When to Change Sex

A handy guide from biologists

Hollywood does sensationalize, so the unnatural sex-role behavior in last summer’s cartoon hit Finding Nemo shouldn’t have surprised fish biologists. In the movie, a male clownfish loses his mate and most of their offspring in an attack on their home within an anemone, but—here’s the extraordinary part—that older male fish continues to act as a father, and the surviving youngster behaves as a son. To be fair, human children seeing the movie have not seemed greatly traumatized by the anomaly. One can only hope that saltwater aquarium hobbyists keep tanks with juvenile clownfish far from the DVD player.

GET SERIOUS. In real life, clownfish belong among the reef fishes that undergo life-history adventures unsuitable for children’s movies.
REGROUPING. The red grouper starts out female but like many other bony fish develops into a functional male. NOAA

While the movie was still in theaters, a study detailing more appropriate behavior for a clownfish appeared in the July 10, 2003 Nature. It reported that clownfish in Papua New Guinea do live as mom-and-dad pairs defending an anemone home. A couple of younger clownfish, often offspring, typically live there too, but they rank lower in the hierarchy, says Peter Buston of the University of California, Santa Barbara. However, his experiments with clownfish clusters in lab tanks showed that when that top female disappears, the surviving fish each put on a growth spurt and rise to the next rung in their hierarchy.

So, if Hollywood had settled for humdrum clownfish home life, in the New Guinea species, Nemo’s dad would have changed physiologically into the new egg-layer and mom in the anemone. And little Nemo himself, assuming he was the only remaining juvenile underfish there, would have moved up the hierarchy and become his new mom’s next mate.

However startling that version of Finding Nemo might be to Homo sapiens audiences, dozens of other species, including plenty of coral-reef fish, shrimp, and shellfish, wouldn’t blink a sensory receptor. Depending on the species, they morph from male to female or vice versa, sometimes transforming multiple times (SN: 10/21/95, p. 266: https://www.sciencenews.org/sn_edpik/ls_4.htm). The immediate trigger for the switches varies. It might be attainment of a certain size or life stage or a social factor, such as loss of a mate.

“We’ve had a good understanding of the basics for a while now,” says Stuart A. West of the University of Edinburgh. “More recently, people are starting to take it in new, exciting directions.”

One of those directions comes from the field called dimensionless biology, which strips away differences of size and life span and looks for patterns that hold true across many species, whether big or little, enduring or ephemeral. Recently, these biologists have reported a rule that can predict when an individual changes sex. This time, some fish biologists were surprised.


Sex change seems especially common in water habitats, and the list of species that manage some form of gender change stretches deep into various lineages. The strategy is quite common among coral reefs. Sponges, cnidarians, and brittle stars all include sex changers, too. The only vertebrates known to switch sex belong to the bony fish class.

The prevailing theme of theories and models so far, West says, is that sex change evolves when age makes a huge difference in one sex’s reproduction success but not the other’s. To illustrate that idea, he speculates what people might be like if they’d evolved into creatures that start out life as one gender and then naturally morph into another.

First, such humans would probably grow throughout their lives, instead of keeping the same bone size after reaching adulthood. “And then males would acquire something with age—size or wealth—that meant a few old males would monopolize all the mating,” says West. This change would plunge the rest of the men into the plight of many coral-reef fish: “Unless you’re the really big, huge guy, you might as well be—female,” says West. Nature would favor changes that let this imaginary H. sapiens mature sexually as a woman but eventually turn into a guy.

West could also have imagined a gender switch in the opposite direction. He could have made women, as many kinds of female shrimp do, reap a disproportionate baby bonus at the peak of age and size. Then natural selection would favor young men who eventually turn into older women.


Traditional approaches to analyzing sex change and other aspects of animals’ life histories have looked for costs and benefits of various strategies, but these trade-offs can be difficult to quantify. So, take a completely different approach, say Eric Charnov of the University of New Mexico in Albuquerque and some other theorists. These scientists examine ratios of a species’ life-history measurements, such as body size or lifespan. They search for proportions, which they call invariants, that remain constant within a group of species. Because this approach doesn’t focus on absolute values, practitioners call it dimensionless investigation.

These biologists say they have identified several ratios that apply to large groups of related species (SN: 10/16/99, p. 249). For example, take the average adult life span in a species and divide it by the age of first breeding. This calculation gives a ratio that stays relatively constant within a group of animals. Elephants, squirrels, and many other mammals live by a 1.4 ratio for female adult life span to age at maturity. That ratio differs from that of fish, 0.5, and of birds, 2.25. Also, reproductive effort—for example, measured by gonad proportions—times adult size equals about 0.6 for fish, 5 for birds, and 1.7 for mammals.

More than a decade ago, Charnov began applying the dimensionless approach to sex change. He first considered factors that might predict which sex a changeable organism would take first. He concluded that an organism would start out as the sex for which increased size yields the more-modest bonus for reproduction.

In the past few years, Charnov and various colleagues have revisited the topic, this time considering when during its life an organism changes sex. The new interest comes in part from noticing that some measurements traditionally associated with sex change factor into ratios that apply across many species. These include the ratio of life span to age at first breeding and the ratio of male age to reproductive success.

Another inspiration came from reconsidering the lives of Icelandic shrimp, all of which start out as males. Unnur Skúladóttir and Gunnar Pétursson of the Marine Research Institute in Reykjavik, Iceland, wanted to find a method to identify the different populations of the northern shrimp (Pandalus borealis) off the Icelandic coast. For 21 sampling areas, the researchers looked at the maximum length of the shrimp and at what is called L50, the size at which 50 percent of the males had switched to become females. The research, analyzing 900,000 shrimp, took 7 years.

As Skúladóttir recalls, one of the ways she plotted the data showed that the L50 was proportional to the maximum size. She and Pétursson published a version of these data in 1999, and, Skúladóttir recalls, “I sent a few reprints to biologists I had contact with.” One was Charnov. He “found this very interesting and saw more out of this than Pétursson and I did,” Skúladóttir says.

Skúladóttir and Charnov collaborated on a paper in 2000 predicting that indeed there are invariant rules for sex change and presenting shrimp data to bolster their claim. Body size at sex change divided by a population’s maximum body size should hold constant across shrimps in various locations, even when their individual growth rates differ, the paper predicted.

West and a colleague in his Edinburgh lab, David Allsop, decided to test this notion. As West remembers it, they started with modest hopes. Charnov “thought the relationship might hold within different populations of the same species or maybe across a couple of closely related species,” West says. “What we found though, quite shockingly, is that it actually holds ridiculously generally across all of the species.”

To do the test, the researches assembled life-history information in all the sex-changing species for which they could find data. In the end, the researchers used 77 animals, including fish, echinoderms, crustaceans, and mollusks. Regardless of the diversity, overall the animals typically changed sex when they reached 72 percent of the maximum size for their species, Allsop and West report in the Oct. 23, 2003 Nature.

To see how powerful a predictor this factor would turn out to be, the researchers performed a statistical test that determines how much of the variability in data depends on a given factor versus other influences. The ratio “explained 91 to 97 percent of the variance,” says West. He pauses in respectful silence and then clarifies, “As biologists, we tend to think that 10 percent is good.”

Allsop says that he didn’t expect these results, which he calls “quite staggering.”

Also, Allsop and West have published a similar ratio when they applied their strategy to 52 species of fish. “Fish change sex when they are 80 percent of their maximum body size and 2.5 times their age at maturity,” they reported in the September 2003 Journal of Evolutionary Biology.

The finding that the body-size ratio is such a strong predictor, “implies that all of the sex-changing animals must share certain underlying trade-offs between life-history variables that shape the decision as to when to change sex,” says Allsop.

Now what?

Reaction since the first report of the predictor has been mixed. “I have very strong feelings about the Nature paper in question,” says Matthew Grober, a sex-change physiologist at Georgia State University in Atlanta. The work shows that, overall, sex change tends to happen when an organism gets pretty big. “We knew that,” he says. The study “grossly oversimplifies or overlooks the fact that many species change sex based upon social cues and that these are quite variable, as is the size at sex change,” he adds. His own work examines fish hormones that increase in concentration in certain social situations and influence the physiology of sex change.

West protests that there isn’t that much variation when you adjust for scale. “The invariant that we found is a statistical property of the average species. The average species doesn’t actually exist. It is just a conceptualization of the real world to help us search for broad patterns.”

The broadness of the rule, though, is a major draw for some fans. These include Skúladóttir, whose shrimp studies as analyzed by Charnov inspired Allsop and West. Skúladóttir says: “I find it fantastic that they have looked at 77 species and proved the rule.”


If you have a comment on this article that you would like considered for publication in Science News, send it to editors@sciencenews.org. Please include your name and location.

Susan Milius is the life sciences writer, covering organismal biology and evolution, and has a special passion for plants, fungi and invertebrates. She studied biology and English literature.

From the Nature Index

Paid Content