool halls, hallways in the university, and so forth. We both observed and participated in social interactions ("chatting") in all of these settings. The informal texts included television interviews and newspaper articles. The nature of these settings is such that video or audio recording was often impossible, so the main data resources for these settings are impressionistic ethnographic fieldnotes (van Maanen 1988) and artifacts from television and newspaper articles.

Much of our data was collected as we worked as field assistants with ecologists over several years (during which we examined both their field research practices, their construction of scientific claims, and the enculturation of newcomers into these practices). In so doing, we were following in the tradition of conducting ethnography through apprenticing in the discipline (Coy, 1989). In addition, we contextualize our analysis for this paper with our past work with students with and without B.Sc. degrees as they interpreted inscriptions and conducted field research projects of their own (Bowen, Roth, & McGinn, 1999; Roth & Bowen, 1999a, 2001a) and with eighth-grade students as they conducted ecology field research (Roth, 1996; Roth & Bowen, 1993, 1994, 1995).

Prolonged and intensive engagement in the field provided the bases for thick descriptions, a foundation for our understandings (Guba & Lincoln, 1989). Ongoing analysis of the field data was conducted to help establish the "credibility" of claims from our ethnographic and qualitative research--a parallel to internal validity (Guba & Lincoln, 1989). Two of the criteria for establishing credibility are peer debriefing and member checking and these both occurred as a consequence of the ongoing analysis of data as the various studies progressed. During the field sessions, active analysis of the collected data was shared between the authors so that interpretations and observations could be critiqued and further questions asked. In addition, analyses could then be checked with the member ecologists or un/substantiated with further observations.

Our analyses of the interviews/transcripts and other texts were informed by the method of Interaction Analysis (Jordan & Henderson, 1995). This method involves both individual and collective interpretation sessions. First, we individually conducted an interpretive analysis of the audio, video and textual resources (including transcripts) by reading them over and identifying both broad themes of interest and specific critical instances of discourse or text. These themes were used to frame assertions, which were subsequently evaluated in the light of supportive or contradicting evidence. We then met collectively to examine our individual assertions and negotiated differences in our interpretations. From this process we developed new assertions or reformulated earlier ones. Through repeated cycles of interacting with each other over our individual assertions and independent analysis, we formulated collective claims from the dataset and established supporting episodes for those claims. This sharing and critiquing assisted in the process of progressive subjectivity (Guba & Lincoln, 1989) and helped guard against developing nonviable interpretations.

In addition to the understanding of the difficulties conducting field research, which emerged in the informal settings, we developed an understanding of the difficulties encountered by new researchers (such as new M.Sc. students) during their first research projects in two other ways. First, we review a study of the process and products of recent biology B.Sc. graduates as they engaged in an independent research project in a professional program. Secondly, we conducted several semi-structured interviews (a short list of pre-determined questions was elaborated in reaction to issues discussed by the interviewee) with "new" graduate students who had engaged in a recent study of their own. Collectively, the data from these many different settings allowed us to develop a better understanding of how and where new researchers in ecology learn about the conduct of field research.

We begin first by reviewing our related studies that render as problematic the interpretations of graphical inscriptions by graduates from B.Sc. biology programs. We then review the research practices enacted by science graduates when participating in their own short-term ecology field studies. To make sense of these studies, we examine the experiences of undergraduate biology students in the formal texts from which they learn disciplinary practices and then analyze the content of the informal settings where graduate students and some undergraduates frequent.

Competencies in Graph Interpretation and Field Research

An understanding of what science practice competencies accrue to students during their undergraduate study provides a foundation to discuss the sources of ecologists’ field research competencies. Our work with graduates of various undergraduate science programs provides insight into the development of competency with inscription use and the conduct of field research projects. We summarize the inscription practices of these students (Bowen & Roth, 1998; Roth & Bowen, 1999a) and then sketch the competencies of these individuals as they pertain to conducting small field research projects.

When asked to interpret a graph, individuals with science degrees (having completed B.Sc. degrees) frequently have difficulty in elaborating an interpretation; these difficulties also exist when they work in groups. Interpretations are often referentially isolated--that is, they are concerned with the form of the graphs rather than with the natural phenomena that graphs might refer to. The individuals get "stuck" within the details of the signs themselves and do not draw on their knowledge of the world to help them read the graphs. This often results in breakdown of the interpretive process; individuals and groups are unable to proceed in their interpretations of some graph. For example, these students used only a small number of resources, such as references to natural populations or mathematical tools in their interpretations of a population graph. In addition, the linguistic resources brought to the graph interpretation task made it difficult for them to make important distinctions between different concepts necessary to develop any interpretation appropriate to the academic field. The resultant ambiguities made arriving at a shared interpretation during group work difficult. In general, students learn to provide correct answers to specific graph-related questions but do not come to make linguistic distinctions, increase their knowledge of specific populations, and do not develop general interpretive skills (Bowen, Roth, & McGinn, 1999). Rather, students learn to apply the interpretation of specific graphs as provided in lectures and seminars.

Science graduates rarely enact canonical research practices or interpretation of data sets. For example, students in a post-baccalaureate program conducted a small field research project in ecology for which they were asked to examine correlations between one biotic feature (such as density of a type of plant) and two abiotic features (such as soil moisture and temperature). Students conducted the study in groups of two or three and then wrote group reports on their findings. Although their reports contained most of the components of scientific reports, there were problems with most regarding the content of each component. Generally, students planned research questions unanswerable by the study design, inappropriately operationalized the constructs to be used, reported and transformed data using inappropriate representations, and failed to match research claims and research questions to data.

In total, there were 12 reports containing 24 research questions. In many cases the investigations were framed as causal investigations (N = 14) that were not possible to conduct given the parameters of the project. For example, questions included such topics as "How does moisture effect horsetail height?" "How does moisture affect growth?" and "How does pollution from cars affect plant productivity?" In seven reports, variables were inappropriately operationalized, replication was problematic or sampling was done such that it was not possible to answer their research questions or to draw conclusions from their data. The subsequent reports reflected some of their difficulty with the conduct of the field research. Tables (N = 14) were used to summarize data but several were structured in non-standard ways and did not aid in understanding any patterns. Ten reports used graphical inscriptions but in such a way that these negatively affected interpretation of the data. They used inappropriate graphs when another type might have better portrayed the relationship (e.g., line graph instead of a bar graph or vice versa), infrequently fitted lines-of-best-fit, and conducted no outlier analysis. In addition, there were often structural problems that confounded interpretation. With this work we gained insight into students’ difficulties interpreting inscriptions. Given that their experience with graphs was such that they could not effectively use them to understand their own research experiences; it was perhaps unreasonable to expect them to be able to re-construct the experiences of others when interpreting inscriptions.

It is possible that these science graduates are unrepresentative of those who go on to actually do field research, however this is not supported by our observations in the community. For instance, it is common lore amongst biology professors that even those graduate students who were top undergraduate students initially have few competencies at conducting field ecology or laboratory research. As was clear from our informants, a notable consequence of this for those who engage in field research, most particularly those who engage in field research during which one is working alone, is that little useful data is collected in the first field season. Graduate students commonly indicate that their first field season did not result in usable data because of the errors they committed due to their lack of knowledge and experience.

Students’ interpretations of graphs contrasts those of experienced field researchers reading graphs related to their work. Experienced researchers engage in a dialectic process in which different kinds of experiences and understanding mutually constitute one another. These include their own research experiences, commonly held ecological knowledge touchstones, examples of live populations, and anecdotal narratives related to them by others. Frequently, we observed a combination of these resources. For example, Jan (a post-doctoral theoretical ecologist) drew together three of these features as he made sense of a graph depicting a population of organisms with density dependent features affecting the birth rate but not the death rate in the population.

When scientists develop interpretations of graphs, they use a combination of resources; as a result, they make sense of individual graph components and the relations between variables depicted. Stories, such as the rabbit story above, are accumulated in a variety of settings and frequently play an important role in the interpretations of experienced researchers.

Given that graduates from baccalaureate science programs experience such difficulties when it comes to field research and graph interpretations, we have to ask "How and at what point do scientists learn to do research and interpret graphs?" Since the common experience shared by undergraduate students and beginning graduate students in biology are the formal texts of science—lectures, textbooks, laboratory activities, and journal articles¾ an examination of those should help us understand the beginnings of ecology researchers.

The formal information sources available to undergraduates present knowledge about ecology in an impersonal style with infrequent reference to the scientific practices and field experiences of the scientific authors; reports thereby constitute "a world from which persons are virtually excluded" (Gross, 1996, p. 70; Roth & Bowen, 2001b). This information is rarely presented against the background of the social contexts in which researchers work, a context critical to understanding the different perspectives and interpretive stances that ground the claims. Our examination of ecology textbooks and journal articles (Bowen & Roth, in press; Roth, Bowen, & McGinn, 1999) suggest that they rarely include any detailed references to the field research and methods. When researchers in formation engage in their first projects, they therefore have limited resources on which to draw for judging the quality of their work. This is especially problematic in in the field science rather than the laboratory including ecology, archaeology, and geology (Delamont, Atkinson, & Parry, 2000). Newcomers to laboratory sciences such as chemistry, physics, or genetics find themselves in settings rich in informational resources¾ post-doctoral students, doctoral students, and professors work in the same locale as the new researchers and academic paper resources are readily available. Field research is often conducted in far-flung settings far removed from other individuals and formal academic resources. In some settings, such as field stations, this is less problematic as there are often others (Nutch, 1996); in many other settings there is substantial isolation and little academic or technical support (Roth & Bowen, 2001b). Thus, researchers in formation often experience considerable frustration when they start their research careers. It is latest at this point that the problematic of formal education become salient to the newcomers.

In this section, we provide an analysis of information sources encountered by students during their formal education. During their time at the university, undergraduate and graduate students obtain access to information about science from four basic, formal sources: lectures, textbooks, journal articles, and formal verbal presentations (at conferences, symposia, etc.). In the following sections we describe these settings and what informational resources for a new field researcher (in formation) are present in each.

Textbooks and lectures are highly similar in that they present a purely factual view of ecology, a view from which human agency has largely been eliminated (Bowen, Roth, & McGinn, 1999; Nespor, 1994; Roth, Bowen, & McGinn, 1999). Both of these information sources provide those in formation with a limited sense of ecology. The information presented is in the form of broad conceptual knowledge claims that constitute compilations of individual research projects. The information is presented as if there was complete agreement amongst scientists within the discipline as to the interpretation of the data; the sources of the information are rarely discussed. Pieces of information are matter-of-fact as if they existed independently of their own constitutive historical foundations. This ahistorical nature also extends to the variables presented conferring to them an a priori nature independent of the knowing subject.

This ahistorical way of teaching ecology was also present in practical activities for undergraduate ecology students. Our observations of a field research exercise for a second-year ecology course suggests that students experience these activities as disconnected from the lecture component of their course and the research methods as inviolable, pre-determined, and standardized. The following field note provides a sense of the experience of participating in such activities.

The students in this field note conducted a cookbook exercise for which they attempted to follow step-by-step instructions. Each group of students only participated in a part of what was described to constitute a whole research activity. Each group collected their samples from one spot along the transect line and only later, after sharing of data amongst groups, wrote their reports in which they examined the data for patterns. The students who participated in this activity experienced ecological field methods as a set of standardized procedures predetermined by an external authority that made all relevant decisions beforehand. Outdoor research exercises such as this one are treated by instructors as activities that are separate from both the lecture material and the exercises in the laboratory. This made it difficult for students to identify the organisms because they had not previously done any identification activities. They not only had to learn new field methods but also the organisms by relying on field identification sheets that they had not seen before; this identification is in itself a difficult task (Law & Lynch, 1990). We noted in our participation that the second-year students were not particularly motivated to accurately identify the organisms in the quadrate and were often satisfied with settling on their first identification without interrogating its accuracy and thus numerous mistakes were made in identification. This was not surprising given the paucity of resources on which they could rely. Even when they realized that errors had occurred (which happened once) corrections in the counts recorded on the table were not made. The students’ most immediate goal was to have the data sheet completed to the satisfaction of the TA so they could later write a laboratory report based on the data¾ their orientation to the task was notably as students, not as researchers with a vested interest in defensible data recording. Clearly, the point of the exercise was that student understanding of the distribution of organisms was to develop from the analysis of the entire classes’ data set, not that this analysis and which variables one should attend to should emerge from observations arising in the field.

The physical discomfort often experienced by field researchers (Roth & Bowen, 2001b) was the one aspect of field research students could in fact experience. However, this "real" aspect of field research was rather under-appreciated by the students who attributed the discomfort to poor planning on the part of the instructors. The poor weather conditions under which the data was collected probably contributed to the reduced motivation for accuracy in identifying species and enumerating them in their quadrat.

University students at the senior undergraduate and graduate levels also experience formal contexts as they read journal articles and attend conferences and symposia. Unlike textbooks and lectures, which portray the knowledge foundations of ecology in broad strokes, these resources focus on reports of individual research projects and how they contribute to the overall understanding of ecology. Many authors (e.g., Gross, 1996) note that the formal writings of scientists dramatically underdetermine the activities in which they engage. They thereby portray a depersonalized image of ecology that reaffirms the objectivity of their work. Here, we conduct an analysis of parts of the "Methods" sections of two journal articles written by an author to whose informal accounts about his research we later return. In the course of his work on a particular fish, this ecologist made observations on bears and their feeding on salmon. We determine what cues these articles offer to researchers in formation to inform current or future fieldwork. We later contrast the information provided in this formal context with that provided in informal settings such as presentations to lay audiences or with graduate students in a bar ("B(e)ar Stories").

The first journal article (in a major journal in the discipline) presents aspects of bear genetics and its implications that had for discrete populations and population distribution. For the present analysis we focus on the part of the methods section dealing with how the "samples" were obtained:

This text describes the body part from which the tissue samples were obtained. A table lists fifteen geographic locations as sources of the samples, how many samples were collected at each site, and who provided the samples (mostly the authors themselves). Thus, sampling was discussed in two forms: location of the tissue samples on and source location of each bear. From the information in the table, a reader might also infer muscle tissue as the preferential tissue for conducting the DNA sequencing technique used and the geographic range and an adequate sample sizes for such a study. The remainder of the "methods" section provides considerable detail of the genetic sequencing activity, although a graduate student interested in conducting such research would find few resources detailing how such samples are actually collected from a bear.

The second journal article makes reference to the first paper and deals with patterns in foraging behavior of bears. It offers considerable detail about doing the research including the geographic location, the number of salmon returning, dates of salmon movements, periods of observation, and number of visual surveys of bears. The variables examined were presented as if they were predetermined before the fieldwork was conducted. A matter-of-fact neutral tone, typical in science writings (Gross, 1996), is used throughout when describing the actions of the researcher, even when describing situations such as following a bear at a distance of two meters. In keeping with our findings of inscription use in ecology journals (Roth, Bowen, & McGinn, 1999) these articles provided detailed captions and explicit readings of the inscriptions so that the readers were channeled into the interpretation of the graphs desired by the writer. So, these articles are structured so that the claims appear unassailable, however they offer few resources to new researchers desiring to conduct such research themselves.

We observed and recorded two formal presentations about this research with bears, one presented by the principal researcher and one by a graduate student, at two different symposia 16 months apart. To be considered "formal presentations" they met the following criteria: a list of speakers with presentation times was printed and distributed to the attendees, there was a common theme to the series of talks being given (e.g., "Nth Annual Vertebrate Symposium"), presentations had a "chair" who introduced and closed each session, presentations conformed to a standardized format (i.e., 15 minutes with slides/overheads, questions from audience) and reported on research. The presentations typically often involved preliminary results of research that is still active. The two presentations we discuss were typical of those we saw at over eight such gatherings we attended.

Both presentations were similar in structure providing an initial background to the research by showing color images with a commentary about what was being seen. These included images of the watershed, a map, the salmon species, the bears, a carcass being consumed or left behind by a bear, and so on. The images of the research site and organisms were followed with a presentation of various inscriptions used to represent the findings. Often detailed readings of the inscriptions with accompanying hand gestures to emphasize the important components were provided. In addition, graphs and tables were presented in clusters in support of the claims presented.

The text of these talks mostly dealt with constructing a persuasive rhetorical argument so that the knowledge claims at the end appeared unassailable. This included the amount of time spent doing the study, the variables chosen for examination and the rationale for their being chosen (from which methodology could sometimes be inferred), the amount of data collected, and the juxtaposition of text and gestures that accompanied the inscriptions. Physical objects (and organisms) constituted privileged information and the human agency during data collecting was downplayed. The following segment of a presentation by the professor illustrates these points:

This description contains information about bear activity and what the data collected, but there is little from which a researcher in formation could gather about how that data was obtained. The ordering of the information, undoubtedly done for rhetorical purposes, pre-supposes information obtained by doing the study in two ways. For instance, the statement that the bear consumes 50% of each carcass requires that the initial fish weight¾ before the bear consumed it¾ was either known or determined, and for a new researcher how this was determined in this study is important information. Additionally, that information is described prior to the study itself, even though it was determined by doing the study. Generally, formal presentations focused on where the research was conducted, what organism(s) the research was conducted on, and what findings were made. Little information was available about the methodology so that these formal presentations bore a strong resemblance to textbooks and lectures. In essence, formal talks present information such that those in formation can gather little about the process of research.

What are the consequences for new researchers of the enculturation to ecology and field research in the manner described above? It is not surprising that our graduate student informants complained about the lack of access to information that leads to undeveloped and underdeveloped understandings regarding the contingencies that mediate field research. One of our informants (a doctoral student) suggested that the main reason for quitting graduate work were the unpredictable factors that impeded with the research. Researchers in formation often experience such difficulties as unique to their situation and attribute the blame to themselves because the formal sources never presented such problems.

In addition, researchers in formation rarely know about the isolation and independence that are also features of fieldwork. Graduate students who had acquired their first field experiences in the past season discussed their isolation in the field, the effect it had on their work, and how they wished to a different supervisor.

This isolation and the lack of input regarding field research methods meant that new researchers were insecure about the quality of their work. These concerns about the quality of data, and the lack of any reference to compare their work to which would allow them to receive affirmation for the quality of work, was a frequent stress factor for researchers in formation.

Researchers in formation often recognize in hindsight that their undergraduate experiences poorly reflected their newly gained experiences conducting research. Donna had completed one field season and had considered dropping out of her program. Her reflections are indicative of her discontent with the realities of conducting field research:

It was the contrast between science as practiced and that portrayed in her undergraduate lectures that caused Donna considerable stress. She wanted to conduct field research that was similar to the science with which she was familiar from her undergraduate education. The uncertainties of field research¾ what should be measured, what variables are important, the dealing with missing data caused by bad weather¾ she found hard to cope with. Further, her research meant spending weeks in the field collecting samples in remote areas so she had no peers to rely on, no email to use to ask questions of her supervisor, no feedback on whether the decisions she was making about research were the right decisions--she considered field methodologies as subject to externally determined criteria of correct or incorrect. Left to her own, she struggled with every situation "guessing this or that was the right thing to do." Donna considered learning to do research as a process of trial and error and was discontent with "not knowing if a decision was right." She felt that her methods would be judged external to her data, her arguments, and the context in which they were applied. If she had known what field research was like, she would have never started a graduate degree in ecology.

Clearly, the formal structures of undergraduate and graduate education in ecology poorly prepare students to do field research. This raises the question, if field research and field methods are communicated so little in the formal settings of ecology, how do new researchers develop competence and confidence in conducting field research? Here, information exchanged between researchers in informal settings provides substantially more to researchers in formation than the formal settings. The paucity of stories of field experiences in the formal texts of ecology, the important role they play in interpretations of conceptual claims, and their potential role in relating field research experiences suggests an analysis of informal settings in which ecologists gather may offer some insights.

During our two-year ethnography we spent considerable time interacting with and observing ecologists in informal settings. These settings included coffee shops, pool halls, bars; cars (as we traveled to conferences or field research sites), informal events at formal conferences, and private homes of a participant. Formal sessions such as symposia were frequently followed by adjournment to a local pub for informal meetings. Finally, we considered presentations to public groups as informal settings (corresponding to our criteria for "formal" settings).

Our observations in the informal settings suggest that different types of interactions occur at different times and places. This is often related to the amount of research experience the ecologists have. Thus, there are more conversations about field research in groups that included field researchers with PhDs in their composition than in groups composed of just M.Sc. students. In the presence of doctoral students but absence of Ph.D. researchers, fewer stories about field experiences are exchanged. In groups composed of M.Sc. students only, fieldwork stories are even less frequent. It is notable that in conversations about research work new M.Sc. students (even those who had participated in numerous research projects as field assistants) often do not participate in conversations about fieldwork. As time passes, even without having gained new field experiences, these students begin to participate to an increasing degree up to the "old-timer" level.

When someone joins a group of ecologists, the ensuing conversations are generally about field experiences if the joining person has a science background. The conversation takes notably different turns if the joining person is perceived as a non-science person. We recorded numerous instances where the conversations became more technical and work-oriented after those present were told that we, the authors, also had graduate degrees in science. However, in the case of non-science people joining a group of ecologists resulted in a change of the conversation.

Unlike the work of laboratory-based scientists, ecological field research varies widely in terms of its geographic locations and circumstances. Even ecology researchers from a single research group can work in geographic locales distributed around the world. This lack of similarity in research settings provides ecologists researchers with variability in their personal experiences. Thus, when groups of ecologists congregate they possess considerable experiential resources in their conversations with other ecologists permitting members to find common aspects in their observations, practices, and conclusions.

The stories we heard were of several different types and served different purposes; they differed markedly both in form and content from the formal exchanges discussed earlier. Substantial information about the research work in field settings was exchanged in exactly these situations and noted by its difference. We noted four general types of stories that deal with the conceptual or experiential nature of ecology research: (a) stories about ecology based on the tellers personal experiences (b) stories about ecology that were being retold (after being initially told by somebody else in a previous setting), (c) stories of the field which are used to reflect typical situations in ecology, and (d) stories of the (hero and eccentric) characters of the field and their activities. These narratives served different roles, including information about surviving and conducting research in field settings (often related as allegories from which a listener could infer what to do or not do in their own research), how to interpret different observations, and in many cases provided a foundation for social cohesion of the community.

Informal conversations constitute the main source of information on field research practices for researchers in formation. Such information takes a narrative form. We illustrate the use of these stories through presenting a series of field notes and anecdotes collected from informal settings over an eighteen-month span. To maintain an interpretive thread with our earlier content analysis of conference presentations and journal articles, many of these stories are about bears¾ in part because, more than any other organism, bears formed a common thread in discussions amongst ecologists across research settings, topics, interests, and locales through this community. This focus on bears was particularly notable because only three researchers in the communities of ecologists we worked with actually did research with bears (and one of these never participated in any of the informal gatherings that constitute our database).

Storytelling, such as engaged in by ecologists in informal settings, does not often appear to be done just for reasons of entertainment. Usually embedded within the story is an allegory about what actions are appropriate or inappropriate (depending on the narrative) when conducting field research. These allegories deal with issues of safety, what to be careful about from a research perspective when doing specific fieldwork (such as collecting lizards), cautions about using volunteers, and warnings about working with specific individuals.

Ecological fieldwork can be fraught with danger. Poisonous snakes, cougars, and bears are but a few of the dangerous wildlife a researcher may encounter, but numerous other hazards also await the unwary researcher. Although these hazards warrant little mention in the formal texts of the discipline, stories about safety issues are frequently exchanged amongst researchers in informal settings--particularly those working in similar geographic settings. Ecologists, either ecologists in formation or those from other geographic areas, rely considerably on information contained in the stories to develop their local knowledge about any hazards in the area. Allegorical stories were quite frequent in informal settings and provide information that pertains to doing effective research.

The example is an allegory that deals with field safety. In groups such as this the hierarchy within the research project does not dictate who contributes to the conversation. Here, the extent of the local knowledge seems to be a better predictor. Information is shared among participants. A notable aspect of this exchange is that field lore such as "how to make yourself big" is not necessarily based on experience but on other information sometimes derived from exchanges with other field researchers. Topics such as this are not discussed in formal situations. The depersonalized nature of the formal texts would not lead a researcher in formation to suspect that working within a few meters of bears really presented any danger. Field researchers who have had interactions with bears relate their stories in informal settings about these encounters with much greater intensity than that expressed by Stephanie:

Safety allegories were also present in stories relating the conduct of field research. The following field note relates a story told as a warning to field assistants to encourage them to pay attention to the surroundings while gathering data. This story was for the benefit of a less-experienced researcher and included the warning that one should listen for twigs breaking and leave rather than risk a confrontation.

These tales of shared common experiences of survival also contribute to the social cohesion of the community of ecologists because they are examples of common experiences individuals working on even quite diverse projects can share. In an exchange between two M.Sc. students near the end of their program the dangers of participating in field ecology research and the precautions that should be taken come to the fore.

Safety issues were ever-present in stories related about field experiences, especially when newer researchers were present. To newcomers, stories like this relate the physical rigors involved in doing field research serving as a warning for what they might expect. However, these stories seemed to be related less for the utilitarian purpose of exchanging information and more for the sharing of common experiences that contributes to the social cohesion between the ecologists.

Other stories were told which quite clearly were meant to provide insights into the conduct of field research--as a warning against or a suggestion for what actions may be appropriate for certain activities in the field. Given that most formal texts contain few resources about how to do fieldwork, allegorical stories dealing with field research practices have particular significance to the enculturation of newcomers into the practices of research. For many newcomers the exchange of stories about field experiences is a valuable resource in learning to conduct research of ones own. Allegorical stories can also possess different meanings for different members of the audience. One of our informants frequently related a story about a volunteer field assistant she utilized in her fieldwork.

This story can be seen to have multiple meanings for different members of her audiences. For those who were helping her in her fieldwork, this story had the message "do not do this to the lizards." (We both heard this story many times making us particularly cautious in our own attempts of capturing lizards.) For researchers who used volunteer assistants in their own work the story was a cautionary tale with the sense that you have to "keep an eye on your field helpers, choose help carefully." This story was often in response to similar stories from other researchers about the problems of using field helpers. Our informant frequently discussed with us her own experiences with field assistants, how some were helpful while others constituted more work than if she was on her own. She hired them for the sole reason of maintaining "good community public relations."

Allegorical stories that inform research practices of old-timers and researchers in formation also involve research aspects other than data collection. This field note excerpt describes a situation where a story was told again and, in this, obtained a particular salience to those present.

Sam told this story because of her irritation with inappropriate causal claims that she heard both at conferences and from the undergraduates she taught. At the time we recorded this note, it did not seem like much. We found out later that this tale was a reflection of the connections between individuals over time and space.

Stories that are told time and again in a community encapsulate particular community concerns and interests. In this case there are two possible cautionary tales about conducting research offered here. The first relates to the general (scientific) concern of making claims of causality without proper consideration of the variables that might affect your results. The second is a warning about using (and misunderstanding) data which you did not collect yourself¾ an activity which field ecologists think that theoretical ecologists do as they models ecological situations. Thus, this narrative both cautions against making inappropriate causal claims and also contains an implicit critique of theoretical models not grounded in empirical data.

Formal conference presentations provide little information about field research methods on which those in formation could draw to guide their own research. There are, however, other opportunities at conferences to gather such information during informal discussions. These informal discussions are an important resource for researchers. We observed that after the principal investigator returned from an international conference, many conversations over the next three weeks turned up as information gathered there, often in informal meetings. However, our informants frequently referred to such information as "anecdotal" because it dealt with topics unsubstantiated by formal observation. Nevertheless, these conversations influenced the research done by our ecologists. For example, in the following excerpt, Sam describes a conversation she had at the conference with a researcher who worked with pit vipers.

The enthusiasm of both belies the detached stereotype of the scientist portrayed in science education reform documents (e.g., AAAS, 1993; NRC, 1994) and as often portrayed in popular media (such as television and movies). Larry's suggestion influenced Sam who spent considerable time over the following weeks building new, much larger enclosures for her lizards. A discussion about an observed behavior in a species quite different from hers prompted Sam to reconstruct enclosures she had used in a previous field season and (plan to) engage in observations that were not at all central to her original research plan. During an interview seven months later, where she was providing a reading of some of her graphs, Sam commented on the value of "anecdotal" stories she heard at conferences and what they offered her.

Sam recognizes that the stories she hears provide considerable insight into her own fieldwork--even those stories that do not pertain to the organism she studies. Informal conversations contribute more than the information contained in formal presentations. Stories contribute particularly to the work of ecologists and help them contextualize their own "anecdotal" observations. Several months later, Sam talked about the information she had obtained in an informal situation about maternal care in reptiles.

This story about maternal care in reptiles, and how it developed and was elaborated in her conversation with Larry, obviously had considerable salience for Sam because she returned to it unprompted seven months later. In that time she had reconstructed her enclosures, structured her time so she could make observations on the lizards when they gave birth, and concluded from her "anecdotal" observations that maternal care did not exist in the type of lizard she studied. In this case, being able to observe a behavior never before reported in her species and the suggestion that a (distantly) related species might engage in the same behavior was enough to add a new research focus (though temporarily). Informal conversations therefore act as a considerable information resource even for researchers with more extensive field experience.

Informal public talks contained far more information about doing field research than formal presentations. They provided field researchers in formation with a greater resource for learning about doing research. An informal public talk about the bear research described above differed considerably from the formal texts on the same topic by the same research group. Whereas the formal texts presented variables as if they were pre-determined, interactions with bears as if they were benign, and little discussed field methods and experiences, the informal presentation were characterized by information on the methods used in conducting the research. This presentation provided details about how initial field observations resulted in a focus on specific features of the environment (in other words, how variables were developed), described in emotive tones the experience of working quite near to bears, and provided considerable information about how the data from which claims were derived was collected.

As with many of the formal oral presentation in ecology, the talk began with maps and diagrams that provided a geographic context for the research, rationales for choosing the locale, and the historical roots of the project. The initial stages of the study were clearly observational as the researcher developed his sense-of-the-field for that site.

This narrative provides graduate students in formation several valuable pieces of information. These include insights on the frequency of observations, type of information to make note of (number of species, what they were eating), how long one surveys the area ("day after day") and the length of time one stays at the research site (from before the salmon arrive until after they are no longer spawning).

Formal scientific texts present variables as if exist they existed a priori and independent of the knowing subject. Formal texts rarely suggest that in field research variables frequently emerge as scientists become familiar with the local setting. For researchers in formation this presents a substantial difficulty, for how should they decide what factors to examine in a setting they have not yet experienced? Informal discussions of research projects, such as the informal talk on bear research, provide a different perspective on variables in ecology research: they may develop as the study progresses. In the excerpt below the researcher talks about how field observations led to making a determination that was to further guide his data collection:

Having determined through field observations that bears consume more salmon than any other organism in the watershed and estuary, the project was narrowed to the impact of the bears alone. In the previous description, a listener could learn information about bear movement patterns not at all present in the formal texts. The researcher then talked about the human agency involved in the research.

This description provides considerable detail, which elaborates the statement in the journal article that the bears were observed at a distance of "two meters to twenty-five meters" (both this description and the one in this informal presentation are still lacking in detail compared to the bear story related in an even more informal setting). For new researchers this narrative is rich in detail about what a field ecologist might encounter when conducting research at nighttime. Notable is that bear behavior at night is such that they cannot see well, do not perceive humans as a potential threat, and will not avoid you if you are on their scent trail. For those about to conduct field research such information, not at all present in formal texts, is of considerable importance.

As is also clear from this passage, and as our field observations support, variables emerged in the conduct of the field research and as the study progressed numerous new variables emerged to the point that at least one new separate project developed from them. How new variables developed in the course of the study was contained in the narrative with such frequency that researchers in formation would have picked up on it. This contrasts formal presentations about his work that implied that variable choice preceded the research.

Information provided in informal settings can also influence the conceptual understandings of researchers in formation and can thus guide future research. The following excerpts show how researchers in formation can learn to do field observations on quite disparate organisms at different locations. Such information also provides insights into animal behavior and ecological principles.

The researcher speculated reflected particular motivations behind the animal’s actions, which is of interest, but not the sort of thing reported in a formal journal article. However, individual field observations gain a broader salience and meaning when other individuals contribute to the conversation. In this case, a contribution made by another group member provided broader ecological context to the speculation.

This comment resulted in the conversation returning to the first speaker who strengthened his comment about the possible reasons for the behavior observed in the salmon based on the comments about the marine mammals.

Thus, observations made of similar circumstances, although with different species, contributed to a broader sense of what was happening in nature. For researchers in formation, such conversations offer insights into how field observations lead to new studies and how knowledge claims develop in ecology.

This paper, unlike most papers in science studies about the construction of knowledge, focuses on formal and informal texts available in formal and informal settings contribute to the information of those in formation. Little work on the exchange of information by scientists in informal settings such as bars, parties, or over coffee has been conducted; previous studies of informal texts instead examined discussions held in the laboratory as work was being conducted (Lynch, 1985), explicit discussions of conceptual issues (Garvey & Griffith, 1971), and information solicited in interviews (Gilbert & Mulkay, 1984). Unlike these studies, we observed our participants as they discussed their research in formal and informal settings unprompted by our intervention. From these observations we conclude that the formal and informal texts are fundamentally different in the information they convey about the conduct of field ecology research. These conversational settings have some similarity to those that were examined in a community of service technicians (Orr, 1990). The conceptual setting of our work differs from this work in that the communities under study engaged in quite different tasks. Orr studied technicians who were engaged in repairing office equipment. They both knew what their purpose was (to repair machines) as well as what the indicators of success at that task would be (a working machine). In addition, the focus of their work, a machine, was constructed of a limited number of parts that can experience a limited number of possible breakdowns.

At the end of four years of enculturation with formal texts undergraduate ecology students have been presented with the broad conceptual issues about which ecologists are concerned, but have few resources on which they can draw to plan and conduct field research. Students experience few opportunities from which they can learn that research is conducted (and papers written) as part of an inter-related series of investigations by a researcher or a group of researchers--this is almost never discussed in lectures and the structure of the laboratory exercises does not emulate this. In fact, school-based research activities (such as the one described above) and the structure of lectures develop in students a sense that the knowledge claims of ecology exist as distinct and unconnected bits (Bowen & Roth, 1998). Most crucially, lectures and other formal teaching situations do not provide students with a sense of how research is done. This lack of embeddedness extended beyond the decontextualized, impersonal, presentation of ecological truisms found in lectures and textbooks to the practice field research activities (the equivalent of physics and chemistry laboratory exercises) in which undergraduate ecology students engaged. In addition, courses rarely offer students the opportunity to examine a series of papers from a single author which would help them develop the understanding of the embeddedness of individual research projects in the broader of the discipline and individual researchers. For example, the Science Citation Index provides evidence that the project on bears is embedded in a series of projects in that area over twenty years involving organisms at different trophic levels. In addition, collaboration with other researchers from other disciplines has allowed their tool-based practices and skills to be adopted for application to the broader ecological interests of the informant. Students have little guidance to aid them in observing research in this manner.

Many authors noted that a characteristic of most formal texts of laboratory science is the removal of all aspects of agency of the actors (e.g., Latour, 1987). Thus, through this stylistic structuring, greater authority is attributed to them because of distancing of the subjectivity of the agents involved in their construction (Gross, 1996). This process of lending authority to formal texts is enhanced by the manner in which the research is discussed ahistorically. The ahistoricity of the presentation in formal texts of knowledge claims and variables is problematic for those in formation planning their own field research. The absence of a sense of the passage of time provides those in formation little understanding of how experienced researchers frame and develop understanding. This includes the understanding of problems, how the concerns of the discipline change, and what variables are important in doing the research.

The formal texts of science present an objectivist perspective of the claims and processes as if they were purely factual (Gilbert & Mulkay, 1984). This differs from informal accounts of field research that has a contingent character. The degree of detail in informal texts about field practices (how observations are made, how features are considered important enough to warrant being defined as a variable for study, and how interpretations change over time) is not found in the formal texts. Discussions of field research in informal settings present this information differently than formal texts. Little or no discussion of this occurs in formal texts. These informal texts (and our ethnographic work in the field) reveal that data is collected because the opportunity arises for it to be collected, not because there is necessarily an immediate or conceptual reason to do so. Thus, the informal settings provide context and information about field research unavailable in formal texts to researchers in formation. For new (and old-timer) researchers detailed descriptions of fieldwork constitute a useful resource informing their (future) field research work. Informal settings provide this very sort of information within the narratives exchanged as researchers talk about their work and discuss their experiences.

In addition, the underdetermined nature of the formal texts has more significance for researchers in formation than for experienced researchers. Experienced researchers can draw on their own experiences to contextualize the formal written texts but those in formation have to rely on the text itself; so that which is unstated in those formal texts, available in some sense to experienced researchers as they use their lived experience to make sense of their readings, is unavailable to those in formation.

Interactions and storytelling in informal settings are quite important to ecologists for reasons other than for research purposes alone--ecologists seem to form a network of acquaintance that is based around the exchange of stories. Stories appear to reinforce this network of acquaintance in several ways and contribute to a social cohesion amongst ecologists. Within informal settings stories of work in the field are told over and again which contributes to cohesion between members that crosses more than just the immediate group of participants. Being a member of this community means being able to participate in those stories in diverse settings as one meets unfamiliar ecologists when traveling or conducting research. We noted that several aspects of informal discussions contributed to cohesion (as well as being resources for new researchers). These aspects include discussions about individuals, discussions of common experiences, discussions of experiences across which ecological parallels can be drawn, and discussions of common (ecological) cultural touchstones that have salience to community members.

Social cohesion develops when individuals can find common ground in their interests and narratives. The excerpts related earlier represent one way in which stories contribute to social cohesion--by offering the opportunity for individuals to find examples of persons they know in common. In addition, such widely shared stories contribute to social cohesion by providing narratives that communicate concerns and issues of the discipline in a manner accessible to any member of a group in which the story is discussed. Thus, exchanging stories allows ecologists to form a community not just of practice and concerns and language, but also of social experience--our work suggests that the social interactions outside of work hours are of considerable importance in becoming an ecologist.

During our ethnographic work who-knows-whom was a frequent topic of conversation in informal settings. We noted that individuals who conduct field research, even in unrelated topic areas, are often broadly known. Researchers were known and talked about because of groundbreaking research they had done in the past and were commented upon in conversation in the context of talking about individuals who had influenced their own research work. In one case, an individual who did graduate work with turtles in the mid-1980s was known by ecologists who worked with lizards, migrating ducks, snakes, tree ecology, and minuscule insects. We also noted that knowledge about various individuals and being able to relate tales of that work appeared to be a key factor that allowed us to be accepted into groups of ecologists. We were struck by the extended network that linked ecologists, for it was a rare assembly where there was not at least one common acquaintance. Clearly, part of his acceptance into groups of ecologists was because Bowen could claim prior acquaintance with other members in the network of ecologists.

Story telling, and the topics and individuals about which those tales are told, provided us access into even newly-met groups of ecologists. In this, as we noted earlier with inexperienced researchers, it was not sufficient to merely have experiences to tell, but to be able to relate them in the context of the work of others that was important. We observed newer graduate students, over many months, at first sit quite quietly in groups, and then begin to contribute field-based narratives into the context of the conversations. As noted, this rarely happened when they were in groups of M.Sc. graduate students (they did not try out their stories in settings of peers) but instead usually occurred in discussions at which professors, Ph.D. students, or post-doctoral fellows participated.

In part, social cohesion develops because of experiences shared in common not directly related to research but arising as a consequence of engaging in field research. For instance, whether one is concerned about ecology of a particular lily that grows on the edge of meadows, studies lizard biology, is interested in insects which grow at the top of thirty-meter tall trees in rainforests, examines distribution and breeding of woodpeckers, or is interested in the utilization of spawning salmon in coastal streams¾ quite diverse ecological topics¾ there are stories which are based in the common experiences of those respective individuals which can be related amongst individuals in those projects.

Informal settings also offer ecologists the opportunity to share observations that they refer to as "anecdotal," in other words, observations that are not sufficiently substantiated to warrant publication. Our observations during informal conversations suggest that these conversations help ecologists make sense of their anecdotes in a broader ecological context by providing a forum for people working on different organisms or in different biomes. For instance, the observations regarding salmon behavior at nighttime in the estuary and porpoise behavior are not related in any formal texts, and were not mentioned even in the informal verbal presentation. Yet sharing these observations helped develop understanding of ecological relationships. For researchers in formation, these anecdotal conversations would contribute to their understanding of the development of scientific claims.

Many of the stories that contribute to social are heroic stories because they relate unusual accomplishments and dangerous encounters. Although these heroic stories occur to some degree in other informal settings, they are notably frequent in bar settings attended by mixed groups of researchers. In these settings stories about getting into street fights in foreign countries, dangerous situations in remote areas, and survival in adverse conditions are common. It is not surprising that much of the stories involved bear because of the geographic location of our research groups and the wilderness settings in which they worked. Thus, in bar settings stories about research on bear-salmon interactions included details not provided in other forums.

In this most informal of settings, we heard stories about working with the bears and salmon unlike any of those related in any of the verbal presentations or formal writings about the work. For researchers in formation, this is a view of research unavailable in other contexts. In contrast to his formal texts, our bear researcher admitted how nervous he was initially. In the public talk he told how close he came to the bears and how he could see them with his night glasses but that the bears were unable to see him. In the bar setting, we first heard him talk about the initial stages of the research when he saw little after night fall but felt and heard the bears close at hand. What our words here communicate poorly is the intensity of the feeling he had about those encounters¾ the initial perception of danger was communicated in his voice unlike that in any other setting.

Heroic stories about bears are not just related by the researchers engaged in bear research, but are common currency in the stories of field research, as the following field note suggests. These heroic stories also contribute to the social cohesion of the ecologist community.

These heroic stories were present in other settings as well, but the social groupings in the bar (where professors sat with graduate students more so than other places) appeared particularly conducive to the relation of harrowing stories of survival and danger in field settings. What is significant about these stories is that while some stories were told and retold in the form of autobiographies, others were retold by individuals who had heard the stories from another ecologist. Thus, these stories, much as those about well-known individuals (such as the east coast eccentric) constitute some of the cultural capital that constitutes membership in the community of ecologists. By learning, and being able to retell, these stories new members are enculturated into the community thus gain access to the pool of knowledge and the individuals within it who share these common experiences and stories.

Becoming a member of the community of ecologists is more than becoming a member of the community of practice, for being an ecologist is more than just engaging in the science aspects of field research and interpretive practices. The stories and other informal exchanges within ecology provide substantial resources for new ecologists to draw upon for their own work¾ often unavailable in the formal texts of ecology. However, informal settings and the narratives related in them also embed new ecologists in a social framework of experiences and activities apart from the academic ones. Unlike participating in many professions, participating as researcher in ecology includes commitment to social gatherings (Friday beers, Wednesday pool, conference drinking) as much as it includes commitment to doing research.

Undoubtedly, part of the reason for differences between the content of formal and informal oral presentations was the amount of time and space available; the informal talk was eighty-five minutes in length compared to formal conference presentations are typically fifteen minutes long. However, the ecology lectures we observed were fifty minutes in length and yet rarely discussed field research in the manner it was presented in informal settings. This suggests that the differences in content reflect an epistemological view that students are there in a lecture to learn the knowledge claims of ecology and not about "anecdotes" of field practices--even though field researchers themselves recognize the importance of anecdotes to the success of their work.

Stories play an important role in the social fabric of practitioner communities. Thus, "knowing the stories and performing in the appropriate style is an unmistakable sign of being a real particle physicist, of knowing particle physics, and of knowing how to make knowledge of particle physics" (Traweek, 1988, p. 121). However, Traweek’s descriptions of the stories suggests that they are often highly technical using specialized language and that the conversations frequently dealt with learning what new claims are made and what new techniques are developed in the discipline. In this, the stories of physicists differ from those told among ecologists which often deal less with knowledge claims and more with doing research; how someone deals with a research problem or survives a dangerous situation. Nevertheless, the outcome of this "gossip" amongst high-energy particle physicists seems similar to what we note in ecologists--it helps structure and connect a small, geographically diverse community. However, whereas stories are used by particle physicists to control who is (to become) a particle physicist, being able to participate in the stories is one of the ways in ecologists are accepted. Informal settings contribute substantially to the success of field research. This needs to be considered when educators examine and plan the programs for those in formation. Divisions that occur in post-secondary institutions between undergraduates, graduate students, and professors are clearly counter-productive; for the most substantial resources of information on field practices rarely interact in informal settings or socially with those who are about to engage in their own work.

The authors would like to thank Sylvie Boutonné who transcribed a number of the videotapes used in this research. We thank the late David Gaskin, Professor of Marine Biology, who distinguished himself through his compelling stories-of-the-field told in his undergraduate classes.

Bowen, G. M., & Roth, W.-M. (in press). Why students may not learn to interpret scientific inscriptions. Research in Science Education.

Bowen, G. M., & Roth, W.-M. (1998, April). Isolation of variables and enculturation to a reductionist epistemology during ecology lectures. Paper presented at the annual conference of the American Educational Research Association, San Diego.

Bowen, G. M., Roth, W.-M., & McGinn, M. K. (1999). Interpretations of graphs by university biology students and practising scientists: Towards a social practice view of scientific re-presentation practices. Journal of Research in Science Teaching, 36, 1020-1043.

Coy, M. W. (1989). Being what we pretend to be: The usefulness of apprenticeship as a field method. In M. W. Coy (Ed.), Apprenticeship: From theory to method and back again (pp. 115-135). Albany: State University of New York Press.

Delamont, S., Atkinson, P., & Parry, O. (2000). The doctoral experience: Success and failure in graduate school. London: Falmer.

Eisenhart, M. (1996). The production of biologists at schools and work: Making scientists, conservationists, or flowery bone-heads? In B.A. Levinson, D.E. Foley, & D.C. Holland. (Eds.) The cultural production of the educated person: Critical ethnographies of schooling and local practice. New York: SUNY Press.

Garvey, W. D., & Griffith, B. C. (1971). Scientific communication: Its role in the conduct of research and creation of knowledge. American Psychologist, 26, 349-362.

Gilbert, G. N., & Mulkay, M. (1984). Opening Pandora’s box: A sociological analysis of scientists’ discourse. Cambridge: Cambridge University Press.

Goodwin, C. (1994). Professional vision. American Anthropologist, 96, 606-633.

Gross, A.G. (1996). The rhetoric of science. Cambridge, MA: Harvard University Press.

Guba, E., & Lincoln, Y. (1989). Fourth generation evaluation. Beverly Hills, CA: Sage.

Jordan, B., & Henderson, A. (1995). Interaction analysis: Foundations and practice. The Journal of the Learning Sciences, 4, 39-103.

Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Milton Keynes: Open University Press.

Latour, B. (1993). La clef de Berlin et autres leçons d’un amateur de sciences [The key to Berlin and other lessons of a science lover]. Paris: Éditions la Découverte.

Law, J., & Lynch, M. (1990). Lists, field guides, and the descriptive organization of seeing: Birdwatching as an exemplary observational activity’, In M. Lynch and S. Woolgar (eds.), Representation in Scientific Practice (Cambridge, MA: MIT Press, 1990), 267-299.

Lynch, M. (1985). Art and artifact in laboratory science: A study of shop work and shop talk in a laboratory. London: Routledge and Kegan Paul.

National Research Council (1994). National science education standards. Washington, DC: National Academy Press.

Nespor, J. (1994). Knowledge in motion: Space, time, and curriculum in undergraduate physics and management. London: The Falmer Press.

Nutch, F. (1996). Gadgets, gizmos, and instruments: Science for the tinkering. Science, Technology, & Human Values, 21, 214-228.

Orr, J. E. (1990). Sharing knowledge, celebrating identity: Community memory in a service culture. In D. Middleton & D. Edwards (Eds.), Collective remembering (pp. 169-189). London: Sage.

Roth, W.-M. (1996). Where is the context in contextual word problems?: Mathematical practices and products in Grade 8 students’ answers to story problems. Cognition and Instruction, 14, 487-527.

Roth, W.-M., & Bowen, G.M. (1993). An investigation of problem framing and solving in a grade 8 open-inquiry science program. The Journal of the Learning Sciences, 3, 165-204.

Roth, W.-M., & Bowen, G.M. (1994). Mathematization of experience in a grade 8 open-inquiry environment: An introduction to the representational practices of science. Journal of Research in Science Teaching, 31, 293-318.

Roth, W.-M., & Bowen, G.M. (1995). Knowing and interacting: A study of culture, practices, and resources in a grade 8 open-inquiry science classroom guided by a cognitive apprenticeship metaphor. Cognition and Instruction, 13, 73-128.

Roth, W.-M., & Bowen, G.M. (1999a). Complexities of graphical representations during ecology lectures: an analysis rooted in semiotics and hermeneutic phenomenology. Learning and Instruction, 9, 235-255.

Roth, W.-M., & Bowen, G. M. (1999b). Digitizing lizards or the topology of vision in ecological fieldwork. Social Studies of Science, 29, 719-764.

Roth, W.-M., & Bowen, G. M. (2001a). ‘Creative solutions’ and ‘fibbing results’: Enculturation in field ecology. Social Studies of Science, 31, 533-556.

Roth, W.-M., & Bowen, G. M. (2001b). Of disciplined minds and disciplined bodies. Qualitative Sociology, 24, 459-481.

Roth, W.-M., Bowen, G. M., & McGinn, M. K. (1999). Differences in graph-related practices between high school biology textbooks and scientific ecology journals. Journal of Research in Science Teaching, 36, 977-1019.

Traweek, S. 1988. Beamtimes and lifetimes: The world of high energy physicists. Cambridge, MA: MIT.

van Maanen, J. (1988). Tales of the field: On writing ethnography. Chicago: University of Chicago Press.

*This work was made possible in part by a Post-Doctoral Fellowship (to GMB) and grant 410-96-0681 (to WMR) from the Social Sciences and Humanities Research Council of Canada.

About the authors...

G. Michael Bowen is on faculty at Lakehead University, 955 Oliver Road, Thunder Bay, Ontario. P7B5E1. All Correspondence should be directed to him via post or e-mail: gmbowen@compuserve.com.

Wolff-Michael Roth is a Lansdowne Professor, Applied Cognitive Science, at the University of Victoria, British Columbia.