Feedback archiveFeedback 2020

Paleozoic Corals and Lunar Recession

Responding to Jerry Coyne

Published: 20 June 2020 (GMT+10)

C.R. wrote in with a question about coral reefs and coral growth rates:

Photo by Joachim Schevencoral
Figure 1: A common Indo-Pacific coral, Pocillopora damicornis, growing on a shoe. Yes, some corals can grow quickly, but this does not necessarily mean that all do, nor does it mean that all coral reefs grow quickly.

I have a friend who is a geologist here at [name of college]. He is open to young earth but read Jerry Coyne’s claims about reef growth being slow. What is your most persuasive article on this common old-earth argument? I think he could be a potential YEC if I give a good answer to some of his questions.

Here is how my friend explained Coyne’s excerpt:

They correlated the radiometric dating to coral growth rings. They got a radiometric date of 380 Ma for a Devonian coral reef, and they also figured out somehow that the rotational speed of Earth is slowing down due to tidal friction. The length of a day increases by 2 seconds every 100k years, therefore there had to be more days in a year in the distant past than in present times.

If we do the math, 380 million years ago there was around 396 days a year, each 22 hours long. Corals today not only have annual growth rings, but daily growth rings (I have no idea how can they identify which is the marker ring what closes the 1 year cycle though), so we should be able to count these rings in the fossil coral, and if the radiometric dates are correct, than we should have 396 daily rings each year. After they counted the rings, they got around 400 daily rings for an annual growth period, which is very close to the calculated 396.

It seems like they were able to biologically confirm the accuracy of the radiometric age of the sample.

Dr Robert Carter, who has a PhD in coral reef ecology, answers.


There is a lot of confusion on this topic. The issues deal with the difference between a coral reef and an individual coral, tides and tidal braking, the recession of the moon, the palaeontology of extinct animals, how the environment influences the growth of a coral, and various types of “banding” and “growth rings” seen in coral skeletons.

I know that Coyne is an extremely dogmatic atheopath, so I love it that I get an opportunity to correct something he wrote. But I wonder what your friend’s real authority is? Is he truly searching, or is this an excuse he is using to resist the idea that the earth was created, and not billions of years ago? Maybe you can read him better than me, so I dug deeply to come up with the best explanation I could muster and I spent a long time looking into the scientific literature. Maybe our joint efforts can help a struggling Christian.

It turns out that Coyne is citing some material from the 1960s. John Woodmorappe reviewed Coyne’s book Why Evolution is True. In the section Coral bands, Earth’s rotation, and isotopic age-dating methods, he said:

One novelty of this book is Coyne’s resurrection of the claimed agreement, based on a cited 1963 study, between coral growth bands and the inferred 400-day Devonian year (pp. 24–25). He is not telling the full story. More recent studies [reference1] include those that cast doubts on the reliability of such methods. There are problems with such things as the placement of band boundaries according to unconscious self-fulfilling preconceptions, the lack of bands grown during stressful years, marine life whose bands are not periodic in the first place, uniformitarian assumptions about the paleoenvironments from which the sampled fossil organisms had been taken, the “reinforcement syndrome” or “consensus” effect caused by the tendency to publish studies that seem to confirm previous ones, and still other problems that could be mentioned.

This is an excellent summary of the problem, but Coyne’s argument can be broken down even further. Let’s study up on a few key issues before getting into the meat of the argument.

Coral vs. coral reef

People have a hard time separating the growth rate of an individual animal (a coral) from the net growth rate of a community of animals (corals, sponges, bivalves, algae, and other things). The growth of a reef over time depends on the growth of many things. But you also have to subtract the negative contribution of those things that are degrading the accumulating calcium (e.g., parrot fish, boring clams, boring sponges, and even rainwater). The rate of growth of a reef has little to do with the rate of growth of individual coral animals.

I corrected an old example that CMI used to use in my talk Spectacular Coral Reefs. Years ago, somebody found a coral growing on a shoe, and many people associated fast coral growth rates with fast reef growth rates. But the reason this is not a good argument is that the coral in the photo is one of the fastest growing corals (Pocillopora damicornis). It does not create a thick skeleton (you can crush it in your hands) and is a very weedy coral. On the one hand, yes, coral colonies can grow rapidly, and the reef must be able to reach the ocean’s surface in only a few thousand years within the biblical timeline. But, on the other hand, just because a single coral species can grow quickly does not necessarily mean that entire reef systems do. We have to treat each case separately.

Robert Carterbaby-coral
Figure 2: a baby coral, Orbicella faveolata, from the Florida Keys. The diameter of the cup is less than a half centimetre.

Just for fun, here’s a picture of a baby coral:

Figure 2 is a picture of a surviving recruit from a coral reproduction experiment my laboratory was performing in the late 1990s. We had collected eggs and sperm from the mass spawning event in the Florida Keys and were attempting to raise corals from the resulting larvae. This little guy already has several polyps and the tentacles, which are retracted in the photo, are already full of the photosynthesizing algae (brown specks) the corals farms for food. This individual is less than half a centimetre in diameter. It did not live more than a few weeks, but such is the fate of such tiny things in any environment. Can you see those spikes beneath the translucent skin? If you zoomed up on those, you would see very small crystals and patterns of crystal growth.

Reefs in the geological record

Another thing many people, including some geologists, struggle with is what a ‘reef’ is in the fossil record. There are no Devonian coral reefs. Oh, there are Devonian corals, but coral reefs as we know them today do not appear in the fossil record until very recent times. And the corals from the Devonian are not related to modern scleractinian (hard) corals. Devonian rugose and tabulate corals are 1) extinct, 2) had a different growth pattern (they do not have 6-fold symmetry), 3) had different biochemistry (calcite skeletons instead of aragonite),2 and 4) never formed massive, interconnected limestone frameworks. Thus, one must do a lot of extrapolating if one wants to correlate the growth bands on these extinct animals to those of modern reef corals.

The lack of coral reefs in the rock record is a testament to the young age of the earth. Think about it. During the height of the ‘last’ Ice Age, the Great Barrier Reef was dry ground. In fact, the crest of the limestone ridge that was to become the GBR was at 100 meters in elevation above sea level. The entire GBR formed in less than 10,000 years in the secular timescale. And, once those reefs hit sea level, they stopped growing vertically, so they took less time than that to form. Evidence shows there was but one Ice Age, after the Flood, that ended approximately 4,000 years ago. Thus, the evolutionary timeframe and the creationist timeframe are highly similar. I have no problem believing the GBR formed in just a few thousand years.

So where are the massive coral reefs in the rock record? The largest coral colony I have seen, in a photo, at a university in Germany, is maybe 2 meters across.3 This could easy to grow in the 1,600 years between Creation and the Flood. The “massive” rugose corals Berkowski and Belka were studying (see below) were all of 30 cm in diameter.

Growth rings and bands

To understand what Coyne is talking about, we have to understand the different types of coral growth “rings”. If you cut a thin slice out of a large coral skeleton and put it on a light table, or if you X-ray the slice, with some species you can see density bands. These are thought to be annual markers, and in many cases this should be true. As a polyp grows, it occasionally lifts up in the skeletal cup in which it is growing and puts down a new floor. Staining data indicates this follows a lunar signal.4 But the skeletal elements can also be thickened over time. So, there are multiple factors that affect skeletal density. The rate that this happens affects the density of the skeleton in that area. Thus, when the polyps are extending quickly, they leave larger spaces between consecutive floors and the skeleton is less dense in that area. This is a seasonal effect, but it is not a growth ‘ring’, like in a tree. Figure 3 is a beautiful example from a recent paper.3 Note that this was from a CT scan and the image was adjusted heavily to increase contrast:

Figure 3: A CT scan of a thin slice of Porites sp. coral skeleton from the back-reef lagoon of Palau shows ‘density’ bands that follow an annual cycle. Some corals literally sit on top of their skeleton and only deposit calcium beneath the tissue layer, while others (as in the species pictured) have a porous skeleton surrounded by tissue. The rate and pattern of calcium deposition affects the density of the skeleton. And different coral colonies can live is very different environments, from shallow mud banks to storm-tossed reef crests, to deep oceans. Thus, this pattern is not so clear in all species or in all places.
Photo: Wikipedia/Mark A. Wilsonfossil-rugose-corals
Figure 4: A fossil Devonian ‘rugose’ coral. In the photo on the lower right, one can see darker bands encircling the coral skeleton. Are these annual or monthly bands? Or do the bands have anything to do with the calendar at all?

But what are the environmental cues that cause density banding? As Berkowski and Belka (2008) said:

In living corals growth-band formation is largely mediated by seasonal variations in temperature but other environmental factors, such as nutrient supply, salinity, depth, water turbidity or insolation can also influence the growth of the coral skeleton. This is why it is not possible to correlate more dense and less dense bands of skeletal tissue with a certain environmental condition … Dense bands are usually associated with warm water and low insolation, while the low-density bands are formed in the season of relatively cooler water temperatures and high insolation. But reversal in band formation can also be observed.5

Thus, any number of environmental signals can cause density banding. The bands are not necessarily annual, especially in the weedier corals living in marginal environments. Keep this in mind as we get into additional details below.

This is not the only banding we can see in coral skeletons. Corals growing near the mouth of a river can pick up tannins in their skeleton. This can also provide a seasonal signal, if, say there is a rainy season and a dry season. The skeleton might fluoresce dimly under black light, with seasonal bands or bands associated with the runoff from major storms.

There are also daily growth patterns, lines and marks and things like that, but you can only see those under magnification. If you can correlate the daily pattern to a seasonal signal, you could theoretically count up the number of days in a year. Yet, the argument has nothing to do with the growth ‘rings’ and ‘bands’ I discussed above. Instead, it revolves around some rings one can see on the outside of the fossils.

Now that we understand the biology of the corals, we must talk about gravity, tides, and the length of a day in the evolutionary deep-time model.

The recession of the moon

The argument about banding in Paleozoic corals revolves around the fact that the moon is slowing the earth down. We know this is true, but the amount of slowing is highly dependent on historical models with many unknown parameters.

Figure 5: The tidal ‘bulge’ is caused by the ocean water closest to the moon accelerating toward the moon faster than the earth itself. The ocean water farthest from the moon accelerates more slowly than the earth. The arrows are in reference to a person on the earth. To a person standing on the earth, it looks like the water is rising and falling as the earth rotates beneath the moon (the “Satellite”).

It is well known that the moon causes tides. Galileo (1564–1642) rejected the ‘occultic’ idea of some invisible force from the moon acting at a distance. He thus incorrectly used tides as proof of the earth’s rotation. Bede (672–735 AD) had it right: the moon somehow caused the tides. Newton’s Law of Gravity (1687) told us how it works in reality. The earth and moon are attracted to each other. Thus, they are always moving toward one another. But, since they are also moving sideways, they constantly miss each other and end up traveling with respect to one another in an elliptical manner. It they were moving too quickly, they would fly apart. If they were moving too slowly, they could crash together. But the two orbit each other in a perfectly balanced, completely stable way.

The gravitational acceleration toward the moon is different at different places on the earth. When the moon is overhead, you are 12,700 km (nearly 8,000 miles) closer to the moon than you will be 12 hours and 25 minutes later.6,7

Since the earth is solid, however, the whole thing moves as one giant sphere.8 But water is free to move across the surface. Water can’t lift off the surface and water does not stretch, so a “high” tide happens when water slides toward the moon from areas not directly under it. Figure 5 should help explain this.

You don’t sense that you are ‘falling’ toward the moon, but the oceans give it away. Water on the side closest to the moon accelerates the fastest. Water on the far side from the moon is ‘left behind’ as the earth falls. To a person standing on the earth, this gives the appearance that the water’s surface is rising and falling as the earth turns beneath the moon.

The oceans cannot continually lap the earth because there are continents in the way. Given a 12-hour period and a convoluted continental system, the net result is a complex mix of tidal ranges across the world (figure 6).

Figure 6: Tides of the world. Average tidal amplitude is indicated by colour. Red denotes extreme tides. Blue shows areas with hardly any tide (so-called ‘amphidromic points’). The white lines show where high or low tide strikes at the same time. The curved arcs around the amphidromic points show the direction that the tides circle.

It is those red areas in figure 6 that are of most concern to us. When water sloshes up onto a shallow shelf, it slows down and bulges up. But since the earth turns more quickly than the moon orbits, the tidal bulge is always lagging the spin of the earth. Thus, it drags on the earth, slowing it down. But that drag is felt by the moon, which is pulled ahead.

We can measure this. NASA astronauts left corner mirrors on the moon. Today, we can bounce laser pulses off those mirrors and precisely time how long it takes the photons to arrive back on earth. This is simple matter of operational science and we know the moon is receding at 3.808 +/- 0.004 cm/yr. We also know that the length of a day on earth is slowing down by about 0.002 sec/century. One way we can estimate this is by comparing what we see today to 2,700-year-old Babylonian astronomical observations.

Based on the current rate of recession, we can estimate the age of the moon and the amount of slowing of the day in the distant past. This is not a simple calculation. The rate strongly depends on the amount of mid latitude, shallow oceanic shelves. This is assumed to have changed greatly over evolutionary time. It also depends on the proportional distance of the moon from the geocentric point. If the moon were orbiting at the same rate as the earth turns, there would be no tidal braking. Currently, the moon is about 10× farther away (385,000 km) than the point of geosynchrony (35,786 km), but this depends on the rate of rotation of the earth. A faster-rotating earth would require a faster-orbiting satellite to obtain geosynchrony. Thus, a geosynchronous satellite would have to orbit closer in. If there were 22 hours in a Palaeozoic day, geosynchrony would be about 95% of the current value. That doesn’t sound like a big difference, but it you are trying to extrapolate out to billions of years you would need to know this. Also, the effect of tidal braking is inversely proportional to the sixth power of distance, e.g. it is very distance-dependent, so the calculations have to be spot-on to be at all accurate.

Yet, as the relationship between a month and a day changes, if the earth-moon system is ever in synchrony, extreme harmonic oscillations in the tidal cycle might result. So, the amount of braking depends on the changing period of the lunar month, the changing length of day, and the relative configuration of the continents. At current rates, the moon cannot be more than 1.3 billion years old. Any further back, and the moon would so close to the earth that it would shatter. This is because the gravitational tug of the earth on the near side vs. the far side would be greater than the self-gravitation of the moon itself. This is called Roche Limit.9

But if the earth is slowing down, where is all that energy going? The First Law of Thermodynamics tells us that energy and mass must be conserved. If energy is transferred, it has to go somewhere. In this case, the kinetic energy of the earth is being transferred to the moon. The rotational energy of the earth is being turned into gravitational potential energy and the moon is moving farther away from us. I use that terminology on purpose. Energy is being transferred to the moon, but it is not ‘speeding up’. In fact, the length of a month is increasing because things that are further away orbit more slowly (compare the 88-day orbit of Mercury to the 248-year orbit of Pluto).

Another way to think about it is to use the Law of Conservation of Angular Momentum. Angular momentum = mvr (mass × velocity × radius). If the earth’s angular momentum is lessening because its velocity is decreasing, the moon’s angular momentum must increase by the same amount. But, not only is the moon tidally locked, with one face permanently pointing toward the earth, you would not expect it to pick up ‘spin’ due to tides on earth. Thus, the distance between the earth and moon (r) increases, i.e. the moon is receding.

So now we have a method to decrease the number of days in a year over time and also to increase the number of days in a month. The ‘banding in Paleozoic corals’ argument focuses on the number of days in a year, but most people ignore the number of months in a year or days in a month.

What annual bands?

I can now address Coyne’s statements. The argument he is using is extremely outdated. In fact, it was contradicted in the scientific literature just a few years after it was made, in the 1960s. I detailed the calculations and inferences in my article ‘Ancient’ coral growth layers, in which I was responding to similar claims made by Biologos. It seems Coyne is just parroting some talking points. They seem strong, but not after carefully analysing them. Please refer to the article for details.

But here’s the gist: you can examine skeletal patterns in many corals. But there is no “this is the dividing line between one year and another” marker, so you have to do a little guessing and then take an average.

But when we think of coral reefs, most people think of clear tropical waters bathed in bright sunlight. If you know about the reproductive habits of corals, you may know about the mass-spawning episodes that happen after sunset in the middle of the summer (I have witnessed many). However, there are also corals that live in more marginal environments. They tend to be weedy, small, and they do not do that mass-spawning thing. Instead, they generally have internal fertilization and will hold the eggs inside the polyps until they develop into planula larvae. Larvae are released on a monthly schedule, following the moon.

In his comprehensive review of the paleoecology of Paleozoic corals, Scrutton (1998) mentioned the day-length argument only in passing.10 And yet Scrutton is one of the key players in this argument. His work from 1964 was incredibly important and was a significant improvement on Wells’ work from the year prior.11 Wells was one of the giants among coral reef scientists in his day. Yet, Scrutton (1964) said he found it impossible to demarcate annual bands, casting a strong negative light on the claims of Wells. But Scrutton did notice strong monthly patterns.

If the bands are monthly, there is no annual pattern to be seen, and so nobody knows what Wells was talking about. Referring to the quote from Woodmorappe at the beginning of this article, Wells might have been subject to “unconscious self-fulfilling preconceptions”. If nobody can back up his claims because there are no clear “annual” markers, we have to reject his methods.

Thus, the whole argument devolves into a circular system that assumes X number of days in a Devonian year, combined with a huge guess about how many growth lines are seen in a ‘year’ of coral growth, while ignoring how many days were supposed to be in a Devonian month. If you don’t make those assumptions, however, you get the same number of days in a month as we have today. I explained this in my ‘Ancient’ coral growth rates paper12 and follow-up communications13 in the Journal of Creation.

If those are monthly signals, the number of days in a month is about the same as today. If you ignore the circular reasoning about the length of a Devonian day, the number of days in a year is about the same as it is today. The whole scheme falls apart.

References and notes

  1. Rosenberg, G.D. and Runcorn S.K. (Eds.), Growth Rhythms and the History of the Earth’s Rotation, John Wiley and Sons, London, New York, 1975. Return to text.
  2. Calcite and aragonite are different polymorphs of CaCO₃. Aragonite is metastable and tends to turn to calcite over time when exposed to water. When this happens, there are obvious changes in the skeleton, like infilling and/or the creation of a ‘mirror image’ of the coral where the original dissolves away and leaves behind a calcite negative. There are good reasons to believe the original skeleton of the rugose corals was calcite. Return to text.
  3. Scrutton, C.T., Periodicity in Devonian coral growth, Palaeontology 7(4):552–558, 1964. Return to text.
  4. DeCarlo, T.M. and Cohen, A.L., Dissepiments, density bands and signatures of thermal stress in Porites skeletons, Coral Reefs 36:749–761, 2017. Return to text.
  5. Berkowski, B. and Belka, Z., Seasonal growth bands in Famennian rugose coral Scruttonia kunthi and their environmental significance, Palaeogeography, Palaeoclimatology, Palaeoecology 265:87–92, 2007. Return to text.
  6. Since the moon orbits the earth in the same direction as the earth turns, one lunar day is 24 hours and 50 minutes long. Return to text.
  7. Gravity is inversely proportional to the inverse square of the distance, so there is greater acceleration toward the moon on the close side and less on the far side. The difference is the tidal force, which is inversely proportional to the inverse cube of the distance. The interested reader can do the math or look up the answer online. Return to text.
  8. There are also ‘earth tides’. Daily and diurnal bulges in the earth occur due to the presence of the moon and sun. But the periodicity is out of step with the oceanic tides. Also, the degree is often smaller. And, even if the earth bulges up by a meter, in places, the water in that region would be lifted by the same amount, so you could not see any effect on oceanic tides. Earth tides do have to be factored in when performing calculations with the GPS system and when predicting particle trajectories in the Large Hadron Supercollider, however, so it is not like the effect is not important. Also, ‘moon tides’ are probably what keeps the face of the moon locked into place. Return to text.
  9. Henry, J., The moon’s recession and age, J. Creation 20(2):65–70, 2006; Return to text.
  10. Scrutton, C.T., The Palaeozoic corals, II: structure, variation and palaeoecology, Proceedings of the Yorkshire Geological Society 52(1):1–57, 1998. Return to text.
  11. Wells, J.W., Coral growth and geochronometry, Nature 4871:948–950, 1963. Return to text.
  12. Carter, R.W., ‘Ancient’ coral growth layers, J. Creation 26(3):50–53, 2012; Return to text.
  13. Carter, R., ‘Ancient’ coral growth layers: yearly or monthly, J. Creation 28(2):55–56, 2014. Return to text.

Helpful Resources

Readers’ comments

Michael B.
Mount Carmel caves which are claimed to show 400K years of human evolution in a single setting are also noted to be situated in one of the best preserved fossilized reefs of the Mediterranean region. I noted the mention of this reef when studying the site for a class I'm doing on how world view impacts your interpretation of the facts. I would be interested in your comments about this reef (for one, is it a true reef?) Thank you for your continued ministry to the Truth of God's Word. Your Brother in Christ, Michael.
Robert Carter
The limestones of Mt Carmel are yet another example of a non-coral-reef. Within those rocks are many horn corals and bivalves trapped within a muddy lime matrix. There is no reef core, no reef structure, no reef framework builders, etc. It is important to note that the term 'reef' is used by geologists, but not 'coral reef'. This is a trap for the unwary.
Ralph W.
If there was no death before Adams sin, i.e. all of God's creation was good, would this not negate the idea of long age dates being attached to coral formations?
Robert Carter
I'm not sure that the two ideas go together. It is not the presence of sin, per se, but the historical record of Genesis that tells us how old things are.
Dustin B.
Fantastic article! Thank you so much for this information.
Nathan G.
One important thing to keep in mind is to bring other disciplines into play, too. Don't get myopic in your focus. Try to give an open and honest answer to your friend's question, but also pull him out of his comfort zone. Creationists tend to get sucked into discussion of a single detail or piece of evidence as if it alone will provide a knockout punch to an evo's beliefs. But the large picture is just as important, since the contradictions to evolution occur in every single discipline. Remember, even a PhD expert in any given area is effectively a novice in all other disciplines. You could use the article by Russell Humphrey's on this website with the top 100 evidences against evolution as a start. Carbon-14 and the unreliable assumptions of radiometric dating methods is always a good place to start. The the amount of salt and sediment in the oceans. Then erosional rates that would flatten continents and destroy all fossils in a very short time evolutionarily speaking. There are lots of ways to go, depending on how your friend reacts.
Kenton Y.
A question occurs to me: What would tide patterns and heights have been like on a presumed single-continent-with-larger-ocean, pre-flood world? Presumably less complex and more uniform. Would they also have been greater? Intuitively, it seems that with greater room to "slosh around," the tides would be higher. Has anyone tried to estimate this that you know of?
Robert Carter
This would be a fun topic to explore, but the possibilities are nearly endless. Everything depends on the geometry of the situation: the presence of bays, inlets, and shallow shelves, and how much 'fetch' (long stretches of open, uninterrupted water) exists and where.
Geoff C. W.
It constantly amazes me how people will latch on to one or two things they have heard about that are claimed to 'prove' evolution, while ignoring the hundreds of well-founded evidences that the earth is young.
Robert Carter
For interested readers: /age-of-the-earth

Comments are automatically closed 14 days after publication.