Let me express it with an analogy to my own line of work (software development).
One of the perennial problems that we have in the software industry is totally unqualified candidates applying for software development roles. In order to cut down on the expense of bringing such people on-site for job interviews, candidates are routinely asked (usually at the start of the preliminary phone screen) to complete a ridiculously simple coding exercise. The canonical example is called “FizzBuzz,” and it asks them this:
Write a program that prints the numbers from 1 to 100. But for multiples of three print “Fizz” instead of the number and for the multiples of five print “Buzz”. For numbers which are multiples of both three and five print “FizzBuzz”.
Even the most junior developers should be able to complete such a task with their eyes closed. But many candidates – some of whom even have PhDs – can’t.
Only if they pass FizzBuzz are they then asked about more advanced topics such as object oriented design patterns, regular expressions, database concurrency, test-driven development, or machine learning. If they fail, the phone screen is cut short, they are thanked for their time, and their CV goes straight in the bin.
Your correct way is the geological equivalent of asking a candidate about object oriented design patterns, regular expressions, database concurrency, test-driven development, or machine learning. Snelling’s claim, on the other hand, is the geological equivalent of failing FizzBuzz.