How to Write a Scientific Paper


How to Write a Scientific Paper

Anthony Raw

Copyright © 2016 by Anthony Raw

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any electronic or mechanical means, including photocopy, recording, or any information storage and retrieval system now known or to be invented, without written permission from the author, except by a reviewer who wishes to quote brief passages in connection with a review written for inclusion in a magazine, newspaper, website, or broadcast.

Shakespir Edition

To my father, John Frank Raw

an excellent observer and a patient teacher


Anthony Raw has lived has lived nearly half his life in Britain and half in Brazil. He is a citizen of both countries. Since 1976 he has lectured in ecology, zoology and biological systematics; 24 years at the University of Brasilia and 11 at the State University of Santa Cruz in Ilhéus in the State of Bahia. Dr Raw’s experience suits him very well to write the present volume. He has written over one hundred scientific articles, books and reports in English and in Portuguese.


Students often remark that the most useful class in my ecology courses was a lecture on how to write a scientific paper. Some of them suggested I write a book on the subject. It has taken a while, but here it is. It is aimed at students starting a career in research and the emphasis is on composing a paper on biological investigations. The book has been improved with the comments of Mr Alan Halley and Dr Paulo Cesar Ribeiro Barbosa, both of Ilhéus.

Table Of Contents




















The researcher must write papers. There is little point in conducting research if you cannot communicate the results. Also, it is a pleasure to see your work in print and know that other scientists will read it.

Even so, nearly half of the manuscripts received by some international biological journals are rejected because they are poorly written. That is more than those rejected on scientific grounds (Endler 1992, Moore 1994).

This volume explains how to write a scientific paper. It contains the secret to success presented in guidelines, advice and some general rules. However, there is no short cut for learning to write well. It takes organization, practice and constructive criticism from your peers. This helps in the organization.

Research students

The number of students pursuing masters and doctoral courses in universities has increased enormously during the past decades. They research a chosen field of interest and their project is supervised. Help is available on project design, on collecting, statistical analyses and the like. Few courses provide adequate, formal training on how to present the findings in a scientific paper.


It is instructive to compare the method of writing of the novice and of the professional.

The novice

The novice does the following.

1. He writes slowly and neatly using every line of the page. That is how he learnt at school. He often stops to think and the pauses are long. He rewrites many sentences and even paragraphs. He often uses a dictionary, checks spellings and consults references. The coffee breaks and conversations are excuses for not writing, but the interruptions just prolong the agony.

2. Soon he reads the first draft carefully and rewrites large parts of it and checks the spelling carefully, then prints out the result as finished copy.

3. The result is disappointing. It is difficult to read because the thoughts are confusing and poorly expressed.

4. After a time, he rereads his work and decides it will be too much trouble to improve it, so he sends it off as it stands or abandons it.

The professional method

The professional’s secret is that he works in stages. Do not forget, you should consider yourself a professional. He starts writing with some confidence and is not in a rush to finish the job. He does not try to make the first draft the final one. While he is writing one stage, he forgets about the others. He leaves time to revise the text carefully. In the following chapters, we will look at this method carefully.

What to include

You might have collected enough information for more than one paper and the task of separating them seems overwhelming. It is not. Decide how many papers you might have. For each one, write a simple title and a brief summary or just a list of the contents. The motive for doing this is to concentrate your attention on one manuscript and not continually wonder what you will do with the other information. You can transfer information from one to another later if you wish.

Organizing the work routine

An hour’s writing when the words are flowing produces more text and of better quality than many hours of forced work. Identify the time of day or night when you write best. It could be morning, afternoon, night or the early hours. Some people have trouble at any time; write when it is less difficult.

Do not feel guilty about taking frequent, short breaks if they help you write better. The brain functions in short bursts of activity of a few minutes and then relaxes, like muscles do.

Take every advantage of the times when the words are flowing. Avoid all interruptions. Do not look up a reference, consult a dictionary or answer the phone. Leave these chores for other times. Organize your workspace and computer so that you work efficiently. Make your pot of coffee before you sit down to write.


Classical structure of the manuscript

This book explains how to produce a classically structured scientific paper. There are four sections, and five if you need a Study Area. Imagine the height of Figure 1 to be the breadth of the subject. Start with a broad Introduction, then reduce the scope to the subject you investigated. Continue along this narrower line. In the Discussion you first discuss your own results, then relate them to the broader body of knowledge on that particular field of interest. That is why in the Figure the last part of the Introduction, Study area, Methods, Results and first part of the Discussion are all narrow.

Figure 1. Classical structure of a scientific paper. The height represents the breadth of the subject.

This structure demonstrates the three demands of a scientific paper.

1. The Methods must demonstrate clearly that reliable methods were used.

2. In the Results you report your work and only that.

3. In the Discussion you discuss your results and relate your findings to the work of others, to the body of existing knowledge on the subject.

The Methods are more important than the Results. Any doubts about the Methods and the veracity of the Results is questionable. Nonetheless, the Results sell the paper.


When the director, Sam Goldwyn was asked about his secret for making a spectacular film he replied that you “start with an earthquake and work up to a climax”. This is wise advice for writing a scientific paper. Try to capture the reader’s attention early in the Introduction. Start with a short, snappy sentence. Daniel Janzen (1976) began a review on the flowering and fruiting of bamboos “In the years 919 and 1114, the mainland Chinese bamboo … seeded en masse”. The reader will be wondering who collected the data, did the interval continue every 200 years afterwards ? He is bound to read the paper.

If you studied the ecology or genetics of a species, you should start with something indicating why you chose that subject or the potential relevance of the findings. Do not start with dry facts on the world distribution of the genus or a summary of its taxonomy. You are not writing a textbook.

1. First, present the subject. For example, in an investigation on how light affected the germination of a particular species you might start with how seed dormancy in other species is broken.

2. Provide references to previous work on the subject and brief comments on the main achievements to date.

3. If appropriate, explain how you came up with the idea. Maybe some preliminary observations gave you the idea, or a comment in a publication, or a discussion with a colleague.

4. Present the objectives clearly. Avoid stating the obvious.

Study Area

This and the next section (Methods) are fairly simple to organize. Provide enough information about the place or places where the fieldwork was conducted so that others can return to the site.

● If you need a map of the place, find one or draw one.

● Give the geographical coordinates of the site or sites in the text.

Cite the distance and direction to the nearest town or other notable geographical point. The reader will like that.

The site is 15 km south of the centre of Brasilia.

Write up a first draft of the description of the Study Area while you are still conducting the work. It is irritating to discover in the midst of writing up a piece of work that you forgot to note some simple piece of information like the breadth of a river or the height of the forest canopy. You will have to return to the site – and if you have moved town you have a problem. If there is little to relate about the Study Area, you might include notes on the location in the Methods section.


The methods you used must be reliable and appropriate. The better journals are far more exacting over them than over the results. Explain them openly and explicitly. Give sufficient information for others to repeat the work. (That is a fundamental demand of scientific method; the method you used must be testable and the results repeatable.) Note the dates and the times of day or night that you worked. Remember to state if it was during daylight-saving time.

Clearly, you used your experience when designing the experiment but, even so, an investigation often has some simple biases. You worked at the place where you knew a species occurred. If you collected only on sunny days, say so. It is essential to write the first draft of the Methods while conducting the work, as for the Study Area.


In the Results section you present your own work. Describe your results in detail. Construct and print out various presentations of tables and figures. They illustrate the interesting points of your but, at this stage, you need not decide which to include in the manuscript. Later you will discover which are the best to hang the writing on.

You can also mention any unsuccessful experiments. They are not personal failures; probably they eliminated useless lines. In that way, they comprise part of the results. The reader will want to know of the errors so that he doesn’t repeat them.

Draw attention to the more interesting aspects of your findings. You collected interesting information so tell the reader about it. Do not simply repeat the data already presented in the table or figure; explain what is important. In a column of data in a table, call attention to the highest and lowest numbers, and numbers that are similar. On a graph, point out the peaks or depressions or the differences between two lines. Also, say which of the data are statistically significant (and maybe which are not).

In the Results, do not say why the findings are interesting or what might have caused them. Make your comments only in the Discussion.


In the Discussion try to follow this sequence.

1. First, discuss your own data and explain your findings clearly. Here you can include all those comments you wanted to make in the Results on why you think your findings are so fascinating. If some data are inconsistent, you might say so. If you used a new method, comment on its usefulness, efficiency and accuracy.

2. Then relate your work to that of others and point out the similarities and differences and the relevance of your work. The reader must be in no doubt about your contribution to the existing body of knowledge on the subject.

3. Conclusions must be short and pertinent. They might be included as a final part of the Discussion or presented as a separate section. Which of the two options you use will depend on how much you write and what the journal accepts.

More than one Discussion in a paper is unacceptable. If you are convinced that more than one is required, you should consider dividing the manuscript into two.

Separation of Results and Discussion

The separation of Results and Discussion is to eliminate any confusion the reader might have over what your contribution is.


It is often helpful to divide some sections into sub-sections. They split the text into parts and that helps the reader understand them better. Use your common sense. Divide sections that are longer than a page or two. For example, if you investigated a pest on a crop, the sub-sections of the Results and the Discussion might be numbers of eggs laid, numbers of adults, the pest’s diseases and predators, the amounts of damage caused.


Before you start writing the text, write a title and a synopsis. They are not to publish so do not have to be carefully written. They are to focus your thoughts on what to include in the manuscript. The synopsis could be a list of the main findings and conclusions. You will write the actual title and abstract later.

The method of working is done in stages wh9ch are quite simple.

1. Put all the information onto paper or the computer.

2. Fill up the four or five sections (cited in Figure 1).

3. Check that each piece of information is in the right section.

4. In each section, check that the pieces of information are in the right sequence.

5. Write for yourself.

6. Write for others.

7. Check the technical material.

8. Revise the text.

9. Obtain colleagues’ criticisms.

Let us examine these points in detail.

1 Transfer the information to the page

You cannot do anything with the information while it is in your head or scattered about in field books, computer files and on scraps of paper. Put it all onto the computer file which is to be the manuscript. You will have a mixture of lone words, phrases, sentences and even paragraphs. That doesn’t matter yet.

2 Place the information in the sections

After you have transferred the information to the computer, place it in the four or five sections shown in Figure 1 (Chapter 3). Use a separate page for each section. This is the skeleton.

3 Check where the information is located

Check that every piece of information is placed in the right section. Move the pieces around until you feel happy they are in the right place. Do not worry if you are not sure about where to place some of them. You can move any piece of information any time you want.

4 Place the information in sequence

Examine each section and place the pieces of information in it in the correct sequence. If in doubt over this task, it might be easier to print out everything to clarify your thoughts.

5 Write the manuscript for yourself

Write everything you want to say any way you want (Chapter 5). Simply write as it comes out of your head. At this stage, it doesn’t matter how it comes out. You can modify, correct, reorganize and improve the work only after it is on the computer or on paper.

6 Write for others

This will be your manuscript (Chapter 6).

7 Check the technical material

Check all the figures and tables and references and related stuff (Chapters 7 to 10).

8 Self-editing

Allow plenty of time to revise the work carefully (Chapter 11). Usually you will do this self-editing several times.

9 Colleagues’ criticism

Pass the manuscript to colleagues for critical reading (Chapter 14). Remember that nothing is truly decided until the editor sends your manuscript to the printer.

As you follow this sequence, you will see how the structure and content of your manuscript are evolving. In the next chapters, you will see how to follow these rules and finish with a respectable manuscript ready for publication. The finished work will be clearly written and well organized and easy to read and understand.


To write well is not easy for the simple reason that to write well you must think straight. And thinking straight is never easy.

Cleanth Brooks and Robert Penn Warren in Fundamentals of Good Writing

The fear of writing

Someone new to writing has a terror of a blank page or of a blank screen. But, ask anyone to talk about their research and the information will gush out. This identifies the real problem. It is not a question of communication, but the form of communication – of writing. The person does not have the confidence to write what he wants to communicate. We have to transfer this ability to writing.

This difficulty is a result of the labours of school and college teachers the world over. In the U.S.A., “Many English teachers strongly encourage students to write poorly by giving better grades to complex, indirect, wordy, and inflated writing than to simple, direct, concise, and understandable writing” (Hake and Williams 1981 in Moore 1994). Likewise in Brazil, Wagner Valentini (1998) observed, “Mid-level teaching focuses on the literary form of writing. This is with long, complex and rhetorical phrases to pass images and sensations to the reader.” That might be fine for literary writing, but not for articles on science.

Like any ability, it is possible to learn the technique. Anyone who is able to conduct research is able to write a good report or scientific article. I am not suggesting that you can learn to write like Shakespeare or J. R. R. Tolkein. We are dealing with means of communication and not of art.

Starting to write

When you have put the information onto the computer or onto paper following the procedure presented in the previous chapters, you will be ready to start writing the manuscript.

In the film Finding Forrester, young Jamal is sitting in front of a typewriter gazing into space. The old man, Forrester asks him what he’s doing.

Jamal “I’m thinking what I’m going to write.”

Forrester “Don’t think ! Write ! You can start thinking after you’ve started writing.”

It might sound crazy, but it is good advice. Get into the rhythm of writing and the words will begin to flow.

Forget those school lessons about writing, when you had to write as correctly as possible from the second you put pen to paper. You had to include every comma and to spell correctly. You knew that only the teacher would read it. At this stage it doesn’t matter about:

writing phrases instead of sentences,

writing stuff that only you understand,

writing neatly,

finding the right word,

repeating words,




Write like a professional. Start off writing quickly. You will often stop to think, but only for a few seconds, then type again as fast as you can. Ignore errors and doubts. You might correct minor errors, but do not allow them to interrupt the flow of thought. You already have considerable experience of writing like this, when you were sitting college examinations.

If a passage is difficult, mark it and move to the next one. When the rest of that section is written, it should be easier to deal with the hard part.

Include notes in the text to check information later. Whenever I have a point I think I may want to check, I type $$ or ##. These marks are easy to find afterward.

Sequence of writing a paper

Follow this sequence.

1. Study area

2. Methods

3. Results

4. Introduction

5. Discussion

The first three in this list are the easiest. The Introduction and Discussion are more difficult. They are closely linked; one tells the state of the topic before you conducted your research, and the other afterward. Those are the sections where you cite references. The Discussion is always the most difficult.

Include everything

Do not wonder if you have written too little or too much. This is not the time to decide what to cut and what to include. Include everything. Do the pruning later. Do not use a dictionary or check a reference at this stage. Why check a word that you might delete ? Some of what you write will make sense only to you. You are still not writing the article, so do not worry about the presentation or about details.

Which word ?

It is often difficult to find the right word. A newcomer is exasperated to see how quickly the words flow from an experienced writer. Do not despair; the more you write, the quicker you find the right word. Often the budding write is in doubt because more than one word comes to mind. If in doubt, write them all down. While the words are spilling out do nothing to interrupt the flow. If you stop to decide which word to use when in the midst of writing, it might be difficult to pick up the thread again. It is far more important to continue writing than to decide on a particular word. I solve the question in the following way.

When you have more writing experience/ training/ skill it will be easier/ quicker/ faster to find/ remember/ recall the right/ appropriate/ suitable word.

To show the groups, the words of the same colour are almost synonyms. Later, when you are re-reading the text, it will be easy to choose the word you prefer.


You write the first draft with your heart and write the second with your head.

The film – Finding Forrester

In the film Finding Forrester, in this one sentence, the character William makes a great statement which sums up the key to good writing.

Choosing the reader

As your work is so fascinating, you might want to include everything you discovered; the reader will be more selective. Bear in mind that you are the author and are not important. Only the reader is important.

The key to an experienced writer’s success is knowing what will interest the reader. Write with someone in mind. Forget your supervisor or anyone else close to your research. Imagine you are recounting your labours to a friend or relative who is not a specialist on the subject and wants to know what you did. Whenever you are in doubt about how to explain a point, think to yourself – how would that person best understand it ? The importance of this approach cannot be over emphasized. Consider the following questions.

1. What does the reader already know about the subject ?

2. What do you want to tell him ?

3. What does he want to know ?

4. How can you best tell him ?

5. If you were in his position, would you be interested ?

Scientific style

Style is about how a text is written as opposed to what is written. There is no great secret in identifying good style; it is what a person enjoys reading and finds easy to understand. A clear, simple style is the most appropriate for any writing whose major aim is to communicate information.

According to the English novelist, Somerset Maugham and Cervantes, author of Don Quixote, this criterion is also valid in literature. If you feel the communication of feelings and sensations requires a long-winded style, read some Haiku.

You are writing for scientists. You are not composing a poem or a political speech or a letter requesting employment. Here is a sentence in everyday style and one in scientific style.

The day we saw the tree it was a mass of large beautiful flowers the colour of ripe lemons.

It was estimated that the tree bore 4,500 flowers on the day it was examined. The flowers are lemon-yellow and 15 mm to 19 mm in diameter.

Good scientific style is clear, simple, impersonal and emotionally neutral. This is because the objective is to describe and interpret information precisely. Write what you have to communicate and no more and no less.

The scientist reader

If you can’t explain it simply you don’t understand it well enough

Albert Einstein

All researchers have far too much to read. You must persuade them to spend time reading your paper. They must be able to evaluate your work quickly and can do so only if you present the information clearly.

Remember that the reader will be an intelligent person. When he finds a passage difficult to understand, he will know it is the writer’s fault. He will think that an author who writes loosely also thinks loosely, and that will compromise your results. He will quickly identify any attempt to cover up failings in the work.


Using words and numerals for numbers

A common dilemma is whether to use words or numerals for numbers. Simple guidelines apply. All are based on what is clearest for the reader.

1. Always use numerals when there is a decimal point involved.

It is easier to understand 2·56 than two point five six.

2. Always use words at the start of a sentence. You cannot write:

The study was conducted in 2008. 162 individuals were collected at Site D. The period of the first sentence could be read as a decimal point.

It is cumbersome to read: One hundred and sixty-two individuals were collected at Site D. Rewriting the sentence solves this dilemma.

A total of 162 individuals were collected at Site D.

At site D, 162 individuals were collected.

3. Use numerals when the unit comprises a whole number plus a fraction, but use words when the fraction stands alone.

The older plants were 2½ X the height of the younger ones.

Grasses comprised three-quarters of the vegetation cover.

The two following guide-lines sound contradictory.

4. Use words from one to nine and numerals from 10 upwards.

5. Mixing words and numerals confuses the reader – as in: seven to 134. Better to say ….7 to 134.

In practice, you can use words from zero to twelve and other short words, like twenty, thirty, etc. and use numerals for longer numbers.

6. You can mix the two forms when it will clarify the situation.

In the forest twelve plots of 10 × 10 metres each were sampled.

Citing decimal places

There is a big difference between 1 and 1·0 because the latter implies ten times the accuracy of the former. For example, 600 m implies a greater level of accuracy than 0.6 km, just as 8000 sq m signifies more than 0.8 ha.

There is no justification for always citing two decimal places. The number of decimal places relates directly to the number of items in the analysis. There is a standard procedure to calculate how many digits you use after the decimal point based on the standard deviation and standard error of the mean. When a number is 35 of 95 individuals the percentage is 37%, it is untruthful to write 36·8% or 36·84%. However, the best advice is to use your head and be honest.

Conjunctions with numbers

While we are considering numbers, remember you can write five to seven, but not five to six because they are consecutive whole numbers. It should be five or six because there is no value between the numbers. This is a point of logic and not of grammar. However, you write 5·0 to 6·0 because it signifies that there are fractions between them.


Ways of displaying the data

Much of the data will be presented as figures and tables. They are easier for the reader (which is important) and simpler for the author (which is convenient). The two are different; a figure gives a general picture and shows relationships, whereas a table gives the values. Some data can be presented only in a figure and some only in a table. Sometimes you have a choice. Use a table when you do not want to lose important details, or you might use a figure and include some details in the text. Prepare as many as you wish before writing the text, then select the most appropriate ones.

Follow these points when deciding.

1. If you can state the information in a sentence or two, you do not need a figure or table.

2. If the numbers are important, use a table.

3. If you wish to show a trend, use a figure.

4. You cannot present the same information in both a figure and a table in the same paper.

Rules for constructing tables and figures

1. Do not include so much data that the reader might be confused.

2. With percentages, averages, etc., state clearly how many observations were used.

3. If you use abbreviations, explain them in the legend or as a footnote.

4. Take special care with the legend.


Figures are graphs, maps, photographs, diagrams and other drawings. They illustrate what would need a long text to explain. Use standard symbols and abbreviations.

The size of the illustration will be reduced when published so do not make it more than two or three times its final size. If you are not sure if it will be legible, print a reduced copy.


The guidelines for composing legends for figures and tables are similar to those used when composing the title of the article (see Chapter 12). There are three criteria to follow.

1. It should be as short as possible.

2. The meaning must be clear.

3. It should convey the necessary information to understand the figure or table without the reader having to search the text for an explanation. This is not always possible, but you should try.

Remember, a legend is placed below a figure and above a table.

Compare the two following figures. Figure 2 and its legend say little. The reader may guess that roman numerals indicate months and that numbers of individuals are plotted on the y axis, but that is all. Figure 3 is self-explanatory. It contains the information the reader will want.

Figure 2. Numbers of grasshoppers per year

Figure 3. Monthly changes in the numbers of the grasshopper, Schistocerca pallens in a rice field


The four types of graphs are histograms, bar charts, dispersion graphs and line graphs.

Figure 4. Bar charts with identical data. The first is arranged by latitude and the second by the number of species.

Histograms and bar charts

Histograms and bar charts show values simply. However, the data are different.

● A histogram shows continuous variables like lengths and weights and dates (Figure 3) and the sequence cannot be altered.

● A bar chart (or graphic bar) shows meristic values; the data are in interrupted forms – like numbers of species and countries. The columns can be arranged in any sequence you want (Figure 4).

A pie-chart is a fancy way of displaying a bar-chart. It may be prettier, but it is more difficult for the reader to compare values. That is why they are rare in scientific papers, though they are common in magazines, books and internal reports. The same data are displayed in two forms in Figure 5.

Figure 5. The same six values are displayed as a pie diagram and histogram or bar chart

Modifying a scale

Sometimes the range of the data is small, yet worth noting. The presentation in Figure 6a is of little use to the reader. Figure 6b contains the same data. The only difference is the scale on the y-axis. That of Figure 6b, ranging from 32 to 35, is more appropriate for the data. However, you must make it clear to the reader what you have done. This is a trick newspapers and magazines use to confuse the reader.

Figures 6a and 6b. Presentation of the same data on histograms with different scales on the y-axis.

Figure 7. Relationship between the numbers of individuals surviving and percentage dying in each generation

Dispersion and line graphs

A dispersion graph plots the relationship between two sets of variables. Each data point in the body of the graph shows the position of two values, one the number on the abscissa (the horizontal or x-axis) and the other on the ordinate (the vertical or y-axis). These are the independent and dependent variables. The dependent is always on the y-axis. The figure may bear a line showing the relationship when becomes a line graph (Figure 7).

A measurement might be an independent or dependent variable. As one value depends on the other, the difference between the two is obvious. In each pair of lines in Table 1, the dependent variable of one is the independent of the next.

Table 1. Examples of independent and dependent variables

Summary of recommendations in constructing scatter diagrams and line-graphs

1. Place the dependent and independent variables on the correct axes (see Table 1).

2. Use scales that are clear and appropriate. The common scales are arithmetic, logarithmic and semi-logarithmic.

3. Use meaningful scales (compare Figures 2 and 3.)

4. Use meaningful legends on the scales (Compare Figures 2 and 3.)

5. Indicate the points clearly. Take care in selecting them. Lots of vertical crosses (+) will make the graph look like a cemetery.

6. Present different points in different forms (x, o, ●, *, ∆).


Align a map with north at the top of the page. Include the data the reader will need to understand the work and omit superfluous information. If in doubt, ask someone who does not know the location to check the map. Use natural lines which are coasts, rivers, meteorological data, vegetation divisions and contour lines. Questions arise about whether to include artificial lines on an ecological map. Though artificial, a highway or railway might impede the movements of animals. The only justification for including political boundaries is to help orientate the reader. After all, the plants and animals do not recognise them.

Provide a scale, and a legend if necessary. You do not need a map of the entire country to show where you worked. To indicate where a biological reserve lies, include a map of the state or county.


Tables must be clear and concise. This demands care with the content and the size of the headings and spacing of the columns. Do not include too much information. In Table 2 the two right-hand columns (in red) are not necessary.

Table 2. Numbers of individuals of four species of butterflies attacked by parasitoid wasps

There are two types of tables. Those dealing with numbers are statistical tables and those dealing with words we can call word tables. All tables comprise at least two columns and at least two rows. Check if the journal requires vertical and horizontal lines between the parts. Use XL or something similar. Journals do not like tables constructed using the Word program.

Summary of recommendations when constructing a table

1. Arrange the rows and columns in the sequence you think the reader will prefer.

2. Do not leave blank spaces. Fill in the empty cells with “0” or “”.

3. Do not make a table too large. Divide or summarize a large table.

4. Long tables are admissible only as an appendix, for example a list of species or the crude data of a thesis.


Mistakes in identification invalidate an investigation as much as employing an inadequate scientific method

Identification of organisms

For the reader to accept the results he must trust the identifications. In a paper on a well-known species like a large mammal or a crop the reader will accept that you recognise the species you studied. When there is doubt, you should state who determined the species (even if you did so). If voucher specimens are available, identifications can be verified.

Citing biological names

Cite the taxonomic names of organisms correctly. Take care with the spelling. Melita is a genus of shrimps, Mellita of sand-dollars and Melitta of solitary bees.

Three international institutions (for Bacteriology, Botany and Zoology) maintain strict control over the citation of scientific names and their authors. The three International Codes of Nomenclature are independent so they can use identical names. For example, Prosopis is a genus of bees and also of trees. Heliconius is a genus of butterflies and Heliconia of plants. The three Codes are very similar. Some differences are explained below.

Points in common in the Codes

Citations of organisms

1. The name of each species is a combination of the generic and specific names so it is a binomial system.

2. All the names of genera and species are in Latin or are Latinised. They cannot include accents.

3. The first letter of the generic name and only that letter is upper case, like Homo sapiens. Even surnames are cited in lower case, for example, Tecoma smithii. There is no exception to this rule.

Differences between the Codes

There are some differences between the three Codes in the manner of citing the name of the author of a species.


1. Only the name of the author who described the species is cited, as in the toucan, Ramphastos toco Müller.

2. The name of the author of the species is always cited in full. The only two abbreviations accepted are L for Linnaeus and F for Fabricius because both named so many species.

3. When a species is transferred to another genus, the name of the author who described the species is cited in brackets. In 1858, Linnaeus named the blue-and-yellow macaw Ptittacus ararauna L. It was transferred to the genus Ara so is now called Ara ararauna (L). The revisor who moved the species is not cited.

Botany and Mycology

1. When a species does not change its genus, the name of its author is cited as in the Zoology Code, for example, Lactuca sativa L.

2. When a species is transferred to another genus, the name of the one who described the species and the reviser are cited (the latter in brackets) and the two surnames are abbreviated. For example, Bermuda grass, Cynodon dactylon (L) Pers. was described by Linnaeus in the genus Panicum and transferred to Cynodon by C. H. Persoon.


The citation of the authors’ names of bacteria is similar to those of plants and fungi. Agrobacterium tumefaciens (Smith and Townsend) Conn was first described as Bacterium tumefaciens Smith and Townsend. For the names of bacteria, consult Bergey’s Manual of Determinative Bacteriology (Holt/ Krieg 1984).

Citations in scientific papers

Citation of authors’ names

Article 51 of the International Code of Zoological Nomenclature (ICZN 1985) is clear: The name of the author does not form part of the name of the taxon and its citation is optional, although customary and often advisable. Most journals insist that the first time you cite a species in a paper you must cite the name of its author. Some journals also demand that the year in which the name was published be included. Some zoological journals state that the authors of animal species be cited, but not of plants.

Citation of generic names

1. All the names of genera and species are cited in italic.

2. The first time a genus is cited in an article it must be in full. Thereafter, only the first letter of the genus is given. However, there are exceptions.

3. The full name of the genus must be given at the start of a sentence. You cannot write: Several species eat leaves. _*A._ communis *eats seeds. It is similar to the question of starting a sentence with a number (see Chapter 7).

4. Sometimes two genera have the same initial letter. If that is confusing, the name of a genus is cited again in full. However, that is not always necessary, as the following example illustrates.

The two wasps, Polybia ignobilis and Polistes satan were found in open areas, while P._ ignobilis *and _*Polistes versicolor hunted on the passion vine (Passiflora edulis). However, P. satan was not seen to visit P. edulis.

Strict use of the rule would demand that every generic name in this example be cited in full, but the reader easily understands these sentences as they stand.


Sources of information

You will find primary and secondary references. When you begin a new subject, start with the secondaries. These are textbooks, reviews in scientific journals and web sites. Then go to the primary references which are articles in scientific journals, theses, technical reports and verbal communications, all with original data. Use filters in your search for information on the web. Writing simply aedes aegypti on Google you find nearly a million hits and on Google scholar you find 115,000. Using “aedes aegypti” “occurrence in Britain” on Google Scholar you have only three hits.

What is a publication

A question is what exactly constitutes a publication. It must be readily available when released and be a permanent record and, if printed, many identical copies must be produced. By these criteria, internal reports are not publications nor, unfortunately for the post-graduate student, is a dissertation or thesis. If you defended a thesis, you should publish some of the data in a scientific paper and cite the thesis so that people will know about it.

How many references to record ?

Note every reference you find on your research topic. Many are available on the Net so you can copy them quickly and without the danger of erring in the typing. You might record dozens that you do not use, but if you did not note a reference and need it later, you could spend days looking for it. The experienced researcher makes a note of almost everything on the off-chance that he might need it.

How many references to include in a paper ?

It is wise to demonstrate to reviewers that you consulted all the relevant references on the subject. You may not wish to include a particular paper, but citing important recent references shows that the literature search was complete. A reviewer might presume you have not seen a paper. Beware; he might be the author of the missing reference.

General rule on citing information

You must provide the reference to every piece of information you cite in your paper. References to published works are obvious (see below). Doubts are over unpublished information.

Unpublished information

Personal observation

If you collected personal information in addition to that of the investigation you can include it, but cite it as such.

The species was seen 40 km north of the study site (pers. obs).

In addition to flies, the birds consumed wasps (pers. obs).

Personal communication

If you discussed your work with someone who gave you information not yet published or a useful comment then give the person credit in the text.

J. M. Smith (pers. comm.) suggested the plants cannot withstand high amounts of soil aluminium.

The animals often roost in forest palms (C. A. Findlay, pers. comm).

You should also thank the person in the Acknowledgements.

Work in progress

For most journals, information is simply either published or not. A publication is a paper, a chapter or a book. It also includes a manuscript that has been accepted for publication, but not yet published when it is cited as “In press”. You can even cite unpublished theses and dissertations and internal reports.

In contrast, the editor places papers submitted for publication, but not yet accepted, with personal observations and communications. All these must be cited, but they are not yet publications. To cite something as “In prep” is more an enthusiastic opinion than a reference. It is of little use to the reader because there is no guarantee where the work will be published.

Published information

In the list of References at the end of the paper, include only those references cited in the text. This is a simple rule that all editors apply rigorously.

The journal specifies how to cite references, both in the text and at the end of the paper. Check the journal to which you intend to send your manuscript and present them exactly as required – down to every comma. There are two general forms for citing references.

Citation-sequence system/ the numbering system

In the text a number is given to each reference in the order that they appear there. The references at the end of the paper are listed in numerical order which is the sequence they appear in the text. Many journals have changed to this system. They include Science, the Annual Reviews and most in physics, chemistry, maths, medicine, environmental studies and engineering.

Name-year system

In the text the name of the author and year of publication are given. The references at the end of the paper are listed in alphabetical order. Most natural science journals use this system.

Citations in the text

1. Place references in chronological order and not by the accident of the initial letters of the authors’ surnames. When the references are (Smith 1968, 1980, Brown 1977) give preference to the earliest work of each author.

2. Sometimes two authors with the same surname published works in the same year. As their forenames are likely to be different, you cite them in the text by the initials of the forenames as (C. S. Smith 1994, F. Smith 1994).

3. The whole point of a citation is to enable the reader to consult the reference. When the information is buried in a large volume, help the reader where to look.

Interspecific competition is common among coexisting animal species that have similar resource requirements (Darwin 1859 chap 3). Charles Darwin touches on this subject several times in The Origin of Species and the book contains 502 pages. The relevant section is in one chapter.

List of references in the name-year system

Two ways are used to present references in this system. One, generally called the Harvard system is used in almost all journals in the natural sciences so it is given greater coverage here. A second system is often used in the humanities (economics, social and political science, history and also the arts). The two differ. Check in the journal where you wish to submit your manuscript to see which they use.

The Harvard system

This sequence is: 1. Author. 2. Year of publication. 3. Title. 4. Name of journal. 5. Volume. 6. Pages. A reference to a paper will be:

Sale, P. F. 1977. Maintenance of high of high diversity in coral reef fish communities. American Naturalist 111: 337-359.

The symbol “ : ” signifies that what follows are the page numbers; an abbreviation that taxonomists have used for many years.

Several references by one author in the same year

When two or more references are by the same author in the same year they are cited with a, b, c, etc. following the year.

Whitmore, T. C. 2000a. Biodiversity. New Scientist p.57, 11 March 2000.

Whitmore, T. C. 2000b. The case of tropical rain forests. The sustainable development of forests: aspirations and the reality. Naturzale 15: 13-5.

Indirect citation

An indirect citation is a publication which had been cited in another. It is not always possible to read the original. It may be unavailable or in a different language. You must state clearly what you have done.

Raabe, E. W. 1952. Über den ‘Affinitätswert’ in der Pflanzensoziologie. Vegetatio, Haag, 4: 53-68. Cited in Southwood 1964, Ecological Methods, Methuen, London. 391 pp.

Institutional references

Some articles and books are the products of a team so authorship might be cited as “anonymous”. However, when all worked at the same place you give authorship to the institution.

ICZN. 1985. International Code of Zoological Nomenclature. International Trust for Zoological Nomenclature. 3rd. edition. Natural History Museum, London. 338 pp.

A translation

Warming, E. 1892. Lagoa Santa et Bidrag til den biologiske Plantegeografi. K. Danske videns K. Slsk. Skr. 6. Copenhagen, Denmark. Portuguese edition: E. Warming & M. G. Ferri. 1973 Lagoa Santa e a Vegetação de Cerrados Brasileiros. Editora Itatiaia Ltda, Belo Horizonte, Brazil, 362 pp.

A chapter in a book

Raw, A. 1992. The forest: savanna margin and habitat selection by Brazilian social wasps (Hymenoptera: Vespidae). In The Nature and Dynamics of the Forest-Savanna Boundary, [Eds. P.A. Furley, J.A. Ratter and J. Proctor] Chapman and Hall, 499-511.

The social sciences system

This system uses the following sequence of information: 1. Author. 2. Title. 3. Name of journal. 4. Volume. 5. Pages. 6. Year of publication.

HEIL, M. Herbivore‐induced plant volatiles: targets, perception and unanswered questions. New Phytologist, v. 204, p. 297-306. 2014.

In the Harvard system the first two pieces of information given are the author’s name and the year the work was published. These are the two that most researchers remember about a reference. The social sciences system places the year of publication at the end. Biologists see this as an unnecessary confusion. No major international biological journal uses the social sciences system.


Writing is mostly about rewriting


After you have finished an adequate draft of your manuscript, leave it for a few days. Then add any new information and ideas you may have and revise the work carefully. The professional will read his manuscript several times before feeling satisfied with it. You will probably rewrite some parts, delete others, and reorganize the sequence of sentences. This is the time to use a dictionary and thesaurus, and correct the grammar.

References to consult for writing

Of the many books on English grammar and style, it is well worth consulting these two classics. Both are available on-line.

Dictionary of Modern English Usage by H. W. Fowler

Plain Words by Sir Ernest Gowers

First person

Some authors never use the first person and some editors never allow it and, for the most part, that is good advice. However, the passive form can mislead the reader. Changing the following examples to the passive voice would be to claim that others had observed the same and writing to avoid that confusion would be ling-winded.

1. If you are the only person who has studied a group of plants in the region, but have not yet published a list of species you could write:

In addition to those specimens recorded in the study area, I have collected four other species of the genus in the vicinity.

2. If you are creating a new system of categorising a group of organisms you might say:

In categorising the 35 species I recognise the following five groups.

Finally, when you are the sole author, using we instead of I merely suggests you are not prepared to face the situation alone. As Mark Twain said:

Only presidents, editors and people with tapeworms have the right to use the editorial “we”.

The Sentence

The following are guide-lines and not rules.

1. Keep the subject and verb together. Whenever possible write directly (subject + verb + complement).

2. Do not include more than one idea in a sentence.

3. Avoid long sentences. Each should be the size that the reader can swallow at one go.

4. Avoid too many commas.

The Paragraph

To help the reader, the text is divided into paragraphs. Each discusses a single topic or aspect of a topic. Construct each one carefully and present them in a logical sequence. Check each paragraph for the following points.

● Do several sentences start with the same word ?

● Are any words repeated too often ?

● If the sentences are similar and monotonous, re-organize a few of them.

● Link the sentences within the paragraph.

● Take care over paragraph length. Editors prefer all paragraphs to be of similar size and common sense tells us that a paragraph of five to ten lines is easiest for the reader to digest.


Sometimes there is a question about which tense to use. State particular cases in the past, like your results. Make general statements in the present.

The winter drought occurs from May to August.

But The drought of 2001 was severe.

The species inhabits semi-deciduous forest.

But The species inhabited a forest fragment in Salvador.

Females lay large numbers of eggs.

But The female laid 260 eggs.

Choosing the right word

Words are tools so take care with them in a work on science. The sloppy use of a word is like using a knife as a screwdriver and damaging the blade so that it cannot be used again either as a knife or as a screwdriver. Beware of homonyms (words with more than one sense); they can confuse the reader.

Use short, simple words

Use technical words if they convey a more specific meaning. Avoid pomposities. Words like phanerogram and pteridophyte simply show that you are trying to impress the reader. Better to write flowering plant and fern. Use only those technical words that are necessary and that the reader will understand.


Avoid fuzzy words. What did the author mean writing: the time the tree flowered ?

a) the occasion the tree flowered

b) the period the tree flowered

c) the duration of flowering

When looking for the right word, it is best to start with the dictionary of synonyms in the Word program.


Innumerable writers on style say the adjective is the enemy of the noun. Use adjectives and adverbs that make the meaning more precise, and not more emphatic. Many adjectives create tautologies. Before using them, ask yourself:

1. Is an adjective needed ?

2. If so, would a more specific one be more appropriate than a general one ?

Using nouns as adjectives

Use a noun in place of an adjective when it clarifies the meaning. There is a big difference between the two following.

Science writing is writing about science.

Scientific writing is writing in a scientific style.

The excessive use of nouns has been called “nounspeak”. It can be difficult to read.

environmental impact assessment study

water resource quality management decision

The following examples are double noun combinations generally accepted in science writing.

tree height

regression line

sample size

transpiration rate

data bank

death rate

scatter diagram

page size

cluster analysis

Some combinations are now accepted as single nouns, like daylight, deadline, honeybee, keyboard, moonlight, sunshine, timetable, keyword.


A homonym is a word with more than one meaning. They are common in everyday speech; a result of adopting words from various languages. Use words with narrower meanings. Here are a few homonyms the biologist will encounter.











Beware of identical nouns, verbs and prepositions. I like Groucho Marx’s remark: Time flies like an arrow. Fruit flies like a banana.

Collective nouns

Various collective nouns are treated as singular or plural depending on the context. These include department, university, government, crowd, company, class and many more. In the following pairs the first indicates unity and the second multiplicity.

The committee announces its decision. The committee members agree on a decision.

The committee are discussing all the questions. Everyone argues.

The class is answering the question. This is clearly a joint effort.

The class are answering the questions. They are sitting an exam.

These options can raise problems for the writer. A number of studies were made. It is pedantic to write A number was made, but then you don’t need to. You can write: Numerous studies were made.

Prepositions and conjunction

Here are three different prepositions. Their meanings differ.

a) The plants flowered after the rains started. This is a straightforward statement, but leaves a question of how long after.

b) The plants flowered when the rains started. This suggests that the flowering and the start of the rains happened together and the two events might be linked.

c) The plants flowered as a result of the rains starting. This is an obvious claim of cause and effect.

Conjunctions can also create problems. It is logistically incorrect to use because when it is not causation.

Incorrect: Many individuals died because of the heavy rains. This statement is not true. There was more water on the ground because of the heavy rain. The floods occurred because the water could not escape. Individuals died because they drowned.

Correct: Many individuals died as a result of the heavy rains.

Do not write and/ or. It demonstrates more about the writer than the research. It is passing the buck to the editor – and he will simply return it because he will not pass it on to the reader.

It is incorrect to think you use between only when two items are involved and among when there are more. Between is a specific relationship often of several pairs. All the predation in a community is between species. Among is a collective relationship. Lizards live among the rocks.

Phrasal verbs

A phrasal verb is a verb plus a preposition. They are very common in spoken English. Try to limit their use in science articles. Most can be substituted by simple, but more specific verbs.


Contractions are used all the time in spoken English. Contractions are not allowed in science writing. You cannot write: can’t, don’t, isn’t. You will notice I have included a few in the present volume, but this is not a research paper. In a research paper you must write: cannot, do not, is not. Do not confuse them with possessive apostrophes which are allowed.


An oxymoron is a combination of contradictory ideas. There are many in everyday English. Avoid them.

exact estimate

small crowd

new classic

found missing

same difference

working holiday


A tautology is repeating a statement or idea with different words. Tautologies are very common. This habit is also called redundancy. They are very common. Avoid them.

temperature values

a definite decision

oval in shape

the true facts

an acute crisis

large in size

height levels

a real danger

green in colour

period of time

equal halves

past history

mutual cooperation

new innovation

future plans

completely filled

continue to remain

often in the habit

Some phrases are tautological.

The two animals discovered the food and both fought over it. Omit both

The diet was supplemented with additional protein. Omit additional.

Hyphens and commas

Using a hyphen between words can alter the meaning.

a) A man eating jaguar. Who is eating whom ? Better to write:

A man-eating jaguar.

b) Three dimensional graphs. There are three graphs.

Three-dimensional graphs. The graphs are in three dimensions.

c) You find a space-bar on your computer keyboard and a space bar in the film Star Wars.

d) A yellow banded wasp. Which is yellow ?

Yellow, banded wasps. The wasps are yellow.

Yellow-banded wasps. The bands are yellow.

A comma can change the meaning. Examples demonstrate the point.

Surprisingly, small birds are aggressive towards hawks. You are surprised at the behaviour of the small birds.

Surprisingly small birds are aggressive towards hawks. You are surprised at the size of the small birds.

a large forest species is ambiguous.

a large, forest species suggests the species is large and occurs in forest.

a large-forest species suggests the species is restricted to large forests.


The word none is a contraction of not one. It is incorrect to write none are. It should be none is.

Take care presenting statistical data

In the following example, the first sentence (a) is vague. Better to write b.

a. There was a direct relationship between nutrition and egg production (P <0.001; r = 0.84).

b. Egg production increased directly with an increase in the amount of food consumed (P <0.001; r = 0.84).

Citing figures and tables

You must refer to all the figures and tables in the Results section. There are good and poor ways to cite them. The accepted way is to refer to some interesting information contained in them and then cite the table or figure in brackets afterwards. This is similar to the best way to cite references in the text (see below).

Poor: Table # shows results on the numbers of visits.

Better: The number of visits varied (Table #).

Poor: The higher rate of predation on sunny days is shown in Figure #.

Better: The rate of predation was higher on sunny days (Figure #).

Where to cite the references

The rule is to cite references only in the Introduction and the Discussion. However, there are two exceptions.

a) When a detailed description of the study area has already been published, you might cite that reference in the Study Area rather than re-describe it.

b) When you used a method described in detail in another publication, you might cite the reference in the Methods section.

Even so, try to put these references in the Introduction.

One strict rule is that you NEVER put references in the Results section.

How to cite references in the text

Present the references in a style that flows easily. Write about the subject and place the references in brackets. The reader is probably more interested in the data than in the author so he can ignore the reference if he wishes. In the following examples, the first sentence is poor and the second better.

a. Fontana (2000) reported that the bees secrete antibacterial compounds.

The bees secrete antibacterial compounds (Fontana 2000).

b. The findings of Jones (1968) suggest that unseasonal rainfall may be a major mortality factor.

Apparently, unseasonal rainfall was a major mortality factor (Jones 1968).

Acceptable exceptions are when you wish to draw attention to the contribution a particular author made to the subject.

1. Warming’s volume (1892) on the cerrado vegetation of Lagoa Santa was a major contribution to our understanding of a large neotropical ecosystem.

2. Charles Darwin’s classification of barnacles (1851-54) is an important reference even today.

Separation of Results and Discussion

The separation of the Results and Discussion is the most important aspect of the structure of a paper. The reader must be in no doubt about your contribution and the simple way to make this clear is to separate these two parts. Even in a short paper, the two should be in separate paragraphs. However, this is one of the biggest difficulties an author has in organizing his manuscript.

● In the Results section you present your contribution to science.

● In the Discussion you relate your findings to the work of others in the field. That is why you never cite references in the Results section.

Where the two are mixed, it demonstrates that the author did not think clearly about how to present the work. He is leaving the reader to try to decide who did what. Remember, the author is not important; the reader is. Even so, some journals accept the Results and Discussion sections combined and even mixed up.

Let us see how to present information in the two sections. In the Discussion try not to “discuss” decimal places; it is best to round off numbers. Do not cite your figures and tables in the Discussion.

The data in the following example are from an article on tomato pollination; the words are mine.

Results: Larger fruits (diameter 51.1 mm) contained 183.9 seeds per fruit, while smaller fruits (diameter 46.6 mm) contained only 59.6 (Table #).

Discussion : An increase of 15% in fruit diameter tripled the number of seeds per fruit.

Imagine you collected grasshoppers in plots.

In the Results you state: There averaged 120 individuals per 100m2.

In the Discussion, you can extrapolate the information provided it is clear that you are doing so. The number is equivalent to 12,000 grasshoppers per hectare. However, you cannot say: There were 12,000 grasshoppers per hectare.

British English and American English

Most scientific journals accept manuscripts in British English or American English, however, you cannot mix the two. Major differences between them are in the substitution of words and in spelling. Some Americanisms are better because the British equivalents are homonyms, some are worse for the same reason.


The harsh truth

Researchers have time to read very few of the thousands of scientific articles published in their field each year. Thousands of people will glance at the title of your paper, dozens will look at the abstract, but very few will read the paper itself. Therefore, the title must persuade the reader to read the abstract, and the abstract persuade him to turn the page. It is up to you.


There are basic criteria to follow when composing a title.

1. It must explain the subject of the paper.

2. It must be clear and concise.

3. It must be as brief as possible and not include superfluous words.

4. Unless the work is on taxonomy and the information is relevant, do not include the author of the species or the date of its publication in the title. Even papers on taxonomy rarely include authors’ names in the title.

Examples of titles

Cockerell, T. D. A. 1933. Descriptions and records of bees. -CXLII. Annals and Magazine of Natural History, Series 10, 11: 372-384.

Uninformative. The author published 141 previous papers on bees. It also suggests that the topic might be on identification.

Shaw, E. J. F. and A. R. Kahn. 1931. Studies on Indian pulses. Dept. Agric. Ind. Mem. Bot. Ser. 19.

Barely sufficient. Of course it was a study, but on what ?

Fleming, T. H. 1979. Do tropical frugivores compete for food ? American Zoologist 19: 1157-1172.

Short but sufficient. The reader of this journal should know that a frugivore eats fruits.

Glance, W. E. 1984. Food and habitat use by two sympatric Sciurus species in Central Panama. Journal of Mammalogy 65: 342-347.

Specialised for the journal. Readers of this journal will know that Sciurus is a genus of squirrels. This title would not be adequate in an ecology journal.

Hutchinson, G. E. 1941. The paradox of the plankton. American Naturalist 95: 137-145.

Janzen, D. H. 1977. Why fruits rot, seeds mold, and meat spoils. American Naturalist 111: 691-713.

Snappy titles. Two intriguing titles on interesting subjects.


Journals request a list of keywords. The title and keywords together are used by internet search engines so you need not repeat words already in the title. In the two following examples, the keywords marked in bold-italic are already in the title.

Brown, J. H. and A. Kodric-Brown. 1977. Turnover rates in insular biogeography: the effects of immigration on extinction. Ecology 58: 445-449.

Key words: Arizona; biogeography; colonization; extinction; insular biogeography; island; turnover.

Terborgh, J., R. B. Foster and P. Nuñez V. 1996. Tropical tree communities: a test of the nonequilibrium hypothesis. Ecology 77: 561-567.

Key words: Amazon; community ecology; meander belt; nonequilibrium hypothesis; Peru; primary succession; tropical forests; tropical tree communities.


A final revision

Check carefully all the points in the following list.

1. Are any data discussed in the Results or have you placed results in the Discussion ?

2. Recheck the division between Introduction and Discussion. Check the publications you want to cite before and after presenting your own data.

3. Are all the references cited in the text in the list of References and vice versa ?

4. Are any data repeated in a table and a figure ?

5. Recheck all the dates, symbols and numbers and citations of figures and tables. Some of these change as you edit the text.

6. Some authors do not like to use the paragraph marker while writing, that is the ¶ symbol. Before submitting your manuscript, use this to clean up the double spaces between words, line-spacing and other errors.

7. Run the spell and grammar checker.


When you have finished all these tasks, pass the manuscript to discerning colleagues for critical reading. Revise the manuscript bearing in mind their comments, and incude any new thoughts of your own.


Acknowledge those who helped. Provide their names and addresses for others to contact them. Include those who gave you the idea, who discussed it with you. Plagiarism is presenting someone else’s idea as though it were your own. It is often just forgetting to say who helped. As Kate Turabian (1996: 74) wrote, Failure to give credit is plagiarism. Say who showed you the site, plant, animal or the method, provided specimens, loaned equipment and who arranged the financial support. Often colleagues help with advice on statistical analyses and identification of specimens. If you conducted the work on private land, say who gave you permission. Technicians and students who worked for you like to see their name in print too. Acknowledge the support of financing bodies and the institution where you worked which provided space, transport and labour.


The authors are those who helped design and conduct the investigation, in the analysis of the results and to write the manuscript. The sequence of the names of the co-authors on the publication should be decided beforehand and it should reflect the proportion of the work each will do.

A second issue is the question of how many authors there might be on a paper. There is no rule against including numerous authors in one article and much research is done by teamwork. However, the description of a new species of tapir was published recently. Its name is Tapirus kabomani Cozzuol, Clozato, Holanda, Rodrigues, Nienow, Thoisy, Redondo and Santos 2013. Every time the species is cited in a scientific paper the names of all eight authors should be cited.

Some papers are astounding. The example below has 176 authors averaging 16 per published page; it may be a record. It is reasonable to ask if the authors compiled enough data for more than one publication so they could split up the work and authorship.

R. J. Mural, M. D. Adams, E. W. Myers, H. O. Smith, G. L. G. Miklos, R. Wides, A. Halpern, P. W. Li, G. G. Sutton, J. Nadeau, S. L. Salzberg, R. A. Holt, C. D. Kodira, Fu Lu, L. Chen, Z. Deng, C. C. Evangelista, W. Gan, T. J. Heiman, J. Li, Z. Li, G. V. Merkulov, N. V. Milshina, A. K. Naik, Rong Qi, Bixiong C. Shue, A. Wang, J. Wang, X. Wang, X. Yan, J. Ye, S. Yooseph, Q. Zhao, L. Zheng, S. C. Zhu, K. Biddick, R. Bolanos, A. L. Delcher, I. M. Dew, D. Fasulo, M. J. Flanigan, D. H. Huson, S. A. Kravitz, J. R. Miller, C. M. Mobarry, K. Reinert, K. A. Remington, Q. Zhang, X. H. Zheng, D. R. Nusskern, Z. Lai, Y. Lei, Wenyan Zhong, A. Yao, Ping Guan, Rui-Ru Ji, Zhiping Gu, Zhen-Yuan Wang, Fei Zhong, Chunlin Xiao, Chia-Chien Chiang, M. Yandell, J. R. Wortman, P. G. Amanatides, S. L. Hladun, E. C. Pratts, J. E. Johnson, K. L. Dodson, K. J. Woodford, C. A. Evans, B. Gropman, D. B. Rusch, E. Venter, M. Wang, T. J. Smith, J. T. Houck, D. E. Tompkins, C. Haynes, D. Jacob, Soo H. Chin, D. R. Allen, C. E. Dahlke, R. Sanders, K. Li, X. Liu, A. A. Levitsky, W. H. Majoros, Quan Chen, A. C. Xia, J. R. Lopez, M. T. Donnelly, M. H. Newman, A. Glodek, C. L. Kraft, M. Nodell, F. Ali, H-J. An, D. Baldwin-Pitts, K. Y. Beeson, S. Cai, M. Carnes, A. Carver, P. M. Caulk, A. Center, Y-H. Chen, M-L. Cheng, M. D. Coyne, M. Crowder, S. Danaher, L. B. Davenport, R. Desilets, S. M. Dietz, L. Doup, P. Dullaghan, S. Ferriera, C. R. Fosler, H. C. Gire, A. Gluecksmann, J. D. Gocayne, J. Gray, B. Hart, J. Haynes, J. Hoover, T. Howland, C. Ibegwam, M. Jalali, D. Johns, L. Kline, D. S. Ma, S. MacCawley, A. Magoon, F. Mann, D. May, T. C. McIntosh, S. Mehta, L. Moy, M. C. Moy, B. J. Murphy, S. D. Murphy, K. A. Nelson, Z. Nuri, K. A. Parker, A. C. Prudhomme, V. N. Puri, H. Qureshi, J. C. Raley, M. S. Reardon, M. A. Regier, Y-H. C. Rogers, D. L. Romblad, J. Schutz, J. L. Scott, R. Scott, C. D. Sitter, M. Smallwood, A. C. Sprague, E. Stewart, R. V. Strong, E. Suh, K. Sylvester, R. Thomas, N. N. Tint, C. Tsonis, G. Wang, G. Wang, M. S. Williams, S. M. Williams, S. M. Windsor, K. Wolfe, M. M. Wu, J. Zaveri, K. Chaturvedi, A. E. Gabrielian, Z. Ke, J. Sun, G. Subramanian, J. Craig Venter. A comparison of whole-genome shotgun-derived mouse chromosome 6 and the human genome. Science (31 May 2002) 296: 1661-1671.

Which address ?

The question over which address to include is simple. You must use the one where you conducted the research. For example, a paper extracted from a thesis must cite the address of the institution where the thesis was defended. If you moved, cite your new address as a second one and indicate that is the one for correspondence.

Selecting the journal

In general, the wider the audience the simpler the presentation. The non-specialist will appreciate your providing additional information in simple ways. The reader who is wondering about the meaning of a word is not paying full attention to the information. See how you might use scientific and vernacular names.

a) For a journal of entomology you investigated a species of Curculionidae.

For a journal of ecology you studied a weevil.

b) In a specialized journal you might say you studied a group of Hirundinidae.

In a more popular publication you studied swallows.

c) Similarly, in a more general journal explain technical terms like arbovirus (arthropod borne virus).


When an editor receives a manuscript that does not follow his journal’s instructions, he will return it. Before submitting yours, read and follow the instructions carefully. Present it in the exact format of the articles published there.

The editor uses several criteria to assess a manuscript (Table 6). If it passes, he will send it to the reviewers. Notice that everyone will examine the methods and how they were employed.

Table 6. Criteria for evaluating a manuscript for publication in a scientific journal

What the editor checks

1. The subject must be appropriate for the journal.

2. The manuscript must be presented in the format of the journal.

3. The method employed must be scientific.

4. The manuscript should be of a suitable length. If it is too long, he will suggest you reduce it.

What the reviewers evaluate

1. The author must demonstrate that he is familiar with the latest research on the subject.

2. The method employed must be unquestionably scientific.

3. The procedures followed to collect the data must be clearly stated.

4. The reasoning must be correct and clear.

5. The statistical analyses used must be appropriate and the results clearly stated.

6. The findings must be good enough to interest the readers of the journal.

More than one journal

While a manuscript is in the hands of the editor of one journal you must not send it to another until you have the decision of the first. If you published a preliminary note earlier then state the fact in the later paper.


The work rarely finishes when a journal accepts the manuscript. The editor will return it with the reviewers’ comments. Proof correction is generally a pleasant job because you know the work will appear in print. This is the last opportunity you have to correct errors, so work very carefully.


If the manuscript is rejected it is not the end of the world. The editor may suggest another journal. Anyhow, the editor and reviewers will have justified their decision – though you may not agree with them. Sometimes the criticisms of the various referees are so different that you wonder if they all read the same manuscript. Having said that, you should consider it a privilege to receive the comments of experienced scientists. Put them to good use. You should be able to improve the manuscript through their constructive criticism.


Improving your style

Everyone who writes well, reads a lot. When reading, do not just think about the subject of the paper; pay attention to the presentation as well. To improve your own style you should read articles in the better science journals and also in science magazines and newspapers aimed at the lay reader. When writing, try to emulate what you enjoy reading. Remember, good writing is best learnt through practice; the more you write, the better you get.

Evaluation of the content

Some papers have elegant methods and others have fascinating results – and some have both. Read papers with the following plan in mind.

1. Is the author well acquainted with the subject ?

2. Was it a difficult subject to investigate – like working under water or following jaguars ?

3. Check the method carefully. Is it convincing ?

4. Were the data collected scientifically ? Check the data before accepting them.

5. Were the data analysed by appropriate methods ?

6. Was it a difficult subject to write about ? For example, did it draw information from many different sources ?

7. Are the results biologically significant as well as statistically significant ?

8. Did the author reply adequately to his question ?

9. Was the work worth doing ? Was the paper worth publishing ? Was it worth reading ?

Exercise on the evaluation of published articles

Making critical analyses of scientific papers is an excellent exercise. When you read an article, give a score to the various parts. A table with nine components, totalling 150 points is useful (Table 7). The exercise will clarify what parts you like and dislike about the papers. You may prefer the Introduction of one and the Results of another. Compare articles in good and not so good scientific journals so you will see what to emulate and what to avoid.

Table 7. Suggestion for the evaluation of published articles

Evaluating your own manuscript

You will soon become adept at evaluating the papers of others. You will then be able to evaluate your own manuscript more objectively. Run your own manuscript through the nine components in Table 5. The result will surprise you.

[]15. LETTER

Letter to the editor of a journal accompanying a manuscript being submitted for publication

Departamento de Ciências Biológicas,

Universidade Estadual de Santa Cruz,

Ilhéus BA,

Brazil 4565-000

[email protected]

1 April 2016

Dr J. Smith,

Editor: Journal of Tropical Insects,

Department of Ecology,

University of Toontown,

Toontown, CA, U.S.A.

Dear Dr Smith,

I should be grateful if would you consider the enclosed manuscript for publication in your Journal. It is entitled:

A founder-controlled assemblage of neotropical social wasps, and its recent severe fall in numbers

Attached are the names and addresses of three specialists who might be prepared to review the work. The first of them has already seen a draft of the manuscript.


Anthony Raw

[]16. THESES

Writing a thesis

Your thesis will have more effect on your career than anything else scientific you will write. Nonetheless, many theses are poorly structured and written. This is partly because the amount of data presented can overwhelm the writer and partly it is a lack of training in writing.

A student who writes well will spend at least three months writing a masters thesis and six on a doctoral thesis. In my experience the student who thinks he will write a masters thesis in two months will spend four.

There are differences between a scientific paper and a thesis. The reader of a paper presumes the author knows how to conduct research. A thesis is the product of a training session. It is to demonstrate that the student has learnt how to conduct research so it should include a more detailed explanation of the methods than would normally be included in a paper.

Evaluating the thesis

It might be argued that the members of an examining board of a thesis are not as demanding as the editor of a journal. Don’t count on it. Better to write a good thesis. It will help to check through a few that your supervisor considers to be the best presentations.

Four aspects of theses are assessed.

Reading. It is wise to show that you have consulted all the important references on the subject even if you must manipulate the text to include them. Ensure that you have included the examiners’ publications.

Quantity of data. The best criterion to assess the amount of information required is that there should be enough for one published paper.

Quality of data. Maybe the supervisor is more responsible over questions of quality. Criticisms are mostly about method, whether the data have been collected by a method that allows scientific evaluation.

Presentation. Some theses are poorly written, but the present volume might help in this respect.

Here are some of the more common problems.

● Often the Introduction, Results and Discussion are mixed up. In particular, data are discussed in the Results section.

● In the Introduction there is no need to start with pages of general information better placed in a text book.

● A common fault is to include far too many sections. I have found 9 Introductions, 5 Methods and Study Areas, 14 Results and 14 Discussions in a single thesis.

You might include an Appendix to place the original data and, depending on the size of the thesis you might compile an index.

The thesis defence

The human brain starts working the moment you are born and never stops until you stand up to speak in public

Comment of Sir George Jessel, a British judge.

In the defence, you need not follow the written thesis exactly. After all, the examiners will have read the entire opus. Most defences are public; a euphemism for saying that you can invite friends and family. Direct the presentation to the non-specialist audience. Explain the subject clearly to show that you understand it well. Use the most interesting figures and tables to show what is so fascinating about the findings. Provided that you include the relevant information and arguments, the examiners will be happier with that approach. You will be on stage so a rehearsal in front of friends is a wise move.


Brooks, C. and R. P. Warren. 1949. Fundamentals of Good Writing. Harcourt, Brace & World, Inc. New York. 523 pp.

Endler, J. A. 1992. Editorial on publishing papers in Evolution. Evolution, 46: 1984-1989.

Fowler, H. W. 1968. A Dictionary of Modern English Usage. Second Edition. Oxford University Press. 725 pp.

Gowers, E. 1969. The Complete Plain Words. Penguin Books Ltd., Harmondsworth, England. 272 pp.

Hake, R. L. and J. M. Williams. 1981. Style and its consequences: do as I do, not as I say. College English 43: 433-451.

ICZN. 1985. International Code of Zoological Nomenclature. International Trust for Zoological Nomenclature. 3rd. edition. Natural History Museum, London. 338 pp. http://www.iczn.org/

Janzen, D. H. 1976. Why bamboos wait so long to flower. Annual Review of Ecology and Systematics, 7: 347-391.

Moore, R. 1994. Writing as a tool for learning biology. BioScience, 44: 613-617.

Turabian, K. L. 1996. A Manual for Writers of Term Papers, Theses, and Dissertations, 6th edition. Chicago, 308 pp.

Valentini, W. C. 1998. Cientista também precisam ter estilo. Jornal do Conselho regional de Biologia, Jaboticabal, São Paulo. p. 7.

How to Write a Scientific Paper

The researcher must write papers. There is little point in conducting research if you cannot communicate the results. Also, it is a pleasure to see your work in print and know that other scientists will read it. Even so, nearly half of the manuscripts received by some international biological journals are rejected because they are poorly written. That is more than those rejected on scientific grounds (Endler 1992, Moore 1994). This volume explains how to write a scientific paper. It contains the secret to success presented in guidelines, advice and some general rules. However, there is no short cut for learning to write well. It takes organization, practice and constructive criticism from your peers. This helps in the organization.

  • Author: Alan Halley
  • Published: 2016-11-18 19:50:16
  • Words: 14008
How to Write a Scientific Paper How to Write a Scientific Paper