Download Variation - Paleontology - Lecture Notes and more Study notes Geology in PDF only on Docsity!
The History of Life
Variation and Evolution
OBJECTIVES
This exercise is designed to introduce you to the concept of variation in populations and its importance in evolution and natural selection. To accomplish this you will assess (by measurement and counting) the variation in two identifiable features of some fossil brachiopod shells and decide whether the samples likely represent one large population or two distinct, separate populations.
VARIATION
No two individuals of a species are identical. Variations may be due to genetic or environmental differences, or simply to chance. The presence of these differences dictates that the description of a single individual is not sufficient to describe an entire species’ morphology, ecology, development, or anything else. Instead, the description of many individuals taken together defines a range of variation that encompasses the species.
Variation in morphology between individuals of the same species is the most obvious kind of variation. For example, domestic dogs ( Canis familiaris ) show a wide range of morphological variation from Chihuahuas to Great Danes. Most studies of evolution, especially those based on the study of fossils, rely exclusively on morphological variation. However, many aspects of organisms may vary, including behavior, ecology, genetic makeup, and growth patterns.
The standard unit of research concerning variation within and between species is the population. A population may be defined as a group of freely interbreeding individuals that share a common gene pool, i.e. , every individual can potentially mate with any other individual in the same population. Typically, species are comprised of several populations that may be partially or completely isolated from each other geographically. All individuals of the same species, however, are capable of interbreeding and producing fertile offspring given the opportunity. For example, two populations of elk may be separated by a mountain range across which only a few individuals typically travel, or two populations of Brazilian rain forest plants may be separated by the Amazon River, which its pollinators are seldom able to cross; however, they are still part of the same species.
Population variation is important to biologists because variation within a population of
organisms is the cornerstone upon which natural selection, the mechanism that
Charles Darwin proposed to explain the fact of evolution, is based. As such, variation within and between populations forms the basis of investigations of the patterns, processes, and rates of evolution.
TYPES OF VARIATION
Variation within a population may be of four types: 1) age-group (ontogenetic) differences; 2) genetic differences; 3) environmental differences; and 4) chance.
- Age-group differences are those differences produced by the presence of individuals of different ages in a population, e.g., adults and children together at Disney World would vary widely in height, but adults or 6-year-olds taken as separate populations would vary much less than the group as a whole. Figure 1 below shows variation in the distribution of shell lengths of a population of blue mussels due to age differences. Note that the distribution of differences can change during the breeding cycle of the organism.
Figure 1. Relationship between age and size in a population of Mytilus edulis from Scotland. Shaded histogram is from April, outlined one is the same population in November.
At the least, age differences will appear as variation in size, but often they are manifested as differences in proportion. The proportions of a human infant’s body are quite different from those of an adult, e.g., compare the relative size of an adult’s head to that of an infant’s. Human adults are behaviorally different from human children, too.
- Different individuals typically possess different sets of genes. [Identical siblings and clonal organisms (such as coral or grasses) are exceptions.] Some variation within a population is clearly the expression of these genetic differences. Hair color, eye color, and number of fingers are all human genetic traits that vary. (Some traits obviously vary more than others.) Another example of variation reflecting inherent genetic differences is sexual dimorphism , where males and females of a species look different from each other. Sexual dimorphism is often, but not always, expressed as a size difference. For example, female rabbits are typically larger than their male counterparts. Sexual dimorphism in fossils presents a particular problem because
Inference : Those organisms which survive to successfully reproduce will be those with features that happen to be better suited to local conditions at that time. Heritability ensures that at least some of these features will be passed on to the next generation.
Therefore, individual variation is the raw material upon which natural selection acts. Through differential birth or death of individuals, the mean of a population will tend to shift toward whatever variants are most adaptive for that particular time and place. This is natural selection.
VARIATION IN FOSSIL POPULATIONS
Fossils provide us with an important opportunity; they are our "window" into biological variation in the past. By comparing variation within and among fossil populations we can recognize and attempt to interpret the process of evolution. Unfortunately, problems inherent in the fossil record tend to smudge and obscure this window, distorting our view of the past by influencing what we see. These distortions can be attributed to three factors: 1) the incomplete nature of the fossil record; 2) the differential preservation of hard body parts; and 3) the physical distortion of fossil specimens.
The fossil record is complete neither in time nor in space. Unconformities caused by breaks in sediment accumulation and/or erosion of existing sediments leave gaps in the history of life as recorded in the earth’s crust. Furthermore, some environments are much more likely to be preserved, which means that organisms in environments with high preservation potential are likely to be better represented in the fossil record. (Remember the marine bias discussed in Lab 1.)
Most of the fossil record is comprised of parts of organisms... those parts which, because of their composition, are amenable to preservation. For example, the hard, mineralized skeletal components of many marine mollusks ( e.g., clams and snails) are commonly preserved. Some plant compounds, e.g., the material coating pollen grains, are equally hardy. Soft bodied organisms like worms are far less well represented in the fossil record. Examination of variation in fossil populations is therefore often limited to the measurement of hard-part variation. Keep in mind that what is preserved is only a part of the organism. There is also a bias against the preservation of small sized mineralized parts, resulting in the potential loss of juvenile individuals from a population during fossilization. This loss of information can result in lower variation in fossil populations when compared to the variation in modern, living populations.
Finally, those organisms for which we do have a record may be physically distorted by compaction or crushing before, during, or after fossilization. Organisms with several distinct hard parts may be broken up and separated (disarticulated) prior to burial, and fossils may be stretched, bent, or smeared like soft plastic while they are buried in the earth’s crust. Therefore, like preservational biases, distortion can alter the range of morphological variation originally present in a population.
HOW TO MEASURE VARIATION
If one were to use only words and pictures to describe and compare organisms, the task would be difficult and far too subjective. To make your study more rigorous, it is helpful to use a combination of numerical techniques that allow more objective comparisons of fossil specimens. These make an account of population variation quantitative, repeatable, and easier to communicate to others. Numerical methods, however, should not be used to the exclusion of careful, detailed description and illustration, for numerical methods can also be misleading if used or interpreted incorrectly. During this lab period you will gain experience with some of the most commonly used techniques for assessing variation. As mentioned before, paleontological studies rely almost exclusively on variation in shape (morphology) of hard parts of organisms, but biological studies can also include variation in the genetic code of individuals.
Morphological variation is measured and described by identifying characters on each individual in your sample population and evaluating the amount and nature of variation in each. A character is any feature of an organism that can be quantified, such as by linear measures, counts, or presence/absence. To save time and provide consistency for comparison in the following exercise, we will suggest characters for you to measure. If you were choosing characters for your own investigation, however, you would have to decide which characteristics are most appropriate. These are a few simple rules of thumb:
- Look for easily and consistently recognizable landmarks that can be identified on all individuals to use for reference points in your measurements. Mucrospirifer , the beast you will be measuring, presents several good landmarks. Examples include the "wingtips" (far ends of the hingeline ), the points where the edges of the central fold meet the shell edge, and the umbo or "beak" of the shell. See the figure below. The line along which the two halves of the shell come together is called the commissure ; you will be using this as a landmark in the exercise to follow.
the blackboard for each character measured. Both populations will be plotted on the same graph to make visual comparison easier. During section, we will together:
- Qualitatively evaluate the results on the board. Do the two populations "look" different? Does it matter whether you base your conclusion on one or the other character?
- Quantitatively evaluate our results through the use of several simple statistics described below. To give more time to discussion during section, we have already measured the same brachiopods and calculated these statistics for you; our results should not differ much from section results. You will, however, have to calculate these values from your own subset of the class data for your assignment, so again, make sure everyone in your group has a copy of the data. Be sure to have a conceptual idea of what the following statistics mean before you come to section.
A FEW SIMPLE STATISTICS
Most biological variation is distributed such that the measurements for a given character cluster around some average value and become less and less frequent as you move away from that value. To characterize this pattern, we use two different types of descriptors: (1) measures of central tendency , to describe the point around which the measurements cluster; and (2) measures of dispersion , to describe how the measurements are distributed around this point.
The arithmetic mean , or simply the "average", of a sample is the most common way of describing central tendency. The mean of a sample, represented by the symbol X, is calculated by adding together all of the measurements for a given character and dividing the sum by the total number of measurements made (for example, the total number of brachiopods measured). In mathematical shorthand, this calculation is shown as:
N
X
X
where the symbol Σ (sigma) simply means add up all the values of X (your measurements), and N is the total number of measurements made for that sample.
Another useful measure of central tendency is the mode , the value of X in your histogram that describes the greatest number of individuals. If the mode (or peak in your distribution) is different from the mean, as in Figure 2c below, then you know that the distribution is "lop- sided", with more individuals on one side of the mean than the other. This can give you valuable information about your population: are there more young (small) individuals han old, suggesting perhaps a high juvenile mortality? Are there more large than small, suggesting that perhaps the small individuals were washed away by water currents leaving only the big, heavy ones behind? A distribution containing two distinct peaks ("bimodal") is reflective of the age-group differences discussed on page 2.
Figure 2. Three frequency distributions with identical means and sample sizes but differing patterns of dispersion. Note the position of the mean vs. the mode on each.
Knowing the mean value is important, but to fully describe the variation in your sample you must also address the issue of dispersion, or how observations "pile up" or "fall away" from the mean. Measuring dispersion about the mean is important because two different samples may have the same mean value for some feature but still be very different in the way the measurements are distributed around it, as shown in Figure 2 on the preceding page. The standard deviation , abbreviated here as SD , describes the average deviation from the mean of the values measured in a population. The larger the SD, the more spread out a sample is around its mean. Compare the two distributions shown below, both for the same number of individuals. Sample A has a greater range of values about the mean than does sample B; sample A’s standard deviation is correspondingly greater than sample B’s.
SOME THINGS TO THINK ABOUT... TOPICS FOR GROUP DISCUSSION
- a) Compare the mean of sample 1 with the mean of sample 2 for rib number. Is there difference?
b) Compare the mean of sample 1 with the mean of sample 2 for umbo-to-commissure length. Is there a difference?
c) Compare the standard deviations of the two characters for sample 1, and repeat for sample 2. Is this a valid comparison? Why or why not?
- Why might one feature be more variable than the other, both within and between populations? Which is more likely to be genetically controlled, and which environmentally? Why might this be?
Technically, one should perform a statistical test called a "t-test" to determine whether the sample means are truly distinct from each other and therefore represent two separate populations. The formula to calculate the t-test is included here for those who are interested; your TF will help. However, to paraphrase Jack Sepkoski, a noted paleontologist and statistician, "if two things look different, they probably are". Beware of researchers that report "statistically significant differences" between populations even when the graphs they include show only fuzzy distinctions at best!
t =
- What benefit do you get from using umbo-to-commissure length for your measurements rather than length of the hingeline? Which do you think could be more variable and why?
- a) Would you say that these two samples represent one population or two distinct ones? Why?
b) Would you say that these two samples represent one species or two distinct ones? Why?
c) What are the problems associated with making decisions like these?
Assessment of population variation through time is done in exactly the same way as assessment of population variation in space. If mean values for a population shift through time, that indicates that evolution has occurred. Tracking the changes in fossil populations through time is a way of "watching" evolution occur.
- Suppose new information has become available stating that sample 2 was collected from a horizon above sample 1, and is thus estimated to be roughly 1 million years younger. Plot the sample means for rib number against time. Put time on the vertical axis increasing downward. Do the same for umbo to commissure. (We will do these plots on the board.)
Do these two points suggest any apparent trends in the brachiopod’s morphology during this time interval? ( i.e., do either of the characters seem to be changing consistently in one direction?) Is it valid to conclude that such trends exist from only two data points? Why or why not?
- To test your hypothesis, you return to the area and collect samples 3, 4, and 5, each from successively higher (and hence younger) beds in the Hamilton Group. The means for rib number and umbo to commissure are the following:
Sample Rib Number Umbo to Comm. 3 12 23 4 10 24. 5 11 29
Add to the plot on the blackboard the means for these additional samples. Separate them by 1-million-year intervals.
a) Describe what has happened through time for each character. Do you see any apparent trends now?
b) What is the nature of change through time? (constant, unidirectional, reversing, gradual, episodic, etc.) Can you really say anything about this issue given the data available?
- How would you distinguish true speciation events from morphological change within a single species? What makes this a difficult task?
MEAN = ________ MEAN = ________
SD = ________ SD = ________
DATA SHEETS Name_________________________
SAMPLE _______
N = _______ (sample size)
Umbo To Commissure Width Rib Number
X (mm) X^2 X X^2
TOTALS Σ X Σ X^2 Σ X Σ X^2
MEAN = ________ MEAN = ________
SD = ________ SD = ________
