Newton's color wheel never was physics until a metric that described the delineation and addition of colors was created. Red plus Yellow equals Green, means nothing without a measure of what Red, etc., are and what their amounts are. The tristimulus values were the metric that made color technology physics. My colleague and office mate, Arthur C. Hardy, enhanced the sophistication of spectrophotometry and with the Handbook of Colorimetry firmly established the color metric in the Thirties. In 1948 I was teaching color to sophomores from the Principles of Physics III: Optics, by Francis W. Sears, also a colleague and office mate.  The last problem set of the term included problems on discussing colors and surfaces defined by tristimulus values. However, technologists rapidly took over the application and development of color printing technology so that, by the mid Fifties, color was left by physicists to engineers. Modern computers contain much programming for color scanning, graphics modification and printing. For example, the RGB color representation that is convenient for color addition is used predominantly in the computer graphics input format and editing, while printers, using color subtraction use CYMK color representation. These two representations span non overlapping areas in the chromaticity diagram, so clever table lookup programs are used to make it possible to print the best CYMK copy from the RGB information.
The metric Newton supplied for use in describing the dynamics of physical bodies was an equation relating a ponderable force to a second time derivative, a parameter that is almost beyond human senses. And the algebra that it obeyed he provided with his method of fluxions.
Einstein's metric for gravity in flat space got nowhere. It was Marcel Grossman who introduced him to the Riemannian geometry of curved space that allowed him to link successfully geometry to gravity to provide the form of Newton's metric in curved space and its algebra.
Schrodinger's metric was a complex relation between the partial time derivative of the wave function and a product of a metric of the system energy and the wave function. And Max Born pointed out to Heisenberg that the funny mathematics he was doing with his metric was what mathematicians called matrix algebra.
As for Mendel, he carefully winnowed the possible plants and species to find those that reproduced true without anomalies. His metric was probability theory and the binomial distribution. Such care in selecting the subjects of study and the means to interpret the results and the need for abundant sampling were ideas beyond the imperative of botanists of his day. But they are needed as would be obvious to the trained physicist that Mendel was. That he stopped at seven traits, the number of chromosome pairs that peas have, is irrelevant. But he would have been criticized by botanists if he had worked with eight traits, a number more than the number of chromosome pairs, based on their incomplete understanding of processes that the chromosomes are subject to. It was not until 1947 that Joshua Lederberg demonstrated that gene recombination did occur.  This process decouples genes widely separated in the chromosome so that they are not linked in reproduction. Mendel was correct because he was careful; botanists were wrong because they did not have the understanding of the genetic substance they thought they had.
Creative experiments are not done to try to falsify a theory. They are done because someone has a revelation about how things work and that person then sets out to see if, indeed, that is how the world functions. The theory and the falsification procedure, and the acceptance of the brand new idea as being intuitively obvious by the now alerted world then follows the initial experiment. There are certainly errors in the original principles employed in the original revelation, but the theorists who follow the successful experimentalist are eager to correct those weaknesses. And the original experimentalist quite probably will fix up the theory herself as an afterthought.