Specimen 1872.10.25.1

Galton,

The cranium of the Great Auk, cataloged as Specimen 1872.10.25.1 in the University College collection, possesses a parietal width that defies the expected Gaussian curve of its kin. You will find the measurement tucked away in the appendix of my 1898 memoir on the reproductive selection of the species. It sits there in a table of sixty-four variables, reduced to a single row of type. Most men read the summary and conclude the species was simply unfit. They miss the asymmetry.

A discovery of this magnitude remains buried because the mind seeks the mean. We are trained to look at the center of the distribution, yet the revolution always begins at the extreme tail. The null hypothesis states that the variation in cranial width is a product of random environmental noise with no selective advantage. The chi-squared test for goodness of fit rejects this null at p=0.004. The skewness is positive and significant. This is not a failure of the data; it is a directional finding. The birds were not dying at random. They were being winnowed by a specific morphological constraint that we have ignored because it appeared as a minor deviation in a crowded ledger.

We fail to reject the null when our samples are small, and we call it “chance.” We reject the null when the sample is large, and we call it “law.” In this instance, the effect size is 0.72. The power of the test is sufficient to confirm that the deviation is not an artifact of the calipers. The truth of evolution is rarely found in the grand proclamations of the naturalists. It is found in the residuals. It is found in the footnotes where the anomalies are quarantined to keep the main text tidy. We have mistaken a lack of attention for a lack of evidence. The signal has been waiting in the fine print for thirty years while we argued over the shape of the beak.

The parietal width of Specimen 1872.10.25.1 is 48.3 millimeters.