Dr. EGAN
GUIDELINES TO FIGURE WRITING
Figures are sometimes considered to be the most important part of a scientific
paper. They contain and relay all the evidence reported in the research. Since
that is the case, all figures and tables must be self-standing, which means that
the reader should be able to understand the figure just by reading the legend.
The reason for this is simple, most scientist are either efficient (or lazy) when it
comes to reading journal articles.
Most scientists read a scientific paper by:
1) Breezing through the abstract
2) Reading the last paragraph of the introduction
3) Looking and reading the figures
4) Then conclude by reading the discussion
Sometimes they may take two additional steps:
5) They may read the results, if it is hard to understand the authors
line of reasoning behind the experiments
6) Sometimes the materials and methods are read, if one of the
techniques described in the figure legend seems confusing, or you
may want to try the technique yourself in your own lab
Why do scientist read this way?
The abstract lets them know what the paper is about
The last paragraph of the introduction lets them know exactly what
the researchers did
Since the figures are self standing, the results are additional and
are sometimes even skipped
The discussion allows them to see how the researcher perceived
the data (then the reader either agrees or disagrees with their
interpretation)
So how do you write a figure legend?
Every figure legend should be written to answer the What-Why-What questions.
What is it?
Why did you do it?
What were the results?
Lets now walk through an example of how to write a figure legend:
Dr. EGAN
35,000
30,000
25,000
Growth
20,000 ucn1
15,000 wt
10,000
5,000
0
0 5 10 15
Tim e
Without a legend you have no idea:
What is growing: ucn1, wt what are they? organisms? What the time intervals
are? Why was this done? Is this normal? To answer these basic questions a
solid figure legend needs to be created. To start off with a title needs to be
made. It is unfair to expect you to create a title when you didn’t perform the
experiment so lets just say the title for this figure is: Comparison of the ucn1
mutant to wild-type cells in 2% LiCl. Notice how the title is not really a
complete sentence, and only contains the most important information. The rest
of the information will be expanded upon in the actual legend. So lets now work
on the legend.
If the experiment was performed on human neuron cells grown in culture and
ucn1 is a mutation in a gene that is necessary for growth in high lithium
concentrations the legend would read.
Figure 1: Comparison of the ucn1 mutant to wild-type cells in 2% LiCl. To
determine the effect lithium has on ucn1 mutant cells, human neuron cells were
cultivated in growth media amended with 2% LiCl. Both the wild-type control
cells and the ucn1 mutant cells were inoculated at concentration of 1,000cells/ml.
Under normal growth conditions, Wild-type cells double every 2hrs; from
1x103/ml to 3.2x104/ml after 10 hours of growth. In comparison ucn1 mutants
decreased by ½ every 2 hours; 1x103/ml to 0/ml in 10 hours.
Now lets examine this figure with the What-Why-What questions:
What is it?
It is a comparison of the ucn1 mutant to wild-type cells in 2% LiCl.
Why did you do it?
To determine the effect lithium has on the growth of ucn1 mutant cells.
What were the results?
When growth should double, it was in fact halved in the mutant
(Notice how in this part you gave actual numbers to show your answer)
Now lets practice writing figure legends on the following three figures. The
“HELP” section indicates data that would be useful in creating the legend.
Dr. EGAN
50,000
colony forming units
40,000
30,000 Temp 37°C
Figure 1
20,000 Temp 50°C
10,000
0
0 2 4 6 8 10 12
Tim e
40
35
30
Figure 2 25
Length in cm 20 water control
15 spray @ day12
10 spray @ day19
5
0
NS RRS NC RRC
Seed Variety
40
35
30
Figure 3 25
NS
Height cm 20
RRS
15
NC
10
RRC
5
0
0% 2% 5% 10% 20%
spray
HELP: NS= normal soy; RRS= transgenic soy; NC=normal corn RRC= transgenic
corn spray is 2% Glyphosate (round-up) unless otherwise stated