How to write scientific manuscripts
in food structure research

Miloslav Kaláb


The author has published over 140 scientific papers on various topics in various disciplines. Since the seventies, he has specialized in food structure studies at Agriculture and Agri-Food Canada in Ottawa as a principal research scientist. He is currently on staff of the Food Research Program, Agriculture and Agri-Food Canada, in Guelph, Ontario, but he works in Ottawa.

M. Kaláb has a wealth of experience both as an author and an editor. From 1982 until 1994 he served as the Editor- in- Chief of Food Structure, an international scientific journal which he established with 3 American food scientists. Now he is a member of the editorial board of Food Science and Technology (LWT) published by Academic Press. He reviews manuscripts also for other journals, including the International Dairy Journal published by Elsevier and for 2 years he served on the editorial board of Scanning published by the Foundation for Advances in Medicine and Science (FAMS).

In the sixties and seventies, he was a freelance abstractor for Chemical Abstracts and prepared over 16,000 abstracts mostly from Russian agricultural, biological, and medical literature.

On this Web page, M. Kaláb shares some of his ideas on how to write scientific manuscripts. He may be contacted at

Scientific papers are reports on findings which have scientific value. Since only a few phenomena may be described without having to carry out experiments, most scientific papers deal with hypotheses, which the authors challenge by experiments, a description of such experiments and their results or findings, and a discussion of what the findings mean, and where they may be useful.

Productivity of a scientist may be assessed according to various criteria. For many years, one of them used to be the number of scientific papers published. A slogan publish or perish reminded the scientists what their true objectives should be. Another criterion is the number showing how many times other scientists have referred to the publications of the scientists being assessed. This information is available in the Citation Index. Neither criterion evaluates scientific productivity correctly - many papers may have a low scientific value and a major discovery may be so far ahead of the times that very few peers may refer to the original discovery.

Like artists who exhibit their pictures or sculptures or have their music or plays performed, scientists strive to publish the results of their work. Most are proud when their papers appear in journals and when they receive reprints. It is a pleasant feeling of an accomplishment because a lot of work was involved before the reprints may be enjoyed. In contrast, other scientists consider publishing their findings to be nothing extraordinary - it is part of their lives: they would not feel that they have accomplished anything and contributed to the treasury of scientific information unless their experiments and the results are published.

Some scientists have an ability similar to composers or painters: they plan their experimental work to examine their ideas in a way that looks like clockwork, where all components are connected in a logical way to each other. These scientists see their next manuscript as a short report or a full-size paper illustrated with many figures to fit the idea and experimental results. They break down a large topic into a series of smaller, more easily manageable tasks and then they publish one part after another. Attempts to keep partial results unpublished so that a series of manuscripts could be prepared after all the experiments have been completed can be disastrous - either someone else will publish a paper on a similar subject or the topic will become obsolete because of developments in another direction. Another slogan suggests: work, finish, publish. It is attributed to Faraday.

At the beginning, there is an idea - a question: 'What happens, if...?' The response may be: 'Let's establish the factors which control heat-induced gelation in milk proteins and the microstructure and mouthfeel* of the resulting gels'. Or: 'Let's examine the relationships between mouthfeel* and microstructure in wieners and in heat-induced milk gels', etc. How did I arrive at such questions? Thirty years ago, I was hired as a food scientist to develop a wiener-like product using skimmilk powder as the major ingredient and have it ready for marketing by 1973. Heat-induced gelation was suggested as the preferable manufacturing procedure.
*Mouthfeel is a new term for the feeling of food in the mouth. There are differences in mouthfeel even between similar foods, e.g., Cheddar and Eidam.

Many preliminary experiments showed that although it was possible to produce something that resembled a hot dog as far as its shape, colour, and smell was concerned - the mouthfeel of that product was horrendous and resembled that of a bad cheese. As a newcomer to food science, I was advised by a senior colleague to ask a microscopist for assistance. The difference in micrographs between my product and a real meat-based wiener was staggering: the real wiener consisted of muscle fibres, fat globules, and little pockets of gelatin (which on heating turned into juice), but my 'milk wiener' consisted of minute globules - casein micelles. No wonder that it was brittle - there was nothing in my product that could have had elastic properties!

This finding aroused my interest in food structure. I learned electron microscopy and started my new career in a new discipline - in a foreign country. I needed to be successful, to have excellent productivity, to show that I am an achiever. At that time, personal computers were introduced on the market but only administrative workers - not the scientists - were allowed to use them for word processing in the establishment where I worked. So I bought my own portable Osborne CP/M computer and brought it to the laboratory. A new era in writing scientific manuscripts had begun for me. Maybe you will find my experience and suggestions interesting.


When preparing experiments to examine the validity of your hypothesis, consider the manner in which you will report your results. Start writing your report (manuscript) even before your experiments have begun. This will allow you to be efficient. Unless you can keep all the information in your head, use your computer to store it for future use. Put everything in writing. You may need to change your experimental schedule based on the results being obtained. The initial plans are not rigid - they need to accommodate preliminary findings. Scientific work is creative. Some paths may not lead to success. Current progress needs to be frequently re-evaluated. Yet, your objective is your goal - you may change directions to get to it. - When you arrive at it and complete your experiments, it will take only days rather than weeks to complete the manuscript - if you find all information in the proper place.

Suggest a tentative Title. The computer allows you to change it easily if you find a better fit. Write your tentative Introduction while doing literature search. Has anybody else done a similar study? Can you use any principles and build on them? Clearly formulate your objectives. Arrange your notes as if you were already writing a manuscript. Also collect information for the Materials and Methods section: take notes of all chemicals and instruments used. Include a title Results and Discussion. This section will occupy you after your experiments have been completed, so leave this section empty. However, list bibliographies in the References section. Characterize most entries with their most important contributions, e.g.: Zittle, C. A., Thompson, M. P., Custer, J. H., Cerbulis, J.: k-Casein-b-lactoglobulin interaction in solution when heated. Journal of Dairy Science 45(7), 807-810 (1962). [Interaction of beta-lcgb with kappa-cas in heated milk - complex blocks reactive sites on casein micelle surface. Probably affects coagulation of milk that had been heated.] - If you have found interesting ideas in other people's papers, take notes about them in the Discussion section. You may store the same information in different sections, so 'copy and paste' using your computer. Eventually all unwanted information, like the notes in the References section, will be deleted. If you intend to use experimental procedures someone else has devised, describe them in the Materials and Methods section. Have you observed something interesting to happen? Always take notes.

Almost all papers on food structure research are illustrated with micrographs. The technology of producing them is rapidly changing and darkroom techniques are being replaced by computer processing. In 1997, the Ontario Science Teachers' Magazine The Crucible published my article entitled From milk to yogurt and cheese which was illustrated with 6 micrographs. No paper was used and both the text and the micrographs were sent by e-mail. This development may also be anticipated for scientific journals. Computer storage makes it easy to describe the most important features of each micrograph and mark them as suggested elsewhere.

Organization of a typical scientific manuscript:

Title Clearly indicates the contents of the manuscript
Abstract or Summary Is short but contains all essential information about the objective of the study, basic information about the experiments and sufficient information about the results to make the reader eager to read the entire manuscript. Each sentence must add new information that is clearly stated. Sentences such as 'The effects of A on B are discussed in detail' are useless, particularly if the title is 'Effects of A on B'. The Abstract is usually written last by the authors but it appears in the journal right below the title. It should contain all important information (numerical values). There are no more freelance abstractors who would improve this section - most reference journals reprint abstracts without changing them.
Example: Cold-stage scanning electron microscopy (cryo-SEM), transmission electron microscopy (TEM), and light microscopy were used to examine the effect of NaCl (2%), MgCl2, CaCl2, KCl, LiCl (calculated ionic strength, IS = 0.43) and 1.5% NaCl (IS = 0.26) on the structure of comminuted raw meat batters (fat added) and comminuted meat mixes (no fat added). From: A. Gordon and S. Barbut: The microstructure of raw meat batters prepared with monovalent and divalent chloride salts. Food Structure 9, 279-295, 1990).
Introduction Is closely related to the subject of the study. It explains what is known about the subject from earlier work by various authors including the author (-s) of this manuscript and it refers to such works, listing them in the References section. If the subject has little history, it explains why there is a need for this study and justifies it. The justification is supported by literature references. The Introduction ends with objectives, e.g., 'The objective of this study was to identify and characterize factors which make Mozzarella cheese meltable'.
Materials and Methods
Experimental Part
The objective of this section is to make it possible for interested readers to repeat the experimental work and to reproduce the results obtained with those reported by the authors. It lists the materials (chemicals, raw materials, ingredients, instruments) used to carry out the experiments. Unless the experiments have been described elsewhere, the description should go into sufficient detail. For example, it identifies
  • the origin of the food under study ('fresh pooled milk from the university Jersey herd', 'commercial pasteurized skimmilk', 'Cheddar cheese manufactured by the procedure described below', etc.),
  • sample preparation techniques ('cubes, approximately 20x20x20 mm, were excised from the centre of 3000-g blocks, and were subsequently trimmed to 1x1x10 prisms),
  • and statistical analysis of numerical data.
  • Results and Discussion This part may alternatively be divided into two separate parts. The text cannot be prepared in advance before the experiments are completed. It is the most important part - the reason why the work was initiated. It is organized in a logical way - from general findings to details. Scientific language is used here at its best.
    Examples: Frying the tofu slices in oil for 5, 15, and 30 min. increased their fat contents to 4.1±0.5, 8.3±0.9, and 12.4±1.1%, respectively. --- Mealy and firm potatoes differed clearly with respect to intercellular contact, cell shape, and cell surface. These differences could be explained by different breakdown of (i) the middle lamella or (ii) the cell wall, or (iii) by the starch content of the potatoes. Last example from: J. T. van Marle et al.: Cryo-scanning electron microscopy investigation of the texture of cooked potatoes. Food Structure 11, 209-216, 1992).
    Conclusion Shows the ability of the author (-s) to observe and to creatively link individual observations to provide a coherent story about the study and its meaning and benefits.
    Acknowledgments provide an opportunity to thank your technician for assistance, your sponsors for a grant or other contributors of materials, and even your internal reviewer for comments.
    This part is helpful to other researchers by directing them to important sources of information. However, it is even more important to the author (-s) of the manuscript. If you have searched the literature thoroughly, you will probably not be informed by the reviewer of your manuscript that a similar paper had been published 15 years ago. Or you will not be notified that similar findings were found 5 years ago to be artifacts from a low freezing rate or a high air humidity. A proper literature search helps the author (-s) to better understand the subject and to state attainable objectives. It is not a good idea to list as many of your own papers as possible in this section to show the scientific community that you are the top expert in this field.
    Tables Provide accurate information in numerical form. However, it makes no sense to use too many digits. Do not list the water (fat, starch) content in your samples at 45.40567% if you realize that the difference between 45.4 and 45.5 is less than 0.25%. Subject your results to statistical analysis. Are the differences statistically significant? At what level? Under what conditions? Is it really important to show tables rather than diagrams? Trends (increase, decrease, fluctuations) may be more easily grasped from diagrams.
    Legends to figures
    Figures may consist either of diagrams or photographs (micrographs). There are many ways to process analytical data into the form of diagrams using line, bar, or pie charts or other graphics.
    Describe the figures in sufficient detail so that there is no need to read the manuscript to understand what the figures show.
    Micrographs should contain micrometer markers (bars) to indicate magnification (e.g., Bar: 10 µm). An alternative way is to identify the value of the image width (Image width = 50 µm). Scientific journals provide instructions on this subject. Structures (components) in the micrographs are identified by arrows (small or large, short or long), asterisks, or letters ('small arrows point to lactobacilli', 'F: fat globules').

    The suggestion to start writing the manuscript before the experiments are carried out is not as outrageous as some old-fashioned scientists would believe. Reading the Introduction before the experimental work on the project is completed is useful. Are you satisfied with the justification? Read your Materials and Methods section while you are carrying out your experiments. Have you included all important information?

    Above all, read good scientific papers and learn from them how to set up experimental work and how to describe it. Learn to condense information into paragraphs which have impact. Also learn from bad papers - analyze them and avoid what you dislike. Writing scientific manuscripts is part science and part art. We all must learn to communicate properly.


    Interestingly, well-planned and executed experimental work is usually well presented whereas poorly written manuscripts reflect deficiencies in both experimental work and writing. There are exceptions, particularly manuscripts received by journals (those which are published in English) from foreign countries from authors who are not proficient in English. However, many inappropriate terms may be found even in manuscripts written by anglophone scientists. There is no 'electronic' microscopy but there is 'electron' microscopy because it is based on a beam of electrons which is used to magnify the image of the sample. During sputter coating, the sample is not 'sprayed with a fine mist of gold' but is 'coated with a thin layer of gold', a platinum-palladium 'alloy' rather than a 'mix' is used for that purpose.

    A short time ago, I reviewed a manuscript where the authors extracted cheese samples, each 40 mm3, in a single step using chloroform extraction for 5 minutes. In the micrographs, they found oily layers and fat globules. They were unable to explain this phenomenon, because, in their words, the samples had been extracted with chloroform. The conclusion was not to use chloroform because it did not remove fat quantitatively. The authors were right - no lipophilic solvent removes fat quantitatively unless the extraction is repeated several times... The size and compact nature of the sample must be taken into consideration when fixing, dehydrating, extracting, or impregnating foods such as cheeses, particularly low-fat cheese, cured meats etc. The most disturbing fact was the ease at which the authors arrived at a wrong conclusion - without adequately exploring what had happened with their samples.

    It is not possible to mention too many examples of deficiencies in this section, so only a few have been tabulated.

    Too broad:
    Microstructure of cheese
    Microstructure of low-fat Cheddar cheese: An optical microscopy study
    Effects of manufacturing conditions on the quality of bread Effect of dough kneading on rye bread loaf volume
    Sentences vague:
    This paper deals with an investigation of manufacturing conditions on the quality of bread.
    Sentences more appropriate:
    The volumes of bread loaves were assessed from photographs of slices taken in the middle of the loaves.
    It seemed that there was a trend towards an increase in the bread volume as kneading of the dough was increased. The loaf volume was increased by 5.2% (P<0.1), 7.6% (P<0.05), and 8.9% (P<0.01) following an increase in the kneading time by 10, 25, and 40%, respectively, as compared with the control samples.
    Unrelated to subject:
    Cheddar cheese is popular in North America and so is Mozzarella cheese which is used, along with string cheese, to make pizza. Recently, the production of Mozzarella cheese surpassed that of Cheddar cheese and the production of other cheeses was also increased, for example, low-fat cheeses, which are also becoming popular. There is almost no information on the microstructure of low-fat cheeses, e.g., low-fat Cheddar cheese, as studied by electron microscopy.
    Related to subject:
    Cheddar cheese is popular in North America but increasingly consumers are demanding cheeses with a reduced fat content. Low-fat Cheddar cheese, however, has a markedly different mouthfeel from full-fat cheese. Several authors [3-7] have pointed to correlations between mouthfeel and microstructure in full-fat cheeses such as Mozzarella [3, 4], Ricotta [5], and Brick [6, 7], where the dimensions of fat globules play an important role. Establishing structural factors which control mouthfeel in low-fat Cheddar cheese would contribute to a better understanding of mouthfeel in cheeses.
    Materials and Methods
    Insufficient information:
    Cheese samples taken from the large blocks were fixed with glutaraldehyde. Then they were washed with a phosphate buffer and subjected to secondary fixation with a 2% solution of osmium tetroxide.
    Proper information:
    Cheese cubes, 1x1x1 mm, were fixed in 20 mL of a 2.5% glutaraldehyde in 0.05 M phosphate buffer, pH 6.85, at 20°C for 4 h. There were 5-6 prisms per vial and the vials were maintained at a steady slow rocking motion.

    All information in each manuscript is important but some parts are more important than others if the authors wish that their work be recognized as a valuable contribution. Reference journals such as Food Science and Technology Abstracts, Chemical Abstracts Service, Current Contents etc. reprint the titles and many also reprint the abstracts.


    Most authors believe that their manuscript is written in an excellent way and that it will probably be published as is. They are very rarely right. We all are so 'impregnated' with our scientific jargon and we think about our 'baby' (the manuscript) so often and love it so much that we frequently suffer from lack of self-criticism. Someone else's look at our manuscript will confirm the truth of this statement. Even the 'most scientific' text should be understood by well educated people who are not necessarily experts in our field. When I was younger and my technician did not understand my text, I explained to her what it is supposed to mean. I realized quickly that she would be the only reader who would obtain personal explanation from the author. Thus, I rather changed my text so that my technician would understand it without any need for personal explanation.

    Peer or internal reviewing is the best way to improve the quality of the manuscript and to save the author from an embarrassment of devastating comments from journal reviewers. Regrettably, however, it is practiced by few authors or research establishments. I was impressed by an American food science department at a relatively small university, where the chairman asked the authors to explain their manuscript to other staff members and to get them involved in selecting the best micrographs for publication. A seminar for graduate students followed a few days later.

    As an editor, I have received some poorly written manuscripts with many grammatical and spelling errors arrive from food science departments. In these cases, it is as if the chairperson signed the letter of submission evidently without bothering to read a single line with the exception of the title. Thus, the precious time of several people (reviewers and journal editors) is wasted and the concluding comments often are:

  • This manuscript would have to be revised and re-submitted for reviewing.
  • This manuscript is not suitable for publication. It cannot be revised because there are principal flaws in experimental design.
    Each reviewer prefers to read excellent manuscripts for which the concluding comments would be:
  • This is an excellent manuscript. It is recommended for publication as is provided that the authors take a few minor comments into consideration.
    Good scientists recognize the important role of reviewers and have their manuscripts reviewed by an expert in the field not because they are not sure whether the manuscript is publishable but because they strive for high quality of their report.

    There are several manuals advising authors how to write their manuscripts, for example:

    Authors are advised to use the Food Science and Technology Abstracts (FSTA) Thesaurus for their selection of keywords. This augmented and revised thesaurus reflects the advances that have occurred in the field of food science, food technology, and human nutrition in the two years since the previous edition. It has nearly 9000 terms, over 500 of them new, and is the definitive indexing and searching tool for FSTA from 1990 onwards. (The section on manuals has been contributed by Dr. Max Rüegg).

    For additional reading, there are several interesting articles on the Internet:

    The author will appreciate receiving comments from scientists, reviewers, and journal editors for future amendments and revisions. Thank you.

    Acknowledgments: This page is illustrated with two clipart images from CorelDRAW (Corel Corporation, Inc.). The author thanks Dr. Barbara Blackwell, Dr. Shea Miller, and Dr. Parviz Sabour for useful suggestions and for reviewing the manuscript.

    Updated: July 26, 2000