From the record
Real cases, written from the published record. Predict first; the record and its sources come after.
1961 · Bethesda
Poly-U
Marshall Nirenberg and Heinrich Matthaei had a test tube of juice from broken bacteria: ribosomes, tRNAs and all twenty amino acids. Add a message, and the tube makes protein.
They added a synthetic RNA made of one letter only: U U U U U U …
What protein did the tube make?
What happened. In May 1961 the tube made a chain of pure phenylalanine. So UUU is a word for phenylalanine: the first word of the genetic code anyone read. Over the next few years, repeating patterns such as UCUCUC… (Har Gobind Khorana's lab) pinned down the rest.
Pause to think. Why did they need repeating patterns, not just single letters?
1943 · Bloomington and Nashville
The jackpot tubes
Some bacteria survive a virus that kills the rest. Did the virus cause the change, or were a few cells already different before it arrived?
Salvador Luria and Max Delbrück grew many small cultures separately, then spread each one on a plate with the virus and counted the survivors.
What did the counts across the cultures look like?
What happened. The counts jumped wildly: most cultures had a handful or none, and a few were jackpots with hundreds. If the virus caused resistance, every culture would have had about the same number. Jackpots mean a mutation had happened early in that culture, before any virus was there, and its cell's descendants filled the tube.
The virus does not create resistance. It reveals it.
1866 · Brno and 1936 · London
Too good to be true?
Gregor Mendel reported 5,474 round peas to 1,850 wrinkled: 2.96 to 1, close to the 3 to 1 his idea predicts. His other experiments landed close too.
In 1936 the statistician Ronald Fisher added all of Mendel's results together and asked how often real counts would land this close to the prediction, this many times in a row.
Fisher found the fit was closer than chance usually gives. What explains it best?
What happened. Nobody knows. Fisher thought someone had known what to expect. Later historians and statisticians have argued every side, including that Mendel's methods were misread. Nobody doubts the 3 : 1 rule itself: it has been confirmed many times since.
What we don’t know. The debate is still open; a 2008 book collected the arguments without settling them.