Random Chance in Evolution - Robin May

Natural selection acts to ensure the ‘survival of the fittest’. But random chance has also played a huge role in the history of life on Earth, from meteorite strikes to massive earthquakes. Randomness also lies at the core of evolutionary processes; the impact of a chance mutation, or the ‘lottery’ of sexual selection.

In this lecture, we’ll look at some remarkable examples of evolutionary chance and reveal why they are sometimes less random than you might expect.

This lecture was recorded by Robin May on 15 November 2023 at Barnard's Inn Hall, London

Robin is Gresham Professor of Physic.

He is also Chief Scientific Adviser at the Food Standards Agency (FSA) and Professor of Infectious Disease at the University of Birmingham.

The transcript and downloadable versions of the lecture are available from the Gresham College website:

[julescaru]

I’m currently watching several of your uploads, wonderful way to spend the afternoon, thanks so much to the folk involved!
All the best Jules

[genier]

A really interesting and understandable presentation. It would be interesting to hear about the 1300 population idea and how it was determined.

[hasato-e]

I finally subscribed to science group via facebook as Gresham recommended.

[jeanneknight]

I a, glad this series came up in my algorithm. Thanks!

[srinivasvaranasi]

As fascinating as the previous lecture.

[mickeysplane]

Very well done and educational. Worth viewing.

[x.s.bleeding]

I feel the same on the Issue of Extinction. Lost Cures. ❤

[oldernu]

The subjective sensitivity of some posters betrays their biases. Getting over yourself can be a Sisyphusian task.

[gulahad]

WONDERFUL!!! WHAT A RIVETING LECTURE!!

[CuriousCyclist]

I love the analogy that you can't gradually evolve from an internal combination engine to an electric engine in a car.

[prototropo]

Wonderful, provocative topic!
I can eagerly accept the role of chance in evolution, and I don't think we should stop there. Contingency, paradox, compounded effects, dynamic filters, dialectical enhancers, redundancy, feedback mechanisms, potentiation, dynamic pulsing, bootstrapping and conditions that act as gauntlets or multipliers or course correctors are all undeniable powers of change in dynamic systems like natural selection. And speaking of dynamic systems, how can we ignore phase-change thresholds and transformations!? In other words, anything that we use today as metaphors or memes in modern life--think siloes, influencers, resonance, opportunism, settler gaze, willful ignorance, disruptive innovation--might be profitably applied to illuminate the long, incredibly complex history of evolution.
And this can happen without denying or defying Darwin's essential idea, in all its brilliant, generative grandeur.

[XH]

Technically, your ICE car already has an alternator and a battery.
You can increase the alternator power, then add an electric motor to assist the propulsion (like in certain motor sports). Congratulation, you have an hybrid car.
You can then reduce the direct contribution of the ICE to propulsion and directly transfer more power to the electric motor until the ICE is only there to power the electric parts (the second mode is more efficient)
Since you don't need the complex transmission anymore, it can start disappearing, and a mutation of the battery to a big lithium one is beneficial. And you can add a plug to charge the battery instead of using only the ICE
Since you use the ICE less and less, its size decrease until it's gone and voila, you have an electric car "evolved" from an ICE with only "small" beneficial changes

[duliofurtado]

Love the topics and the subject matter and I enjoy your enthusiasm in presenting it.
Is it possible for the videographer to focus the camera on the slides rather than on the presenter, when discussing the slides?

[rfvtgbzhn]

As a physicist I noticed that the concept of "fitness landscape" seems very similar to the physical conceptof (scalar) potential. Which actually kinda makes sense, as every natural science is ultimately based on physics.

[SteveLomas-kk]

That's one problem with natural selection, it works great as a thought experiment, as we constantly act according to our own anticipated/desire future. But take that human anticipation away, and natural selection is a much weaker force against a backdrop of chaos.

[Ted_Eddy]

I love the Founder Effect....i've often thought some evolutionary reasoning istautological based on wide assumptions.

The founder effect explains how random differences can develop between two populations that don't offer any advantages .. but without careful thought we migh infer the difference signal some advantage.

[videos_not_found]

All species seem wonderfully adapted and to offer different ideas to survive in a changing Environment. There is a considerable confusion about Randomness. for example cosmic catastrophes such as meteorites or chance events like founder effects are clearly on the selection side. Mutations on the other hand concern the production of Proteins, made of up to 35 000 amino acids. And even small changes can change the folding and function in dramatic ways.
May points to genetic drift as the source of mutations, which is reasonable but seems unproven. How multicellular organisms transform in cohorts is not known, pretension we do won't help.

[petersq]

there are two types of evolution: incremental and leap. incremental is a slow visible gradual change driven by constant evolutionary selection. leap is when something complex appears all of a sudden ( EV on the market). the background of it is gene duplication and latent mutations. functional gene got an extra copy which can start accumulate mutations while the other part functions as normal. when a, set of such genes manage to cooperate in a cascade of functions and either by evolutionary pressure or just that the last modul of a, system clicks into place the new trait appears and will be subjected to evolutionary selection.

[davidmurphy]

"If you roll enough dice enough times sometimes you can get 25 sixes in a row"

Ok, I'm going to take what was almost certainly meant figuratively literally like a pedant.

If you were only rolling 25 dice then that's 6^25, there's one microstate that satisfies the macrostate condition, which if you were rolling one a second would take about 600bn years before you'd expect to roll all 25 as sixes.

Obviously that changes massively with more dice. If you roll trillions of dice then the chances of it not happening become vanishingly remote with a single roll.

So yes, even pedantically, it's a valid statement.

[marvinmauldin]

I hadn't heard about the bottleneck 900, 000 years ago, but I had heard about the one 72, 000 years ago that left about the same number, and seems to appear on the chart. So why wasn't this later bottleneck mentioned?