Stephen Spotte is a marine scientist and writer. He has published 19 books, including four volumes of fiction, a memoir, and a work of cultural theory. He is a Certified Wildlife Biologist of The Wildlife Society and also holds a U.S. Merchant Marine officer’s license. His popular articles about the sea have appeared in National Wildlife, On the Sound, Animal Kingdom, Explorers Journal, and Science Digest. As a life-long researcher, Stephen holds a soft spot for the possibilities of science’s astonishing unrealities to be mined and their contents allowed to metamorphose into strange shapes and patterns in his fiction writing.
an organism that lives in or on another organism (its host) and benefits by deriving nutrients at the host’s expense.
In January 1965, soon after obtaining an undergraduate degree in biology, I went to work at the public aquarium in Niagara Falls, New York, then under construction. At a staff meeting one night Bill Kelley, our boss, passed around a bottle of bourbon while we lounged in the motel room that served as a field office and said to listen because he had a story to tell. After we’d topped off our glasses he told about the candiru, a small, slender catfish inhabiting the Amazon River and certain of its larger tributaries. Bill had spent considerable time in the Amazon basin studying the fishes, and he was especially interested in this one.
The candiru is nothing extraordinary except for its life-style, which requires it to feed exclusively on the blood of bigger fishes. Hence its nickname “vampire fish.” To obtain a meal a candiru swims into the gill cavity of its victim, locks its jaws onto a gill filament, punctures the fragile tissue with needlelike teeth, and begins sucking blood. When satiated it returns to the river bottom to digest its meal, doing nothing in particular until hungry again.
It was a fascinating story, especially when Bill mentioned that since Europeans began exploring the Amazon in the 18th century tales have emerged of candirus occasionally swimming into the urethras of unsuspecting bathers. The invader then erects its spiny gill covers, becoming lodged in place and causing excruciating pain. According to legend, when a man is attacked a local shaman must be summoned to amputate the penis, thus saving the victim from a fatal infection.
That night Bill also told us how Dr. Eugene W. Gudger, an ichthyologist (fish specialist) at the American Museum of Natural History in New York City, had developed an armchair interest in these legends and combed through the world’s arcane literature, perusing dusty books and long-forgotten journals and noting every mention of the subject he found. After exhausting the last leads he wrote a short book, published in 1930. Only a few copies were printed, and today the volume is exceedingly rare. Gudger was never able to find conclusive proof of candiru attacks, but if they did happen then the candiru would, according to him, be the only known vertebrate parasite of humankind. Ironclad evidence would be an astounding discovery. So Grudger posited, assuming, of course, that candirus are actually parasitic on humans.
One online dictionary defines a parasite as “an organism that lives in or on another organism (its host) and benefits by deriving nutrients at the host’s expense.” If candirus attack humans, would Gudger be right in his assessment? Not in my opinion. Mosquitoes, fleas, ticks, black flies, tapeworms, and other such organisms indeed derive nutritive benefits from their hosts by sucking blood, including that of humans. But what benefit could the candiru, a specialized bloodsucker, possibly gain from invading a human urethra? None I can think of. And what would be the evolutionary advantage of entering an orifice from which there’s no escape? Again, nothing comes to mind.
With its gill spines buried in urethral tissue a candiru would be stuck inside a very narrow space unable to wriggle either forward or backward. It would ultimately die there. The gills of a big fish, in contrast, are suspended in water allowing a candiru to easily detach and drop away, and the gill covers of its victim are like open doors. Escape is easy. Candirus obviously evolved to suck the blood of other fishes and not seek nourishment in the relatively bloodless human urethra. Were this not so such attacks would be routine, and we would have more than adequate documentation. Furthermore evolutionary theory posits that an individual of any species must survive to reproduce and leave progeny that themselves are capable of reproducing. A candiru that dies inside a human urethra fails such a test. Deprived of water to stay moist and unable to obtain oxygen it would quickly suffocate. Still another strike against the candiru-as-human parasite theory.
The candiru dilemma continued to puzzle me, as it did many other scientists. As the years passed I followed in Gudger’s footsteps, collecting copies of every report he found and others he missed or that appeared after 1930. And in the process I developed three essential questions. First, has there ever been an actual candiru attack on a human? Existing reports, written over three centuries and in several languages, were anecdotal. In no case was the offending fish ever recovered, much less identified. And if candirus indeed have an affinity for mammalian urethras, why should humans be special?
Later observations and experiments I made with South American colleagues showed that candirus attack larger fishes of many species more or less nonselectively, so why not also attack the aquatic and semi-aquatic mammals with which they live in daily contact? The Amazon has many such potential species: manatees, river otters, tapirs, and two species of dolphins, to name a few, and the anatomies of all are thoroughly described. A candiru has never been reported from the urogenital system of any mammalian species indigenous to the Amazon River basin.
Second, what exactly constitutes a candiru? The ichthyology literature indicated dozens of putative species loosely labeled “candirus.” Which was the culprit? No one knew, of course, because no specimen obtained after a candiru attack had ever been examined by a scientist.
Third, assuming candirus actually did sometimes attack humans, what attracted them? Mosquitoes, for example, are drawn to the ammonia we humans exhale. Some of the reports Gudger and I found speculated that urine was an attractant and possibly aqueous ammonia too. (Most fishes release it into the water from their gills as a metabolic byproduct.) The candiru’s normal victims (large fishes) slough mucus routinely as they swim, leaving a trail of dissolved organic compounds (a sort of chemical “footprint”) in their wakes. A catfish has olfactory and taste receptors located inside its mouth and on its exterior surfaces. It tastes and smells using its whole body and could presumably track a potential victim by following along downstream. Candirus, as mentioned, are catfishes. I prepared samples of candiru skin at the Marine Biological Laboratory in Woods Hole, Massachusetts, and examined them using a scanning electron microscope. The external surfaces are riddled with sensory pores.
By the late 1990s my calendar had cleared sufficiently that I could investigate candirus part-time as a kind of hobby. I began by translating into English all reports in my files from German, French, Spanish, Portuguese, and Italian. With the addition of those published in English (including Gudger’s book) I was confident I’d missed nothing important. Thus armed with the known history of candiru biology and lore I went to the state of Amazonas, Brazil, in search of a documented case of a candiru attack on a human.
Through word of mouth my Brazilian colleagues from the Instituto Nacional de Pesquisa da Amazonia (INPA) and I tracked down a urologist whose office was in Manaus, a large city in the central Amazon where the Rio Negro and Rio Solimões merge to become the Rio Amazonas. Several years previously he had removed a candiru from the urethra of a young man attacked while partly submerged and urinating in the main channel of the Amazon near Itacoatiara,175 kilometers east (i.e. downriver) from Manaus. The physician, whose name was Dr. Anoar Samad, had extracted the fish, which by then was dead, using an endoscope equipped with forceps and a camera. We went to his house where he showed us the video on a home computer. Remarkably, Samad had kept the fish, which he turned over to us to be deposited in the collections of INPA. The next day we photographed and identified it.
Soon after we packed equipment and supplies and drove north to the state of Roraima, where we set up camp and a makeshift laboratory on the banks of the Rio Jauaperi. Over the next few nights we caught several candirus using push nets while wading in the rapids, and also by using as bait large fishes obtained by hook and line and tethered in the river to attract candirus. We then conducted a series of aquarium experiments testing the effect of various substances (e.g. ammonia, human urine) to see if any were attractants to captive candirus. According to our results, they weren’t.
I later wrote a book titled Candiru: Life and Legend of the Bloodsucking Catfishes, published in 2002, and my Brazilian colleagues and I collaborated on a scientific journal article describing those field experiments beside the Rio Jauaperi and in a makeshift laboratory in colleagues downtown apartment in Manaus.
The final question begging an answer is this: are candirus parasites? Not according to the classic definition of the term. Do they occasionally swim into human urethras? They certainly do. My colleagues and I, having found the smoking gun, can say so with assurance. Then why would they do something so obviously counter-productive if the act serves no evolutionary function? I don’t know. A candiru takes a wrong turn into a dark, dead-end orifice on the false notion it’s found the gill cavity of a big fish and subsequently a legend is born? Maybe, but conjecture is all there is at the moment. We can say this much: the candiru isn’t the only vertebrate parasite of humankind; actually, it isn’t a parasite of humankind at all.
 Spotte, S., P. Petry, and J. A. S. Zuanon. 2001. Experiments on the feeding behavior of the hematophagous candiru, Vandellia cf. plazaii. Environmental Biology of Fishes 60: 459-464.