I enjoyed Cornell. That's nothing lasting my memory, but I enjoyed mostly the math and the physics, not so much the anatomy and the memorization, which I was not good at. I took a year leave of absence at Cornell. I worked with Dr Walter Riker, who was a wonderful person. And he proposed a hypothesis, which was so off-the-wall. He said that acetylcholine's primary action was on the nerve terminal. He did not believe in the postsynaptic action of ACh.

I was given the job of recording from neuromuscular junctions with microelectrodes. I was the only one at Cornell who was doing that work, so I had to learn it from scratch, which I love to do. It did take me a full year to get the equipment up to date.

In fact, I came up to Columbia to work with a man named Ernie Amatniak, who had started a company building amplifiers, which would work with high-resistance electrodes. These electrodes had a very fine tip to penetrate muscle and nerve cells. They were filled with potassium chloride, had to be made fresh every time they were used because the KCl would eat away at the glass casing. I would pull these electrodes on an ad-hoc puller which had a filament that heated the glass filament. And then at a key time they would pull away and the glass, both the sides, would be pulled down to a fine tip. I'd say 50% of the electrodes were... workable. When they were filled with 3 molar KCl they measured 20 to 80 megaohms of resistance, so you couldn't use the usual amplifier with them. You had to get a high-input impedance amplifier. That's what Amatniak pioneered at Columbia. This wasn't long after the first microelectrodes were used to record from nerve and muscle cells. It may have been a decade later, but it was not universally employed yet.