An irritating new weakness seems to be creeping into scientific papers, calling for further scrutiny.  I refer to the growing tendency of authors to write long, eloquent abstracts that are actually introductions rather than summaries.

 I would like to help authors by offering a few suggestions. First, start the abstract by telling the reader at once what the paper is: new data, a review of progress, a new technique, a synthesis, or whatever describes the nature of the paper.  To be sure, this recommendation can in principle be followed by a well-designed title.  But if the title does not make it clear what the paper is, the abstract should, preferably in the first line.

 A second suggestion: write the abstract in a terse, almost telegraphic style, saving your eloquence for the body of the paper.  The abstract is not an introduction to the paper, but a freeze-dried version of it, so to speak, intended as a condensation and concentration of the essential information in the paper.  It should be written for quick reading and unnecessary descriptive phrases (“it must be pointed out”) should be left out.

 A final suggestion: pack as much specific information into the abstract as possible, reporting on results, but not commenting on conclusions.  The abstract should be written after the paper is finished, when you know exactly what you are summarizing.

*Taken from “The Abstract Rescrutinized” by Paul D. Lowman, Jr.
Goddard Spaceflight Center, Greenbelt, Maryland
 

ASSIGNMENT: Write an abstract (75 words or less) of the above paper

 
 This paper presents three suggestions for better scientific abstracts: begin the abstract by briefly describing the nature of the paper (new data, review, etc.); write the abstract not as an introduction to the paper but as a tersely styled summary of its essential information; and include as much specific information as possible.  Write the abstract after finishing the paper, to avoid the common fault of abstracts that are good introductions but poor summaries.