He accidentally, blithely, produced his most cited paper by far.

From The Flamingo's Smile Reflection in Natural History by Stephen Jay Gould.  From Chapter 14, Losing the Edge.

There is a postscript to Losing the Edge chapter which begins:

Funny business. I labored for three years to write a monograph on the evolution of Bermudian land snails, and only nine people have ever cited the resulting tome. I wrote these few hundred words in a quarter hour’s flood of inspiration during an interminable round of speechmaking at my son’s annual Little League banquet (good for something besides sliced turkey, I always thought)—and it has already received more commentary than most of my technical papers combined.

He then goes on to address various items and arguments that have arisen in correspondence centering on the arguments in the essay.

But even in that first paragraph, there is a worthwhile observation to do with patterns and forecasting and variance.

Gould notes the stark contrast between a study he labored on for three years and the relative little attention it drew regardless of how important that study was to him and how much time he spent on it in comparison to what, for Gould, was essentially throw-away ephemera, the essay on variance in baseball.

I recall my father-in-law, Dr. Frank Kinard, making a similar observation one time.

He was a physicist who worked at DuPont's Savannah River Nuclear Plant after World War II.  He told the story, perhaps mildly exaggerated, at his bemusement and pique that across all his research career, his doctoral thesis on which he spent some years and hundreds or thousands of hours in the lab and also in the writing was only cited a handful of times.

In contrast, at some point early in his career, as part of some larger project, he encountered some particular technical challenge (I don't recall what it was) in the lab such as being able to quickly isolate different isotopes from a common ore.  For him, it was a tactical problem to be solved in pursuit of a larger problem.  

He spent a day or two on the on the problem, lighted on a solution which worked, dashed off a short paper and sent it to a journal while returning his focus onto the larger project.

As he described it "My PhD was important to me but it was only cited fewer than half a dozen times.  My technical note turned out to be the solution to a problem apparently many scientists had run into and were wrestling with.  I dashed it off without much consideration and yet it was my most cited research, many dozens of times."

Scientific research, songs, stories, art, theatrical plays, styles of food - it is extraordinarily difficult to accurately forecast what will catch fire and gain attention and what merely quietly smolder and then extinguish.  Systems which generate a lot of variation will, often unpredictably, create extreme values "without any special reason rooted in the intrinsic character or meaning of the extreme values themselves.”

Dr. Kinard was working in that first big blossoming of nuclear research in post World War II where there was extreme variance in and among different research labs.  He could not have easily known that the tactical problem he solved in a couple of days was a tactical problem that had been plaguing many other scientists.  But the reality was that extreme variance in the system drove extreme values in the outcomes for no predictable reasons known to him.  He accidentally, blithely, produced his most cited paper by far.