Charles Darwin, evolution simple and testable01/07/2015
Evolution is testable like all good normal science, writes DR DAVID PENNY.
This article was originally published in New Zealand Science Teacher in issue 116, 2007. We’re republishing it here as a useful resource for classrooms.
A striking feature of evolution is that it is made up of a number of very simple parts – each of which is fully testable. Nevertheless, taken together, the ideas put forward by Charles Darwin – and developed mathematically over the last half century − can appear complex. The aim here is to identify the main components of the theory and show how they relate to developments in physics and geology in order to demonstrate that each part is both simple and testable and thus show that evolution is just good normal science.
Charles Darwin’s family and science background
We will look quickly at Charles Darwin’s family background, and then the state of biology at that time. His paternal grandfather, Erasmus Darwin, was by the late 1700s a leading doctor, poet, inventor and scientist (a member of the London-based Royal Society, made up of leading scientists). Erasmus also proposed evolutionary ideas, but unlike his grandson two generations later, Erasmus’ ideas lacked a testable mechanism.
Charles’s maternal grandfather was Josiah Wedgwood, inventor, scientist, and founder of Wedgwood potteries. Both families were leaders in the early Industrial Revolution that introduced the start of modern scientific and technological society as we know it today. Both families were progressive liberal, and very anti-slavery. Erasmus Darwin also founded a private school for girls (Ashbourne) because at that time there were so few opportunities for their education.
Charles’s father, Robert, was also a doctor and had published some early scientific work. Charles himself tried studying medicine at Edinburgh but dropped out after two years – apparently he did not like the sight of blood! He then attended Cambridge University, and was expected to eventually earn a respectable living as a village vicar. (At that time there weren’t many professional career opportunities for sons of the well-to-do!) However, as an undergraduate he became more and more interested in geology and biology, and this led to his being invited to be the naturalist on the voyage of the H.M.S Beagle, starting at the end of 1831.
Before we go on, we need to take a quick look at the scientific understanding of ‘species’ at that time. Until about 1700, almost everyone in Europe accepted continued spontaneous generation –the belief that some living things can arise suddenly from inanimate matter. In this theory life continued to arise from Earth through the power of a creator, as described in Genesis.
From around 1700 the hypothesis that ‘species’ had a permanence over thousands of years became popular in scientific circles. Experiments showed that, at least for multicellular organisms, all life came from eggs (or seeds). So the next idea was that ‘the work of the creator’ was finished – species had all been formed by special creation, but only in the past!
Strictly speaking, this is consistent with species being created and then evolving, but the idea of species being unchangeable fitted in with the philosophy of an ancient Greek (Plato, from the 5th century BC) who postulated unchangeable ‘essences’.
This idea of an unchangeable essence from Plato has been very influential in Western philosophy. Thus the theory of unchangeable species being created only in the past was a mixture of two components – a creator, and the ideas of Plato, the pre-Christian philosopher. There was no evidence for Plato’s unchangeability of species, and surely an all-powerful creator could make a system that evolved!
This switch from continued spontaneous generation to special creation occurred between the time of Abel Tasman’s (1642) and Captain Cook’s visits to New Zealand (1769). In North American history the change would have occurred between the arrival of the Pilgrim Fathers (1620) and the War of Independence.
Thus, by the time the young Charles Darwin was studying it in the 1820s, biology was (apart from a few doubters who lacked any testable mechanism) firmly in the natural theology phase we now call classic creationism. We will return later to follow these changing concepts through time, some of which are shown in Figure 5.
The present is the key to the past
Darwin’s almost five-year voyage around the world on the British navy survey vessel HMS Beagle was the defining event leading him towards evolution. His first major scientific step was convincing himself of the validity of the ‘new geology’ advocated in the early 1830s by the Scottish geologist Charles Lyell. Lyell (following an even earlier Scottish geologist, James Hutton, whose main works were in the 1780s and ‘90s) aimed to explain ancient geological events by mechanisms that can be studied in the present – citing ‘the present was the key to the past’. During the voyage, Darwin saw, for example, how earthquakes in Chile raised the coastline several metres—and then found evidence for a series of raised beaches and/or remains of fossil seashells higher and higher up the adjacent hillsides, and later to the tops of the passes over the Andes.
The simplest explanation was a series of similar earthquakes in the past – the present processes (earthquakes in this case) were explaining past events. At the same time, geology was gathering evidence for the Earth being at least hundreds of millions of years old, rather than the 6000 years favoured since the late 1600s. The new geological timescale allowed plenty of time for similar events to have continued back into the past.
Thus the ideas underpinning the scientific advances in geology were an important key. And geology itself was fitting with the new models from physics and mathematics. Ever since Sir Isaac Newton, physicists explained the movements of planets by the combining the measurements of mass, gravity and acceleration that they could make here on Earth. They postulated a continuous series of intermediate steps to the position of planets, thus bringing in calculus with its integration and differentiation – similarly assuming continuity of intermediate states. So physics, calculus and now geology were adopting models based on a continuous series of intermediate states with small incremental changes; would the same approach work with the more complex systems in biology?
But again, coming from physics and geology, there had to be a mechanism, or force, that could be studied independently.
Darwin’s next major step was to transfer to biology these principles he had learned from geology, using current mechanisms to explain past events. For example, as the HMS Beagle sailed up and down South America, he observed that a given species could vary − finches on the Galapagos Islands appeared more similar to South American species than to any other. The simplest explanation was that South American finches had got to the Galapagos Islands (after the islands had formed from volcanoes) and then these South American finch species had adapted to local conditions.
Figure 1. Subspecies of Brassica oleracea - artificial selection can modify species well beyond the limits of the species seen in nature. (a) The wild form is subspecies ssp. oleracea and is found on rocky cliffs. (b) Kale and collards are members of the ssp. viridis; curly kale (not shown) is ssp. sabellica. Jersey kale (not shown) can grow to be 2 metres (7 feet) in height and belongs to ssp. palmifolia. (c) Brussels sprouts are classified as ssp. gemmifera (d) Kohlrabi as ssp. gongylodes. (e) ssp. capitata includes red and green cabbage. The English word ‘cabbage’ comes from the French word ‘caboche’, which means head. (f) Savoy cabbage belongs to ssp. sabauda. Image source: http://serc.carleton.edu
Other hypotheses were possible – a Great Gardener could have taken species from anywhere on Earth – in which case there was no reason for Galapagos’ species to be closest to South American species.
However, because of these regional similarities of species, it was a simpler hypothesis that natural
mechanisms were responsible for finches getting from South America to the Galapagos. By the end of the voyage of the H.M.S. Beagle, the question of the origin of species was uppermost in Darwin’s mind. Although Darwin’s main publications from the voyage were in geology, he increasingly transferred the mechanistic approach into biology.
The importance of plant and animal breeding
On his return to England, Darwin searched for mechanisms that would help explain how populations could change with time. He quickly recognised that plant and animal breeding was an excellent model, in the sense that new ‘varieties’ of plants and animals had been generated by breeders, and that these were now very different from their parent species.
Consider a giant Great Dane and a small Chihuahua; they are so different in size and appearance that if they were the only dogs known then they would be classified as separate species, and almost certainly in different genera.
Similarly, for plants, one original species (Brassica oleracea, if you really want to know) has led to cabbages, cauliflowers, kale, Brussels sprouts, kohlrabi, and broccoli – see Figure 1.
So in this sense it was ‘known’ for both animals and plants that there were no absolute boundaries to variation within a species – a very fundamental point that contradicted the application of Plato’s ideas to species. Thus, artificial selection by humans could change existing species well beyond the limits of what was found in nature. Plant and animal breeders already knew that there was variation within species and varieties, and that some of the variation was inherited. Yes, Mendel and his laws of genetics came later, but people already knew both about inheritance (‘breeding true’) and also about environmental variation.
Left: Figure 2. Bold simplicity; evolution is the simplest possible hypothesis. The basic postulate is that there is a continuous series of generations back through time to ancestral species. If we allow an average of 25 years per human generation, then after 10 generations we are back 250 years, back 1000 years after 40 generations, 10,000 years after 400 generations, and so on. Nothing could be simpler. To suggest that everything appeared at one point in time: thousands of galaxies, our solar system, the Earth and its organised geological strata, and 5–10 million species (each with extremely long DNA sequences that look as if they have shared common ancestors, see Figure 3) is just unbelievably complex.
For example, botanists had transplanted plants from their natural environments into gardens, and had observed that some differences persisted, while others disappeared. Thus some differences were inherited and persisted, while others were dependent on environmental conditions (genotypic and phenotypic in modern terminology).
Again, it was known that some of the inherited differences were ‘useful’ to plant and animal breeders – organisms having certain inherited features were selected. All that was required for Darwin to make the intellectual leap to his ‘mechanism’ was that some of this inheritable variation increased (or decreased) the chance of an individual surviving in nature and leaving offspring. Note that this statement is quite weak – ‘some’ of the variability affected the probability of survival and reproduction – we will come back to this later. The inherited variability is important; if all individuals in a species were genetically identical, there would not be any genetic change between generations.
Correct use of mathematics vital
However, inheritable variation (genetics) was not sufficient by itself, and so we turn to the population parts of the theory; the first is the potential for a geometric increase in population numbers. It is well known that Charles Darwin was impressed by the mathematical rigour of Robert Malthus’ calculations on the potential for increase in population numbers. For example, a unicellular individual dividing into two offspring per cell division has the potential to increase to 2, 4, 8, 16, and so on, in succeeding generations.
As an aside, Malthus (a pastor) used his calculations to come to very conservative conclusions about not helping the poorer groups in society for fear that they would increase in numbers exponentially. As such, this was ‘politically incorrect’ in Darwin’s extended family circles. Even decades later, Darwin (in his autobiography) almost had to apologise to his family for reading Malthus – his excuse was that he was reading it ‘for amusement’. Without knowledge of birth control, alternative conclusions from Malthus’ calculations were not obvious at that time. Nowadays we know better (or should know better) that the mathematics is neutral; it is our use of it that causes problems.
Figure 3. Testing the human, chimpanzee, gorilla grouping. The figure shows 70 nucleotides from a long (>10,000 nucleotide) non-coding region of DNA from humans, chimps, gorillas, orangutans and rhesus monkeys. Unless indicated otherwise, all species are identical to the human nucleotide in the top row. For example, with the 4th nucleotide all species, except rhesus monkeys, have an ‘A’ for adenine. The rhesus monkey differs most from the other primates, indicating it branched first (as shown on the right hand side), followed by the orangutan. Because (over the whole sequence) there are many places where humans, chimps and gorillas share the same nucleotide (as for the 9th nucleotide shown), it can be directly concluded from this data that these three species are more closely related to each other than to either orangutans or rhesus monkeys, thus supporting Charles Darwin’s prediction. In addition, chimpanzees and humans are somewhat closer to each other than to gorillas. Estimating times of divergence is a bit more complex, but in the case of humans, apes and monkeys the order of branching of the tree is straightforward (given long enough sequences).
Below: Figure 4 A: A comparison of predictions from the theories of ‘intelligent design’ (A) and ‘descent with modification’ (B) (right). The test compares protein sequences of photosynthetic enzymes in plants living in very hot, dry climates. From intelligent design, the enzymes might be expected to be designed similarly, at least compared with enzymes from plants living in cooler moist conditions. Here we compare a cactus and a grass living under the same desert conditions (under water stress) with a grass (or you could use a cactus relative) that lives in moist temperate conditions, that is, without either temperature or moisture stress. The theory of ‘descent with modification’ predicts that the sequences of the two grass species would be more similar to each other than to another plant living in the same environment, but not sharing a recent common ancestor. It is essential always to consider ideas or explanations as hypotheses to be tested.
Returning to the potential for the increase in population numbers, it was also known that, on average, and although numbers varied between years, such an exponential increase didn’t occur. Yes, the potential was there, but there were obviously other limitations. Darwin’s next logical step was his awareness of the limited nature of resources. For example, for plants the amount of sunlight for photosynthesis is limited by the surface of the globe – there is a maximum amount of light energy per square metre.
Consequentially, for herbivores there are limitations on food from the biomass of plants. Further downstream, some bacteria and protists in the rumen are limited by the amount of grass a sheep or cow eats, and so on.
Obviously, many other factors can be limiting under some circumstances. The simple consequence, given the potential for increase in numbers and limited resources from ecology, was that there must be competition for resources – both within populations and between them.
Mechanistic and probabilistic thinking
At this point, there are two important generalisations about Darwin’s theory. Clearly it was very mechanistic in that it was aiming to explain past events (changes in species over time) by mechanisms that can be studied today in the laboratory and in the field.
The other novel aspect for science was that his theory used probabilistic thinking – a specific outcome depended on chance events. This was well before physicists adopted probabilistic thinking for quantum effects, so physicists were initially sceptical of Darwin’s mechanism – they were thinking far more deterministically in the mid-19th century.
But Darwin reasoned that if you had large numbers of individuals, and long periods of time, then you could make strong statements about the ‘average’ (or expected) outcome. Today, we are quite used to statistical reasoning, but it was a radical step for science then. Thus the use of known mechanisms, and probabilistic reasoning, were major features of his theory, and major intellectual leaps for his time.
Microevolution, then, is the combination of these populational, ecological and genetic processes, and results in a mechanism for ‘natural selection’. However, we should be careful here − calling it ‘natural selection’ makes it almost like something that ‘exists’ as a separate entity. Rather, referring to natural selection as the inevitable outcome of microevolutionary processes is a more accurate picture (but more tedious in everyday conversation).
Nevertheless, it is an error to consider ‘natural selection’ as an entity − it is just the automatic consequence of the above processes. It is very important to note that all those components of Darwin’s proposed mechanism, as is summarised in Table 1 (below), can be tested independently.
|component||is it testable?|
|exponential increase in population numbers||yes|
|limitation of resources||yes|
|competition (intra and inter-specific)||yes|
|inherited diversity within populations||yes|
|some inherited variation advantageous||yes|
|leads to improvement over generations||yes|
If we combine the population, genetic, ecological and resource limitations, together with the hypothesis of a continued series of generations in the past, then we expect some limited changes between generations (though the amount of change per generation can obviously vary). A representation of this continuity between generations is shown in Figure 2, and is the simplest possible hypothesis for the origin of biological diversity − I have labelled it as ‘bold simplicity’. This continuity of states (generations in this case) links evolutionary biology into the explanations used in physics, chemistry and geology – it is standard science.
The continuity between generations means that mathematics also fits well with evolution − evolutionary biology is one of the most mathematically developed parts of biology. This continuity, further and further back in time leads these microevolutionary processes to Darwin’s ‘theory of descent with modification’.
This is usually expressed as an evolutionary tree − even though it was known from 100 years earlier (the time of Linnaeus) that hybrids occurred. Thus, the tree is always a simplification. Changes certainly need not be equal between generations − Darwin expected rates to be variable depending on any number of factors. Nowadays, some biologists seem just to accept the evolutionary tree, but as scientists we expect the tree to be able to make predictions − the tree is a testable hypothesis.
Predictions and testing from new data
A mark of any scientific theory is that it makes predictions that can be tested when new data becomes available. For evolution, the availability of DNA sequences is now the most powerful form of evidence used for reconstructing evolutionary trees (and New Zealanders are among the leaders in this area). Consider humans and the great apes as an example. Based on morphological (and probably behavioural) evidence, Charles Darwin suggested chimpanzees and gorillas were our closest relatives. In the late 1980s, long lengths of DNA sequences (over 10,000 nucleotides) became available for several primates. A very small proportion of the data for (what is now) an inactive form of haemoglobin is shown in Figure 3; being an inactive copy means that most of the changes we see are chance events.
The human sequence is given at the top: chimps, gorillas, orangutans and rhesus monkeys are identical to humans unless an alternative nucleotide is shown. The figure strongly supports Darwin’s prediction that humans are most closely related to chimpanzees and gorillas. DNA sequences go further in grouping chimpanzees and humans, with gorillas just slightly older.
Theory predicts that occasionally there will be support for either human and gorillas, or gorillas and chimpanzees. These exceptions are very important because they allow estimates of the population size of the common ancestors – but that is well beyond our scope here. The fundamental point is that hypotheses about relationships can be tested as more data becomes available.
Now we have whole genomes from humans, chimpanzees and gorillas, and so the testing gets even stronger. For example, are there any new genes inserted into the human genome by a kindly creator or a group of itinerant space travellers that might lead to higher wisdom and intelligence?
Unfortunately, not. We have had to make do with millions of small changes to existing genes, but the idea of special new genes was worth testing.
As scientists, we must always be open to new ideas, but must subject them to testing. Alternatives to mainstream evolutionary thought are no exception, and Figure 4 shows one example of comparing predictions from the theory of descent and from intelligent design. The latter assumes some greater force has general oversight of the running of the Universe, whilst not getting actively involved in its day-to-day running. It is a bit like a mixture of personal trainer/coach/ referee that helps individuals, improves overall strategies, but also ensures the rules of the game [laws of nature] are obeyed. A simple prediction from intelligent design might be that it would be sensible to use similar enzymes that do the same function in similar environments.
Above: Figure 5. A range of hypotheses about evolution, ranging from continued spontaneous generation to full acceptance of natural law.
Until about 1700 it was assumed forms of living and non-living matter could interconvert. This was followed by a special creation phase that combined a mixture of a separate creator with Plato’s philosophy that ‘essences’ could not change. There is a range of intermediate steps until a fully naturalist condition is accepted.
There is a wide range of views and, in practice, everybody places themselves where they are comfortable. However, all of the views lead to predictions that are testable, and as scientists we must always try to test our favourite ideas.
For example, photosynthetic enzymes in plants that live in very hot and dry climates might be expected to be designed similarly. The example compares a cactus and a grass living under the same desert conditions (under-water stress) with a grass that lives in moist temperate conditions, that is, without either temperature or moisture stress. I am deliberately leaving it as a prediction (and not telling you the result) so that you can concentrate on the logic behind the test.
You could make a similar test with the proteins that make up hairs in a mammal. A polar bear and a snow rabbit living in the Arctic might be expected to have proteins well-designed to maximise insulation – at least compared with a sun bear or to a rabbit from more tropical conditions.
Where are we up to now with evolution?
I have emphasised the testability of all aspects of Darwin’s evolutionary theory, but obviously there are still many conflicting views. My approach is to show a spectrum of views (Figure 5), and look for the predictions that can be tested. We have already considered the traditional view of continued spontaneous generation, and then the mixture of Plato and special creation in the past, that arose around 1700.
From this point everyone will have their own variants of the alternatives − I find the version in the figure helpful.
There are those who are happy just to describe nature and not to consider origins – this was almost the norm in early science where origins (such as the origin of the solar system, or the origin of species) were outside the realm of science.
Again, there are others who accept that normal evolution works for all species, except for humans! They might argue, despite the extensive evidence, that there is nothing in the communication systems of the great apes that helps us understand the origin of human language.
Somewhere in the middle of the spectrum is ‘intelligent design’, referred to above, that accepts an ancient Earth and normal microevolutionary processes – its ‘just that some overall guidance is required’ – a bit of trial and error perhaps.
Another view, espoused initially by North American academic Marxists, was that ‘some major component was missing’. For example, they could not accept that competition would be sufficient for major evolutionary changes (macroevolution) – some ‘great principle’ (presumably Marxist) must be missing.
The official Catholic position is that everything is natural law, apart from two interventions – one for the origin of life, and one for the origin of humans. While individual Catholics probably vary as much as everyone else, the previous sentence refers to the official position. Finally, the mainstream Darwinian view is for 100 per cent natural law (with lots of interesting things still to discover).
My approach in teaching evolution, after covering the material here in more detail, is to:
• give this range of views
• say that I am comfortable with 100 per cent natural law
• say that you (the students) should place yourselves where you are comfortable for the present
• and then say that I will never ask you where you are ‘comfortable’.
I do add, however, that, being a science subject, I expect you to be able to give the standard evidence from a scientific understanding.
As far as I can gather, students do not find this approach threatening, and appear (to me) to accept such an overview as allowing them to study the subject without being challenged beyond their present comfort zone. The intent is to allow personal space to consider the questions, and to allow growth in the future. Note that all views in Figure 5, possibly excluding the last, have advocates of a religious persuasion.
It is important to remember than continued spontaneous generation was the accepted belief for most of the last 2000 years. The main issue is to see evolution as perfectly normal science that makes predictions that can be tested.
Indeed, in reading Charles Darwin’s autobiography, he repeatedly comments on the critical role of theory in science. He certainly fits into the scientific approach that Karl Popper espoused about generating hypotheses, making predictions, and testing the hypotheses – just in the way Popper thought the great scientists should do. Darwin actively looked for apparent weak points where his theory could have difficulties, and sought to understand the biology better. Darwin’s theory is excellent testable science, and continues to generate interesting and testable ideas.
About the author:
David Penny was awarded the NZAS Marsden Medal in 2000 in recognition of his outstanding service to science and the profession of science. He is a Fellow of the Royal Society of New Zealand and, in 2004, was awarded the Rutherford Medal in recognition of his distinguished contributions in theoretical biology, molecular evolution, and the analysis of DNA information.
You might also like to read:
- To the Ice
- Gene editing - a train coming down the track
- The gene editing future is here - Royal Society NZ panel formed
- The future of farming - teachers learn about agriculture careers
- The exponential generation - student workshops look to the future
- Network of inspiration: Assoc. for Women in Science