I don’t want to give you the impression that all my lecture subjects came from computer-geeks in Cambridge, Massachusetts. Far from it. My net was spread far wider than that, and my number-one crowd-pleaser in those early days had nothing to do with computers or any other manifestation of inorganic hard science. It was, in fact, a bunch of worms.


James V. McConnell

James V. McConnell

The planarian worms were the province of Dr. James V. McConnell of the psychology department of the University of Michigan. They weren’t much to look at — tiny, thin, the color of milk — but the important thing about planarians (about Dugesia dorotocephala, if you’d like a formal introduction) was that, although they were pretty dumb even as worms in general go, they were not totally ineducable.

Given an experimenter with a significant amount of patience they could be trained to acquire a Pavlovian conditioned reflex to certain kinds of stimuli, and Jim McConnell had that patience. Other factors that made planarians attractive research animals included the facts that they were easy to raise and not at all cuddly, so they didn’t have the legions of defenders of, say, gerbils or bunnies. They were also pretty cheap, all of which qualities added up to the fact that McConnell had large quantities of the creatures to experiment on.

And he did in fact experiment, in many ways, employing large numbers of planarians. One series of experiments — don’t ask me how he got on to this train of thought, because I don’t know — produced some quite curious results. It goes like this:



Suppose you take a batch of planarians and divide it into three parts. The first group, for reasons we will come to in a moment, we will name LUNCH, the second WELL-FED and the third simply CONTROL. The LUNCH group goes to school. That is, you do the Pavlov conditioning thing with them, repeating a given stimulus that naturally causes the worms to twitch (perhaps a flash of light) along with a second (maybe a mild electric shock) that doesn’t until the worms now twitch at the shock alone. This is the definition of a conditioned reflex, but when they have attained it you don’t give the LUNCH group caps to throw in the air. Instead their graduation ceremony amounts only to being chopped up into small pieces and fed to the second, or WELL-FED, group.

You might imagine that even a planarian worm could have objections to cannibalism. They don’t, though. WELL-FEDs will gobble LUNCH planarians down as enthusiastically as any other part of their diet. Then, while the WELL-FEDs are chowing down you start a new series of Pavlovian flinch lessons for them and the CONTROLs, and this is where Dr. McConnell found a quite unanticipated result.

Both WELL-FEDs and CONTROLs learned to do their assigned reflexes as they were supposed to, but there was a difference. The WELL-FEDs learned faster. They acquired their new reflexes after significantly fewer repetitions than the CONTROLs. To Dr. McConnell that could mean only one thing. Some learning was being passed along through the digestive system. Part of what the WELL-FEDs “knew” came not from what they had been taught but from what they had had to eat.


As you can imagine, I had a lot of fun with that idea, particularly with college audiences and especially when senior members of the faculty were right there in the room — and in particular when they were old friends like Tom Clareson or Jack Williamson. I got a lot of mileage out of those dim-witted little creatures. (By which, of course, I mean to refer to the planarian worms.) Sadly it has begun to seem that the effects McConnell (and I) described may have been illusory, because when other experimenters tried to replicate his results they failed.

That’s not to say that no one took McConnell’s work seriously, though. There was at least one person who evidently did, and that was a troubled recluse by the name of Theodore Kaczynski, better known to American history as the “Unabomber.” Apparently Kaczynski didn’t care for what McConnell was doing with the planarians, because he sent one of his bombs McConnell’s way. It didn’t kill anyone, but the noise of the explosion left McConnell partially deaf for the rest of his life,

But by then I had moved on to other subjects anyway, because a generous Providence had dropped in my lap a whole new comedic (and serious) topic to talk about.


  1. Stefan Jones says:

    Larry Niven used the “memory RNA” notion in one of his better stories. “The Fourth Profession” I believe.

  2. Marc says:

    Memory RNA and its possible uses is a beautiful concept. Sadly? it ‘seems’ Ockham’s razor (my favour law, which I use daily with great glee) triumphs again. Ironically Ockham’s “Theory of knowledge” touched remarkably close to memory RNA back in medieval times.

  3. Nix says:

    Maybe Niven used it in _The Fourth Profession_ (transferrable memory yes, not sure about memory RNA, and I’m fairly sure that can’t give you the ability to do miracles), but memory RNA was *definitely* used in _Rammer_, which became the first part of _A World Out Of Time_. Definitely one of Niven’s better works, gave me sensawunda like nothing he’d done since _Ringworld_.

  4. Bruce says:

    I hadn’t thought about planaria since my biology course in high school. At first I thought “The Planarians” was a parody science fiction fan club… Back on topic: it’s too bad learning by eating didn’t pan out. That whole concept struck me as a little “off” when I heard about it, but it would’ve been so convenient to rationalize some ancient human traditions about gaining the power of your prey by eating it.

  5. Anton Sherwood says:

    Were CONTROL worms fed other untaught worms? If not, maybe WELL-FED learned quicker because they ate better, not because they ate smarter, as it were.

  6. Gerry Quinn says:

    Transferring memories by way of cannibalism is a major theme in Gene Wolfe’s _Book of the New Sun_ also.