Ideas for figures - comments welcome

| 37 Comments

I’m working on an article discussing “Trends in Evolution”, to be written for a broad audience interested in science, but not experts.  I’ve already narrowed it down to a few sub-topics to focus on, but I would like some advice on figures.

I am working on figures to succinctly illustrate how evolution works, while also addressing common misconceptions. What do you think? How can I improve these? What have I unwittingly misrepresented? Which do you like best?

1. Evolution is the gradual change of populations over time, not distinct transitions between species.


Here, I’ve already received feedback that I should remove the word “gradual”. I agree that “gradual” is a relative term, and in many cases, evolution happens very quickly. Given that I work on long-lived species, I tend to use “gradual”, and like it for this example.

 2. Individuals mutate, populations evolve


3. Individuals represent a subset of variation within populations. Populations evolve.
 

I think this one might be assuming too much background information, or just be too vague.

37 Comments

I think the first pic is great. Also, across the board, the use of small shapes and colors to represent indidivuals within a population is excellent.

The second picture is better than the third. You’ve just introduced people to thinking those small shapes are individuals within a population, and then on the left side of pic 3 you reverse that and use the same small shapes to represent whole species. That may be confusing to some readers. For figure 3 (left), you might simply be better off using small pictures of related animals themselves, instead of trying to genericize it to shapes and colors.

I have no problem with the word ‘gradual’ in the first one, I don’t think it detracts at all.

Small quibble only, but you might consider making the color schemes consistent. In the first you use blue as the parent species with purple and red being the daugters. In the second and third pics, purple is the parent and blue and red are the daughters.

Depending on the space available, I would consider rotating pic 2 90 degrees CCW and combining it with pic 1. That might require a fourth color as you then have blue parent, purple daugter, then red and (green?) granddaughter populations. Maybe something like that could be used later in the chapter, to wrap up the ideas discussed and give a ‘whole view’ figure combining the parts that the student has seen individually.

For 1, the text at the bottom jarred me a little. I suggest the top and bottom lines to be:’Evolution is the gradual change of populations over time. There are no distinct transitions between species.’ The bottom squares of this figure that represent distinct species can be removed, and replaced by 3 overlapping brackets. The leftmost bracket spans the mostly blue population. The middle bracket spans the mostly purple population, and the right bracket spans the mostly red population. Again, the 3 brackets overlap each other so the bracket under the blue also includes some purple, and so on.

I like the comments from eric above. One can take it a step further and point out that figure 2 also does a great job of supporting the point made in figure 1. So much so, that you do not need fig. 1 at all!. Just use the picture from fig. 2 in fig. 1, then repeat it for fig. 2. This also takes care of the consistency issues that eric mentions.

I am not entirely comfortable with ‘gradual’ because it does have the connotation of a long/slow time span, which is not helpful to the argument. ‘Evolution is a ‘series of small changes within populations over time’?

In the top diagram each object represents what? An individual organism? If so, then the blue square should read something like “member of ancestral species 1”

Instead of squares for a given ancestral species draw something explicit? Example ~ finches with various sized beaks leading to a modern species where the range of sizes has settled on a large size. Such a diagram would enable you to show why species are designated well after the fact. Perhaps a dotted line representing where one species population becomes another ~ illustrating why the dotted line is a convenient shorthand fiction that disappears at high resolution.

I think evolutionary explanations that I’ve read are not good at diagramming how fuzzy the whole process is. Ring species would be a good graphic exercise with time slice layers to show the flows across geography/time. I haven’t seen a good digram of that [BOTH geography and time that is]

In figure 2, are you trying to imply from the colors that the blue/red alleles are co-dominant, as expressed in the purple ancestral phenotype, with loss of one or the other leading to red/blue ancestral populations? That’s a pretty subtle point, especially since it is overlain on the square/round phenotype. Only squares turn blue, and only circles turn red, which seems confusing unless there is some kind of linkage thingy. The various shades imply yet another phenotype that is not linked to square/circle. I fully appreciate how difficult this kind of work is, but this figure would be a real challenge to explain if someone asked you ‘why are squares only blue and circles only red’?

I think figure 1 is fine the way it is, because it does a good job of refuting the standard image that most people have of evolution.

On Figure 2, I’d suggest two changes. First, individuals don’t mutate. Genes mutate … or rather, genes vary (mutation is not the only source of variation). Individuals are selected. In fact, that’s the way I think of evolution: genes vary; individuals are selected; populations evolve. I’d also change the colors, because using the same colors to mean different things in figure 1 and figure 2 could be confusing. I suggest that figure 2 use green for the ancestor, and blue and yellow for the descendants.

I’m not sure what point you’re trying to make with figure 3. Founder Effect? Allopatric speciation?

I don’t like figure one, as it implicitly assumes a problematic definition of “species” and “speciation”. The most common species concept is the so-called biological species concept, which really can’t be applied to ancestor-descendant pairs. “Species” is a relationship between coeval populations, and it’s fairly meaningless when you attempt to extend it over evolutionary time. Now, what you’re trying to do here is to introduce the concept of gradual, continuous change in a single lineage. That’s evolution, but it isn’t speciation, and it doesn’t result in new species. What you have here are chronospecies, and even paleontologists frown on those nowadays. Take out the references to species and your fine. If you are instead trying to communicate something about species, then I think you need to think about what that is, and what you mean by “species”.

Figure 2 shows divergence. It might show speciation, depending on species concept and the reasons for divergence. If it’s speciation, at least according to the BSC, there should be some isolating mechanism. Or it could be just geographical isolation at work here. Who can say?

I’m not sure what Figure 3 is trying to show. Are you setting up a coalescent? Unclear.

Um, remember people these are supposed to be simple figures aimed at a general audience. Of course deep thinking uncovers errors in a simplified figure. But the audience is not going to be thinking that deeply. On a different note: through this, mathbionerd, I have discovered your blog and you now have a new fan.

My notion is that simple figures for a general audience should attempt not to teach egregiously false points, especially when the correct point would be a simple. Now in evaluating each figure, the main question should be what point the author is attempting to get across. I’m guessing that the headlines are hints in that direction. In Figure 1, for example, the bottom row (species 1, species 2, etc.) is a distraction from the message.

Words or diagrams are still all unpersuasive to the general population. In fact they think species evolve from previous different looking species. Learning its from a few within a population that survive and then create a new species is making it more unbeleivable.

You must ensure they understand there is great extinction in order for the few individuals with new mutations to stand out and become the new species. People will question what happened to the majority who did not mutate/breed. Selection is not just bringing forth successful new types but mostly eliminating the unsuccessful types who were successful for a while.

The point of evolution is the power of mutationism. Mutations are the engine for speciation and not selection. Without mutations one could not create what was created. Then selection holds in place every new mutation that has started a breeding group. Graphs can’t make it a easy thing for bugs to become buffalos. Mutationism acting on a few in a population is the origin of species. The unmutated majority of the population by observation/fossil record cease to exist. Of coarse creationism sees things otherwise.

I’d think the simplest case would have only one thing varying among the individuals.

Some cases beyond the simplest:

Have two or three things that vary, one of which changes its average over the duration of the chart.

Have two things that vary, with no initial correlation with each other, and over time become correlated with each other, leading to isolation of the two groups (i.e., speciation).

Hybridization of two species.

Polyploidization (is that how you spell that?) where the offspring are suddenly a different species in one generation.

Henry

Thank you everyone for the comments.

I should have clarified that I’ll be choosing just one of these, and am considering making another figure that highlights the abundance of extinctions throughout time. The introduction is quite short (only 1500 words), and is to be written for a broad readership interested in science and technology. You can make of that what you will.

Eric, I admit that after I made 1 and 2, I immediately realized I should have been consistent with the colors, or switched them completely. It’s one of my pet peeves too, but these are just drafts.

Thanks, Mark!

If you would explain exactly what you are trying to communicate, the comments might be more effectively focused. If you have multiple objectives, consider the possibility that one figure might not achieve those goals, and any attempt at multiple goals might prevent all of them from being achieved.

Wilson,

I like the graphics. My big change would be to the first figure:

1. I agree you should drop the “gradual”. 2. I think you should also drop the arrows with the red NO symbols over them. 3. I think you should rename “Ancestral Species” 1+2 to “Ancestral Fossil” 1+2. 4. I think you should change the caption to: “Fossilisation only preserves a small number of individual organisms, which can give the illusion that there are distinct transitions between species.” Or some such.

The problem I have with figures 2+3 is that they show a speciation event but nothing in the graphics mentions speciation. If you’re just trying to show that “individuals mutate, populations evolve” you’d be better off showing a single non-bifurcating stream. I’d repurpose this graphic for the specific goal of demonstrating speciation with a title like “Populations often evolve into distinct species.”

Good work, though.

I think Byers’ comments are for once helpful. Showing a new mutation occurring and then spreading through a species gives the impression Byers got: that the successor species come from only one or a few members of the present species. But that does not take recombination into account: I can have a new mutant allele from one individual in my ancestry, and at another locus another new mutant alleles from another individual who was neither an ancestor nor a descendant of the other individual.

You’d somehow need to show ancestry at two or more loci.

M. Wilson Sayres said: I should have clarified that I’ll be choosing just one of these, and am considering making another figure that highlights the abundance of extinctions throughout time.

In that case I would take fig 1, because it is the most intuitively understood yet gets the big message across. If you can, consider sticking fig 2 on the right side of it (i.e., making a y- between two daugter species).

Eric, I admit that after I made 1 and 2, I immediately realized I should have been consistent with the colors, or switched them completely. It’s one of my pet peeves too, but these are just drafts.

Well its not a major issue if you’re picking one of the three. In that case, I’d suggest you get the opinion of a graphic designer rather than a scientist - some color combos are just better than others, but damned if my artistic sense can figure out the good ones.

I’d suggest changing the bifurcation in diagrams 2 and 3. They’re shown as having already distinct populations by the split, rather than identical - this illustrates an abstract idea of speciation perhaps, but misrepresents the causes of speciation like geographic isolation. Also, changing it to very similar populations existing for a short ways after the split, it illustrates the possibility of hybridization like mules and ligers.

I know you want to be concise, but I’m a little leery of “Individuals mutate” - perhaps “Offspring differ from parents” or something along those lines? It’s a trait obvious to everyone and doesn’t suggest the three-legged concept of mutations that too much of the public has. And to be honest, individuals don’t mutate - that occurs during DNA replication before there’s a new individual.

Turn diagrams 2 and 3 sideways, like 1. The vertical ascent/descent thing implies a lot, mostly misleading.

I think I’d also lean towards more shapes, myself. People may concentrate on the shapes and not the distinctions of color within, leading them to believe that the squares from the ancestral population are the same as the squares of the current species.

However, overall I think the concept’s great, and covers a lot of the misapprehensions people have. Showing the variations in a given population and how the average shifts without giving up variation is strong. I don’t think it’s possible to illustrate any concept perfectly - someone’s always going to get the wrong impression - but the visual aid of minor changes progressing over time corrects a lot of the confusion I’ve witnessed, at least. I don’t want nitpicks to take away from the predominantly positive effect I suspect this will have.

I love the symbology.

If you are stuck with just a single figure, I would recommend #3. I like the distribution of individuals with differing attributes that vary over time and space. I agree that the specific distribution of attributes isn’t particularly accurate, but that’s okay. It’s a very simplified schematic, not a accurate representation.

However, if it’s not too late I would recommend some changes.

First, invert the figure: past at the bottom, present at the top. You want to implicitly emphasize evolution, not devolution. Yes, it is subtle and misleading, but it plays to and reinforces something the audience already suspects, slotting it into what they already know, making it easier for them to grasp the new information that you really want to show.

Second, label the frog of the fork with whatever speciation trigger you want to emphasize: geographic separation, sexual selection, etc. Something to indicate that there is a “cause”, rather than just drift. Yes, “drift” plays a part as well, but that’s not what you are trying to emphasize.

Third, increase the number of individuals. Double it if you can. Or more. The less distinct the differences between nearby individuals, the better.

Fourth (and most importantly), for the three-dot stick figure, select the three dots from among the evolving population. Use yellow circles around the dots “in place”, showing how the fossilized examples that we have today are simply small discrete samples from a larger population, which give the impression of discrete speciation “events”. With the circled dots in place, you can replicate those dots outside the evolutionary “stream” connected with sticks. Or, you can just leave the circled dots in place, and connect them with sticks.

In this way, #3 would combine the ideas from both #1 and #2, and puts the bottom row from #1 into the more accurate context of #2.

Best of luck!

Robert might gain some credibility(?) by showing a credible explanation of the (his ‘theory’) evolutionary transition from Wolf (mammal) to Thylacine (marsupial)

Appreciating the variability within populations is key to understanding differentiation. If your mental image is coupled to an image of a “type specimen” it’s unlikely that any of these diagrams will be useful.

As an absolute non-biologist, the image that I like is of a women facing a chimpanzee. The woman is holding her mother’s left hand with her right hand. The chimp is holding her mother’s right hand in her left hand. Both lines continue into the past. At some point the lines join where a mother is holding the hands of two daughters, each belonging to different lines.

Any further complication is mere biology.

Any further complication is mere biology.

Or in software engineer speak: That one page flow is the design; the rest is implementation.

Henry

I like the figures!

Since you said you’ve narrowed it down to some sub-topics the next part might be off-topic/unhelpful.

It would be nice to have a figure explaining the importance of natural selection in building complex systems (perhaps not in the article, but somewhere). Perhaps comparing random chance to natural selection. This seems to be almost always overlooked and creationists prey on the misunderstanding by using probability arguments.

This is a shot (just the concept, obviously these aren’t actual figures):

Random Chance, number represents number of mutations building complex system(~20% mutation rate in both directions): 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 1 0 1 0 1 0 0 0 1 0 0 0 2 1 0 0 0 0 1 1 0 0 2 1 0 0 0 0 1 2 1 0 2 1 0 0 0 0 1 2 1 0 2 1 0 0 0 0 1

Natural Selection, number represents number of mutations building complex system(~20% mutation rate in both directions, selection is 100%, favoring complex mutations):

0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0 1 2 1 1 1 2 1 1 2 2 2 2 2 2 2 2 2 2 3 2 2 2 3 1 2 1 2 2 3 3 3 3 3 3 3 3 3 3

Random Chance, number represents number of mutations building complex system(~20% mutation rate in both directions): 0 0 0 0 0 0 0 0 0 0 – 0 1 0 0 0 0 0 0 0 1 – 0 1 0 1 0 1 0 0 0 1 – 0 0 0 2 1 0 0 0 0 1 – 1 0 0 2 1 0 0 0 0 1 – 2 1 0 2 1 0 0 0 0 1 – 2 1 0 2 1 0 0 0 0 1 Natural Selection, number represents number of mutations building complex system(~20% mutation rate in both directions, selection is 100%, favoring complex mutations): 0 0 0 0 0 0 0 0 0 0 – 0 1 0 0 1 0 0 0 0 0 – 1 1 1 1 1 1 1 1 1 1 – 0 0 1 2 1 1 1 2 1 1 – 2 2 2 2 2 2 2 2 2 2 – 3 2 2 2 3 1 2 1 2 2 – 3 3 3 3 3 3 3 3 3 3

Thanks again!

A few quick responses:

1. Yes, yes, I completely agree. It should say genes mutate (or DNA mutates), not individuals mutate.

2. Alan, do you have an example of that figure. It doesn’t sound like something I could make, with any sort of aesthetic appeal, but I do like that sentiment for representing evolution to the public. But, I’m concerned that this might still reinforce the idea that individuals are evolving if it isn’t made clear that these are parent-offspring holding hands.

3. In this figure I wanted to emphasize the variation within populations - I can definitely increase the number of shapes to try to take the focus away from any one individual.

4. Someone else is writing a whole section on Natural Selection. Strangely, IMO, the section “Natural Selection” was given 2500 words, while the more general “Trends in Evolution” section was given 1500 words. You can see where the editor’s preferences lie.

I see the problem. I have used the word ‘image’ but when I heard it, it was described in words. You cannot really draw it without precise knowledge of what all the creatures involved actually looked like back through the generations. Certainly, such a line exists for any particular woman, it is this that makes it so convincing, but it hides the variation that is occurring in the rest of the population. Mutations will pass from mother to daughter but those passed by males have been ignored.

Maybe include something to explain how positive feedback causes accumulation of changes that emphasize or optimize a trait or ability, without it being dependent on any one particular change along the way. (Though maybe less technical than how I said it?)

A lot of the “arguments” talk about the odds of particular mutations, but that’s the wrong question. One needs to ask if there are any mutations (or recombinations) that would optimize some trait that would increase reproductive success.

Henry

I think that increasing the number of shapes would muddy the lesson. Again, you want one slide to teach one lesson. Are you trying to talk about the spread of alleles in a population or about sheer population variation? The one is best addressed by showing the spread of that one allele, i.e. only two shapes or two colors (and not both). The other is best addressed by showing just a snapshot in time with lots of shapes and colors. I wouldn’t go for both at once.

Now if you want the former, I think you might try a coalescence figure. You don’t have to tell anyone about this, but you could have a constant-size population of shapes, introduce a mutant (perhaps more then once), show lines for mating, with each offspring having, say, a 60-40% chance of resembling each parent when the two are different (just to guarantee fixation), and just let that run to fixation. Again, you don’t have to talk about how this picture was arrived at; it just shows alleles changing in frequency until one of them goes away, and the population is transformed. You can easily find the raw images (uncolored) for such a figure online.

May I take a completely different view? I like to picture species as sets, and I came up with this diagram: you have two dimensions measuring some features (doesn’t matter which ones exactly), and a species is just a cluster of points in (say) upper left corner, with a boundary. Over time as you take snapshost, set of points moves slwoly towards the lower-right, always with an overlap to the previous position, but if you wait long enough the envelope of the first and last sets do not intersect. Children are always within the same envelope as their parents, but as the envelope moves, eventually the great-great-great-grand-children will be sufficiently different from their ancestors. You could also do an animation to illustrate this point.

Just my .02 $

That sounds sort of like an amoeba crawling across the landscape.

Here, I’ve already received feedback that I should remove the word “gradual”. I agree that “gradual” is a relative term, and in many cases, evolution happens very quickly. Given that I work on long-lived species, I tend to use “gradual”, and like it for this example.

I much prefer “incremental” over “gradual.” As you note, “gradual” carries connotations of slow rate, while “incremental” doesn’t (so much), and descent with modification is almost universally an incremental process (speciation by polyploidy to the contrary). “Incremental” fits your example just as well as “gradual.”

That sounds sort of like an amoeba crawling across the landscape.

Across the fitness landscape, yes. I guess the concept of a set boundary moving over time would take out the obvious misconception of a chicken being born to dinosaur parents. But then, maybe I am too much of a mathematician…

I’ve always liked this one. It’s different from the diagrams, using simple text to make the same point http://i.imgur.com/xWpvw.jpg

What would also work would be two copies of that blue text, side by side, the second one changing from red to orange to yellow to green. Then you’d have two new species, the blue and the green, descended from a common ancestor.

I don’t know if it is too late but I have an image I made to respond to ArthurianDaily some years back. He couldn’t understand that evolution is changing populations and kept insisting that species had to have only a single breeding pair for an ancestor. I made the graphic but he left and I never bothered with it after he left. I hope it has some use.

http://postimg.org/image/nmzljuygb/

The parent generation A thru L gives birth to 5 sets of offspring carrying 5 variations that then disperse thru the population. No generation would be incapable of breeding with its parents but generation 11 is different enough from the parent generation that they are probably reproductively isolated from the parent. The point was that at no time was there only ever a single breeding pair. An entire population changed into another population.

Thank you for sharing, dornier.pfeil.

About this Entry

This page contains a single entry by M. Wilson Sayres published on July 25, 2013 10:48 AM.

Kentucky science standards derided as fascist, socialist was the previous entry in this blog.

Bruce Chapman—“The Pseudo-Science Guy” is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Categories

Archives

Author Archives

Powered by Movable Type 4.38

Site Meter