The body proportions of many animals change as they get larger or older. This can be easily seen in humans, where babies have much larger heads (relative to the rest of the body) than adults. These changes in proportions are often accomplished by different growth rates between body parts, a process called allometry. In the case of humans, the head grows slower than the rest of the body and it said to be negatively allometric. When a structure grows faster than the rest of the animal, such as the antlers of a deer, is called positive allometry. When the two features grow at equal rates (and as a result the proportions do not change) this is said to be isometry.
Understanding the growth dynamics of living animals is an important part of their biology. The same is true for extinct animals, but in these cases we can’t observe the animals as they grow, and can only infer growth rates by looking at patterns across multiple individual fossils. To further complicate things, most extinct animals are not known from vast numbers of fossils, and have a limited sample upon which to base analyses of relative growth. This research paper investigates allometry in fossil animals in two ways:
- A review of other published studies was conducted to document the sample sizes used for studies of allometry, and then to compare these results between animals alive today, and those known only from fossils. The result, unsurprisingly, shows that analyses of fossils generally rely on fewer specimens (less information) than those based on living animals.
- A dataset of skull measurements for the American alligator (Alligator mississippiensis), was evaluated for allometry, and then reevaluated using fewer and fewer specimens to test the effect of reducing the sample size. The result was that at small sample sizes, the correct allometric trend was rarely identified. This is called Type II Error and often happens when sample sizes are small.
When the data from the two different analyses are combined we see that the sample sizes normally used for studies of allometry for fossils (from the literature review), are in the same range of sample sizes that experience high amounts of Type II Error (from the Alligator study). Studies of allometry based on fossils need to ensure the sample sizes are large enough to avoid large amounts of Type II Error, which will alter the results. This work helps help to highlight the continuing role of museum collections in scientific research – without a large number of specimens available to researchers the conclusions that can be made are limited.
Read the paper here: https://peerj.com/articles/818/
Caleb Marshall Brown and Matthew J. Vavrek, “Small sample sizes in the study of ontogenetic allometry; implications for palaeobiology,” PeerJ (March 10, 2015).