It's amazing the coincidences one comes across while reporting about autism:

The autism rate rises in tandem with increasing numbers of vaccines that contain a known neurotoxin, ethyl mercury.

Public health authorities say that's coincidence.

Parents say their children became autistic after receiving mercury-containing vaccinations, sometimes several shots in one day.

Pediatricians call that coincidence, too.

Another remarkable fact that caught my attention: Autism was first identified in both the United States and Europe at almost exactly the same time. Child psychiatrist Leo Kanner published his landmark paper at Johns Hopkins University in Baltimore in 1943; pediatrician Hans Asperger published his -- about a slightly less severely affected group of children -- in Vienna in 1944. Cut off by a world war, neither knew of the other's work.

Coincidence, say the experts, who attribute the timing to improving diagnostic techniques in both countries.

What else can the experts say, literally invested as they are in massively funded genetic research to find the presumed cause? If it's not a coincidence that autism arose simultaneously on separate continents, that suggests something happened in two places at once to trigger the disorder. And that would suggest genes are not the fundamental factor, though they certainly could be implicated in making some children susceptible to whatever the new exposure was.

The first Age of Autism column in early 2005, titled "Donald T. and Fritz V.," made this point, noting that the first Austrian case report and the first American case report "were born within four months of each other, Fritz V. in June of 1933 and Donald T. that September."

Because Kanner's kids became known as "autistic" and Asperger's as having "Asperger's disorder," the overwhelming commonalities have not been fully appreciated; Kanner's study of Donald and 10 other children was titled "Autistic Disturbances of Affective Contact," and Asperger called his study of Fritz and three other children "'Autistic Psychopathy' in Childhood."

At different times in different places, they were seeing the same remarkable disorder. These kids were all "on the spectrum," as we say today, and that raises a question I put this way in that first column: "Was it coincidence the first few cases of these strikingly similar disorders were identified at the same time, by the same term, in children born the same decade, by doctors thousands of miles apart?

"Or, is it a clue to when and where autism started -- and why?

"The question reflects a huge, and hugely important, debate. If autistic children always existed in the same percentages but just were not formally classified until the 1940s, that would suggest better diagnosis, not a troubling increase in the number of autistic children.

"If, however, autism had a clear beginning in the fairly recent past ... then the issue is very different. That would suggest something new caused those first autism and Asperger's cases ... something caused them to increase, and something is still causing them today."

At that time, I had no clue about a possible connection. But now, after reporting and writing more than 100 Age of Autism columns over the past two years, I do.

The clue could be the simultaneous arrival of ethyl mercury -- but not, necessarily, in the vaccines that some parents blame for the huge rise in reported cases over the past two decades. What I've learned is that this especially dangerous form of organic mercury also was used starting around 1930 in fungicides. Morris Kharasch, the same American chemist who patented its use in vaccines -- where it is called thimerosal -- also pioneered its use as a seed disinfectant.

Remember, this type of mercury didn't exist in nature; it's man-made, and Kharasch is the man who made it marketable.

Two companies, one German and one American, built their ethyl mercury fungicide, called Ceresan, on those patents. In a joint venture, they sold it in both Europe and the United States. (Mercury-containing agricultural products were phased out decades ago after their effects on humans and the environment were recognized -- though ethyl mercury still remains in most flu shots given to pregnant women and young children. Go figure.)

So what might have happened -- warning, hypothesis ahead -- is that some early exposures to ethyl mercury came from inhaling or otherwise coming into contact with it via that agricultural route. And some of the children exposed to this novel and neurotoxic form of mercury developed a novel neurological disorder called autism.

Speculative, yes. But everything about the cause of autism at the moment is speculative. And as I showed in a column earlier this year titled "Mercury Link to Case 2," the first three cases diagnosed in the United States can plausibly be linked to such exposures.

Case 2, in particular, is compelling, because documents show that the father of that child was a plant pathologist experimenting with ethyl mercury fungicides for the U.S. government at the time his child was born in 1936. The father of Case 3 was a forestry professor -- not a very different occupation from plant pathologist -- in the South, and Case 1 lived in a town called Forest, Miss., near sites where ethyl mercury was first tested as a lumber preservative.

Plants, forests, timber, the South.

Now check this out: Among the earliest cases seen in Europe were 10 identified by a Dutch researcher named D. Arn Van Krevelen. One of the 10 fathers was a horticulturalist; another was a florist's salesman.

Other early studies in the United States found a clear "chemical connection" via the occupations of a similar percentage of parents, a connection overlooked as the gene-hunting juggernaut gained steam.

Maybe that's no coincidence.