Sir Arthur George Tansley, was an English botanist and a pioneer in the science of ecology. In 1917, he published a paper on the competition between Galium saxatile and Galium sylvestre in different soil types. The paper itself, consists of the relevant elements any typical paper should, ranging from a clear hypothesis, to a detailed experimental methodology section, revealing an exceptional grasp on the material, especially given that this was over 100 years ago. Nevertheless, Tansley’s research maintains a single blatant discrepancy. Let’s dive in.
The findings from Tansley’s work suggest that both species in question were capable of surviving, however, differences were revealed with regard to soil type. Although all of the data are indeed there, a statistical analysis seems lacking. Today, this form of testing can be regarded as a hallmark for research papers.
Upon further analysis, we deduce that the reason for which Tansley did not include a statical analysis in his paper is actually quite humorous. Statistical testing hadn’t been invented yet! Statistical analyses were in their infancy period. The t-statistic was only introduced in 1908 by William Sealy Gosset, making this methodology too recent for Tansley to include in a publication.
This valid excuse evokes deeper thought. One may ask, “How has scientific thinking evolved in the past 100 years?” And, “How much further do we have to go?” That, my friends, is a question to which nobody knows the answer. All I can say is, I hope they won’t go inventing many more tests. There’s only so much us students can handle!
Applied Plant Ecology Winter 2020 