Monday, October 11, 2010

Palaeoporn 19

Another holy grail

Ignore the cute baby buffalo. See that unprepossessing roadside quarry with the people sitting on a pile of rubble (click to enlarge)? Well that rubble is the type section for the Chengjiang fauna.

Yeah! Not so unprepossessing now, is it?!

The Chengjiang fauna occurs in the Maotianshan Shale, a member of the Lower Cambrian Chiungchussu formation. They are found about 5 km northwest of the Fuxian Lake, and 6 km northeast of Chengjiang, Yunnan, China.

The age of the fauna is between 525-520 million years old, significantly older than the Burgess Shale (at 505 million years ago), and the Emu Bay Shale (approx 515 million years old).

The Chengjiang fauna rivals that of the Burgess Shale in preservation and diversity. Although not in difficulty to get to. The Burgess Shale is a 3 hour hike. Here, you can drive right to the outcrop!

In this shot (click to enlarge) you can see the quarry from photo one in the middle distance. There is a bus parked next to it with some people in the road. The hill to the right is Maotian Mount (hence Maotianshan Shale)

In this photo (click to enlarge), I am right at the back in short sleeves and a broad-brimmed brown hat (I'm in the same position in the first photo). I'm there partly because it's in the shade, and partly because there were some nice fossils there (all went to Nanjing University for study).

The depositional environment was delta front prograding eastwards into an open sea. Most of the fossil layers appear to be episodic events onto the marine muds in front of the delta, possible storm induced deposition of clays and fine sand. There's not much evidence of transport, so most organisms were locals and were buried by a series of turbidity flows.

This is a smaller quarry close by (click to enlarge). The dark colour is the fresh colour of the Maotianshan Shale. It is black when fresh, but rapidly oxidises to a tan colour on exposure to air. This is near the top of the Maotianshan Shale. The pick axe handle in the middle of the shot is marking the topmost Maotianshan Shales. The very top of the handle is resting against the first influx of sands, which coarsen upward until they are topped by a large lenticular sandstone at the top of the sequence above my head. This represents the prograding delta as it moves out over the muds of the Maotianshan Shale - just like the sands of the Mississippi Delta are prograding out into the Gulf of Mexico.

Monday, October 4, 2010

Evolution Among the Trilobites - Part 2

Meet the family. Estaingia (right) and Xystridura (left).

In Part 1 we looked at the growth patterns of the Early Cambrian trilobite Estaingia bilobata, using certain measurements from the head, or cranidium. In this part we'll compare and contrast those growth patterns with the Early Cambrian trilobite Xystridura templetonensis. You can click on any of the images and graphs to get a larger version.

The change through time in the appearance or rate of development of ancestral characters is known as Heterochrony, which comprises two basic phenomena (pay attention, there'll be a quiz at the end!):

Paedomorphosis - the retention of ancestral juvenile characteristics in a descendant adult. Paedomorphic forms usually pass through fewer morphological stages during growth than their ancestors (in other words less time as a juvenile you stop being a juvenile early, before you've shed all the baby characters). There are three types of peadomorphosis
Deceleration - the rate of morphological development is reduced during the juvenile phase (slower development, but the same amount of time is spent as a juvenile, so the same size as the ancestor, but you have not completed the juvenile development so have a baby-faced adult)
Hypomorphosis - the onset of maturity occurs at an earlier stage of development (same development but less time as a juvenile, so a smaller adult then the ancestor - still baby-faced though).
Post-displacement - a change in the timing of the onset of certain features, with one or more structures starting to develop at a later stage (the same amount of time is spent as a juvenile, so the same size as the ancestor, but the structures are smaller in the adult)
Peramorphosis - the appearance of ancestral adult features in the descendant juvenile stage (beards on a baby!). Here to there are three processes:
Acceleration - increasing the rate of morphological development (if maturity is also accelerated then adult will be smaller then the ancestral adult, if the onset of maturity is not affected, the the adult will be the same size as the ancestral adult).
Hypermorphosis - delayed maturity so the juvenile stage is extended (spend longer as a juvenile, so the adult is bigger)
Pre-displacement development of structures occurs at an earlier stage of development (same time as a juvenile, but structures more developed and bigger).
Before we look at a comparison between Estaingia and Xystridura, lets refresh what it is we are actually measuring (see Estaingiacranidium (or head) at right). The relevant measurements that we are interested in are; Cranidial Width (CW), which is literally the distance between the eyes; the Pre-glabella Field (PGF), which is the area in front of the large bulbous glabella (incidentally the pre-glabella field only runs to the shallow trench towards the front of the cranidium. In front of the trench is the doublure. That is folded underneath the cranidium in life, along the trench, and pops up during moulting); Glabella Length (GL), adding PGF and GL gives us a value for the length of the cranidium; Pre-Orbital Glabella (POG), which is the bit of the glabella that lies in front of a line drawn between the front tip of the eyes; and the distance between the end of the Axial (or Occipital) Furrow (the trench behind the glabella) and the back tip of the Eye Lobe (AF-EL). (Note that the figure is missing the free cheeks. This is because they are usually lost or repositioned during moulting. Trying to measure missing of displaced portion of the head would not allow accurate measurements. The measurements here, therefore, are all done on cranidia (without free cheeks).

The story is that Xystridura (Middle Cambrian) has evolved from Estaingia as represented by forms from the Emu Bay Shale (Lower Cambrian) through changes in the rate of development of certain characters, or hererochrony.

Now, to look at changes over time we will be comparing ratios, not individual characters. Individual characters will usually grow over time, but we are interested in how that character changes with growth. If the character increased with growth at a greater rate than other characters, then the ratio increases. If the character increases with growth at a lesser rate than other characters, then the ratio decreases. If the character increases at the same rate as other characters then there is no change in the ratio.

The first character we will look at is the distance between the back of the eye and the occipital furrow (AF-EL).

Measurements of meraspid Estaingia and Xystridura.
measurements adapted from McNamara (1981).

Here, the ratio AF-EL/CL compared with cranidial length is plotted with growth (cranidial length is taken as a proxy for growth) for meraspid (juvenile) Estaingia (yellow) and Xystridura (black). Here the smallest (youngest) meraspids have the largest ratio, and as they grow that ratio decreases. This means that, with growth, the feature is increasing at a slower rate than the overall growth rate. We can also see that Estaingia and Xystridura plot together. This means that the juvenile growth pattern is the same for both. The big difference is that the Xystridura meraspids continue beyond the transition from Estaingia meraspid to holaspid (adult) (yellow dotted line). The Xystridura meraspid to holaspid transition (black dotted line) occurs later than Estaingia and so Xystridura meraspids grow to a larger size before the holaspid stage than do Estaingia meraspids. So the onset of maturity is delayed and the juvenile growth phase has been extended.

The largest Xystridura meraspid appear to have a smaller AF-EL/CL ratio than Estaingia meraspids, but the spread of data is large.

Obviously, the character itself - the distance between the back of the eye ridge and the occipital furrow - increases throughout growth in absolute terms, but that increase is at a slower rate that the overall growth of other characters, so the ratio decreases.

No lets look at what happens during holaspid growth.

Measurements of Estaingia and Xystridura.
Xystridura measurements adapted from McNamara (1981).

The first thing to note is that the meraspid growth pattern of reducing AF-EL/CL ratio is halted at the meraspid-holaspid transition. In the holaspid growth phase the AF-EL/CL ratio does not change with growth, which means that it is now increasing at the same rate as the length of the head. Also the spread of data at the meraspid-holaspid transition for Xystridura meraspid settles down in the holaspids suggesting that the timing of the transition has a bit of slop in it.

The second thing to note is that Estaingia and Xystridura have the same growth pattern - exactly the same growth pattern. Having a similar growth pattern could be chance - maybe this is a common growth pattern for trilobites. But having exactly the same ratio values is unlikely to occur by chance. This is strong evidence of an evolutionary relationship. Xystridura has inherited this particular ratio pattern and values from Estaingia.

Note also the "NSW Estaingia". This is Estaingia bilobata from Cymbric Vale in New South Wales. It is slightly younger than the Estaingia bilobata from the Emu Bay Shale. There are few measurements, but this form plots on the same trend and ratio values as the others - albeit a larger form than the Estaingia bilobata from the Emu Bay Shale.

Lets look at another character, this time the ratio of width compared with length of the head CW/CL

Measurements of meraspid Estaingia and Xystridura.
Xystridura measurements adapted from McNamara (1981).

Here again are the meraspid measurements for Estaingia and Xystridura. Again the character ratio decreases with growth. This means that as the head grows, the width is growing at a slower rate than the length. Also, the Xystridura meraspid growth phase is extended. This time however, there does seem to be some difference between the Estaingia and Xystridura meraspid growth phases. Unlike the last example, where the two growth patterns seemed to be in step. here the Xystridura meraspids appear to change the growth pattern. The larger Xystridura meraspids appear to stop the trend of decreasing CW/CL ratio and, by the time they reach the meraspid-holaspid transition, they appear to be slightly increasing the CW/CL ratio.

Things should become clearer when we look at the complete growth patterns.

Measurements of Estaingia and Xystridura.
Xystridura measurements adapted from McNamara (1981).

Here the growth patterns are markedly different. It should be pointed out that holaspid growth changes usually occur at slower rates than for meraspid forms, so that holaspid trends are less marked than meraspid trends. The Estaingia growth pattern shows that the reduction in the CW/CL ratio occurring in the meraspid growth phase is slowed at the meraspid-holaspid boundary. In the holsapid growth phase, the growth pattern shows that cranidial width continues to grow at a slower rate than cranidial length, but the difference between the growth rates is reduced somewhat.

However, for Xystridura, the apparent change in growth rate that occurs in the latest meraspids is carried into the holaspid growth phase, where the CW/CL ratio not only equalises (that is the width is growing at the same rate as length) but the width growth rate actually appears to be slightly exceeding the length growth rate.

Critically this change in Xystridura growth pattern occurs in the meraspid phase, but at the point where Estaingia would be expected to transition into the holaspid growth pattern. It isn't very clear in the graph (so click on it to enlarge) but the new growth pattern of the larger Xystridura meraspid after the Estaingia meraspid-holaspid transition (yellow dotted line) shows that the width is actually growing faster than the length (CW/CL ratio is increasing). At the Xystridura meraspid-holaspid transition (black dotted line) this growth rate actually slows down so that the width growth rate only slightly exceeds the length growth rate.

The NSW Estaingia actually plots between the Estaingia and Xystridura, suggesting that this change was underway by the time of Cymbric Vale sediments deposition.

The one thing that controls the cranidial width is the large bulbous thing in the centre of the head - the glabella. What these measurements are telling us is that during Estaingia meraspid growth, the glabella increases in width at a slower rate the the increase in the length of the head. During holaspid growth, the rate of growth of glabella width actually increases, but still not to the rate of increase of the length of the head - the glabella has increased the pace of its growth, but not enough to match the rate of growth in length the head is achieving - resulting in a slowing of the trend that is reducing the ratio of width to length, but not enough to stop the trend. In Xystridura, the glabella width actually starts to grow at a faster rate than the length - it is increasing in width ata faster rate than the head is growing longer - and so the ratio starts to increase.

If the glabella is changing size in one direction, maybe it's changing in another. Lets see.

The length of the head is composed of two measurements, the glabella. and the field in front of the glabella - the Preglabella Field (PGF) Any change in the glabella will affect the PGF. If the glabella increases in size at a faster rate than the length of the cranidium, then the PGF will reduce, if the glabella increases in size at a slower rate than the length of the cranidium, then the PGF will increase. If the glabella increase in size at the same rate as the length of the cranidium, then the PGF will remain the same. Measuring the PGF then, provides a proxy for glabella length growth.

Measurements of meraspid Estaingia and Xystridura.
measurements adapted from McNamara (1981).

For Estaingia meraspid, the PGF is increasing with increasing meraspid size. This means that the glabella is growing at a slower rate then the length of the head resulting in an expanded PGF. By the Estaingia meraspid-holaspid transition, this pattern has stopped.

The Xystridura meraspid growth pattern follows the Estaingia meraspid pattern in showing an increase in the size of the PGF up to a certain point. Then, just before the Estaingia meraspid-holaspid transition this trend reverses, and the PGF reduced in size. This means that the glabella is now increasing in size at a greater rate than the length of the cranidium, resulting in a reduction in the PGF.

What does that mean for the holaspids?

Measurements of Estaingia and Xystridura.
measurements adapted from McNamara (1981).

Lets take Estaingia first. After the meraspid-holaspid transition, the PGF ratio decreases slowly. This indicated that the glabella is growing at a faster rate than the cranidial length, so the PGF is getting smaller, but only slowly.

After the Xystridura meraspid-holaspid transition, Xystridura shows the same pattern of slowly reducing PGF and so slowly increasing glabella. But, as the Xystridura meraspid phase lasted longer, the trend to reducing PGF (increasing glabella) starts in the meraspid period and is accentuated. So much so that, by the time of the Xystridura meraspid-holaspid transition, the PGF is so small that even at the slowly reducing rate of the holaspid form, the PGF reached zero shortly after the Xystridura meraspid-holaspid transition. This means that the the glabella has been growing at such a faster rate than the cranidium, that it is so close to the front of the cranidium by the time of the meraspid-holaspid transition, that it quickly reaches the front of the cranidium early in the holaspid phase. PGF = 0.

Interestingly, the NSW Estaingia plot with the other Estaingia. This indicated that this change had not yet started by Cymbric Vale time.

All it took to produce a descendant form that looks significantly different from the ancestral form is the simple process of slightly delaying the onset of the holaspid phase, resulting in some of the slower holaspid growth pattern being incorporated into the faster meraspid pattern. In other words the large bulbous glabella of Xystridura is caused by a portion of the holaspid growth pattern being incorporated into the meraspid growth phase. "But wait a minute", you say (or you would say if you were paying attention) "the first character didn't change at all, and that involves the glabella - it measures the distance between the back of the eye ridge and the side of the back end of the glabella". That's true, but the growth of the glabella is focused towards the front of it. It is the front portion that has grown, not the back portion. So the back remains unaffected by the change.

Why is the glabella important? Well, that's where trilobites keep the stomach. So big glabella, big stomach.

So, what type of heterochrony is this? Go back to the list at the start and work it out.

I'll give you a clue. We have delayed maturity so the juvenile stage is extended. As a result, the the adult is bigger.

We are dealing here with Peramorphosis, and more specifically Hypermorphosis. One or more populations of Estaingia has evolved, by Hypermorphosis into Xystridura. An event that occurred around the Lower-Middle Cambrian boundary, during a time of eustatic sea level fall that would have reduced shallow-water living space.

The take-home message isn't that trilobites are cool (they are), but that this brings out a very important point about evolution, and a good refutation of the old creationist canard, "if evolution is true where are the half-way transitionals? The half reptile-half bird?"

What these results show, is that evolution doesn't happen to all features at the same time, or at the same rate, producing a neat half-and-half transitional form. Some features change relatively rapidly (the expansion of the frontal glabella), some features change relatively slowly (head width to length ratio), and some don't change at all (the distance from the back of the eye to axial furrow distance as a ratio of head length). So there isn't a transitional which has all features exactly half way between the ancestral and descendant forms. What we find are transitionals with a mix of features depending on the rates at which those features are changing. We should not expect to find exact half-and-half transitionals. Evolution doesn't work that way.

de Beer called it "Mosaic Evolution". That isn't an excuse for the lack of half-and-half transitionals, it's a description of how evolution operates.

McNamara, K.J. (1981) Paedomorphism in Middle Cambrian xystridurine trilobites from northern Australia. Alcheringa, 5: 209-224. DOI: 10.1080/03115518108567002