Thursday, July 25, 2013

Ideas for figures - comments welcome

Co-posted at pandasthumb.org.

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. 

13 comments:

NickM said...

Re: gradual, the term is much misunderstood. Padian likes to note that when Darwin saw an earthquake in Chile raise the coastline by 10 feet, he called this a gradual change. The term relates to graduation, graduated cylinder, etc. -- i.e., it actually means "stepwise".

From there we can ask how big the steps are. Does most change in system X occur in many small steps that occur at a roughly constant rate (e.g. mutations and substitutions in DNA), or does it occur in fewer somewhat larger steps, e.g. if morphological change is concentrated at speciation events, as asserted by Punk Eek, an observation that appears to be commonly true (e.g. with invertebrates) but not always true (e.g. with large mammals).

Even these punk-eek steps aren't THAT big -- it usually takes experts to tell apart different fossil clams -- but it's different than the constant uniform rate model.

So anyway, it would be dangerous to assume constant-smooth-rate anagenesis in your first figure, unless you specify that the change you are talking about is e.g. neutral substitutions instead of morphology. Cheers! Nick

NickM said...

Re: gradual, the term is much misunderstood. Padian likes to note that when Darwin saw an earthquake in Chile raise the coastline by 10 feet, he called this a gradual change. The term relates to graduation, graduated cylinder, etc. -- i.e., it actually means "stepwise".

From there we can ask how big the steps are. Does most change in system X occur in many small steps that occur at a roughly constant rate (e.g. mutations and substitutions in DNA), or does it occur in fewer somewhat larger steps, e.g. if morphological change is concentrated at speciation events, as asserted by Punk Eek, an observation that appears to be commonly true (e.g. with invertebrates) but not always true (e.g. with large mammals).

Even these punk-eek steps aren't THAT big -- it usually takes experts to tell apart different fossil clams -- but it's different than the constant uniform rate model.

So anyway, it would be dangerous to assume constant-smooth-rate anagenesis in your first figure, unless you specify that the change you are talking about is e.g. neutral substitutions instead of morphology. Cheers! Nick

Anonymous said...

My comments aren't as detailed as NickM's but I prefer the second figure for two reasons - Firstly it represents a more branching evolution than the linear style a lot of people think of ie humans are one part of an evolutionary bush rather than an end point of a linear progression. Secondly I think that the second figure shows some of the novelty that doesn't make it to the point in time where you define the species - there are partial red squares in the blue branch and vice versa, so making the point that mutations don't all get carried through in a population.
Hope that helps!

mathbionerd said...

Thanks for the comments Nick.

I'm thinking about how the term "gradual" is used by the general public. I think in practice it is used to mean, slow, in reference to time. For example, "gradually loosen the bolts", or "gradually add the flour".

Would it be best to just remove it? Or use a different term?

dwbapst said...

Following off from what Nick said, looking at figure 1, is the non-constant change example supposed to be three individuals or three species/populations?

If the latter (species), I agree with Nick that it looks like its claiming that trait change is very constant, when we know that isn't the case, as the fossil record is full of morphospecies that persist through time with very little change in their preserved parts (in trilobites, brachiopods, snails, graptolites... and even large Cenozoic mammals, Nick).

I think this overall theme of individuals don't evolve, populations do, is very important in a world full of pokemon fans (although pokemon can be useful in their own way).

dwbapst said...

With respect to the use of the word gradual, I think of gradual as meaning constant. I think that's what most people in the public would see it as meaning.

It strikes me that mine and Nick's complaints about non-constant change are maybe a bit off-center from the message... Although I admit in my graptolites, well, their evolutionary patterns do actually look like taxon A -> taxon B and so on. Perhaps this is more of a time-scale issue, that hopeful monsters isn't how change actually takes place at the scale of individuals but you zoom out and it can look less, heh, 'gradual'/constant.

mathbionerd said...

David, when I started putting that one together, I was thinking of the evolution of man figure, and how it implies that one individual changed into the next.

I think it is very confusing for people representing a single individual as representative of the entire species. We have to do it, but it can be confusing, as it is here.

I think I probably will avoid the first figure, for a lot of reasons. I'm leaning towards the second figure.

Regarding constant change or not, that is well-beyond the scope of this article. I'm really introducing the concept that populations accumulate mutations over time, and the boundary between where one species ends and another begins can be fuzzy until they are quite diverged from one another.

Anonymous said...

I think it would be good for you to explain how the eleven body systems of vertebrates (fish, mammals, birds, amphibians and reptiles) evolved. This would include the cardiovascular, respiratory, renal, nervous, skeletal, muscular, endocrine, integumentary, digestive, lymphatic and reproductive systems. If you could provide credible explanations for these body systems then it would very helpful for the broad audience that is interested in science.

Adam Retchless said...

Would it help to replace "gradual" with "continuous"?

Figrues 2 & 3 might create confusion with their use of "ancestor". I tend to think of an ancestor as an individual, not a population. Especially in Fig 3, the word "ancestor" is juxtaposed with "individual", which could throw off someone who has not yet absorbed the entire figure.

mathbionerd said...

Adam, yeah, figure 3 is out. I was going for something to try to illustrate that it isn't just one individual at each node, but a population. That totally got lost.

What do you think about saying, "ancestral population"?

strangetruther said...

Nice. In particular, the divergent populations model is of the type of speciation event that most interests me - speciation when two populations are NOT physically separated beforehand. However, I've heard it suggested that most speciation IS because of some kind of separation, typically geographic. In that case, you might want to make the tops of the trouser legs the same, only differentiating after separation. After all, the divergence is between populations rather than species :-) .

NeedleFactory said...

Assuming you will eventually have many slides and will be presenting various topics: it might be useful to adopt consistency with colors & shapes.

For example, your first slide shows evolution from blue squares to brown circles; your next slide shows two evolutionary paths from purple.

If you or your audience begin referencing concepts on a particular slide by identifying shapes or colors, it might be nice if all slides in a given presentation had some consistency in their timelines.

Otherwise, I think your slides are good. If you aren't familiar with The Visual Display of Quantitative Information by Edward R. Tufte, check it out!

mathbionerd said...

Yes, the color variation immediately popped out to me when I had finished them. I'll make them consistent for future use. :)