I've been a bit sex-obsessed lately. No, no, not that way—it's all innocent, and the objects of my obsessions are all fish.

A little background explanation: one of my current research projects is on the genetics of behavior. This is a difficult area, because behavior is incredibly complex with multiple levels of causation, and one has to be very careful when trying to tease apart all the tangled factors that contribute to it. It takes numbers and lots of controls to sort out the various contributors to a behavior.

What we've done so far, though, is to identify and quantify a few simple, robust behaviors that show consistent differences in different strains of zebrafish. Basically, we've got some easy measurements we can make of responses to specific stimuli, and we've got one genetically defined line of fish that do one thing, and another genetically defined line of fish that do something measurably different, and as closely as has been possible, the environment in which these fish have been raised is identical. So far, this has been straightforward, but now everything gets more complicated. The next step is to 1) replicate the observations in multiple populations and generations, 2) look at the behaviors in hybrids of our two strains, and 3) modify the environment of the developing fish to see how that influences their behavior. Another side project is to do a developmental study of the behaviors to see how early we can detect them—one drawback to working with zebrafish is that it takes 3-6 months to get adults from eggs, and if we could measure earlier we'd be able to progress a little faster.

The reason for the sex-obsession, then, is that these experiments need lots of fish, and worse still, many of them need to be obtained in large cohorts laid at the same time. Unfortunately, I have a small lab, and we're also limited because all of my new fish have to come from the populations that we expanded from 5 fish each last year to a few hundred animals this year. This strains the capacity of my small colony, and I really can't afford to have any slacker fish playing celibate. I've become a real slavedriver, cracking the whip and telling all my little fishy minions to get with the program and start having sex. It's a hard life in my lab, I tell you.

So what does a fellow have to do to pander to a fish? There's a couple of things I provide. Here they are in their human equivalents:

• Silk sheets: fish really like clean, tasty surroundings, so I've just finished some major tank scrubbing. They have new filters, and the water is the very best lab quality RO filtered stuff, with just the right amount of trace salts added back. You can't just throw them into sterile water, though; these fish are schooling animals that secrete interesting pheromones into the water, so I like to let them condition it for a while.
• Champaigne and chocolate-covered strawberries: The fish versions of these romantic foods are good, greasy, oily stuff and arthropods. Hungry fish don't breed, so I've been religious about stuffing them with fatty calories, and raising brine shrimp that they get fed every day.
• Sex toys: Zebrafish like a little variety in their environment, and get bored with the same ol' stuff. I've noticed that they seem to like colorful or distinctive rocks, and that plastic grass is a major turn-on.
• The Coolidge effect: Nothing stimulates a good round of 'getting to know you' sex than fresh faces in the tank. Shuffling partners around can trigger a healthy burst of activity, although the trade-off is that breaking up a school is also stressful and disruptive.

And also, I like to watch. Zebrafish are early morning breeders who go into a round of mating behavior at sunrise, so I've been going into the lab early in the morning to observe them. Zebrafish sex is spectacular. They are beautifully acrobatic, males and females chasing each other and swirling and swooping around, with many members of the school joining in. At its most strenuous, you get a ball of tightly packed fish wrestling at the bottom of the tank, with occasional eruptions of milky/pearly milt and eggs puffing out of the mass (followed by a mad scramble as everyone rushes in to have a little caviar breakfast). Someday I'm going to have to make a movie of the activity—so far, though, I've been stymied by their vigor and athleticism, since the last thing they'll do is pose quietly in one place while they're doing it.

Anyway, what prompts this longish revelation about my early morning perversions is a recent paper that describes the courtship behavior of my fave critters, the zebrafish, published in a newly launched journal titled Zebrafish. Darrow and Harris have characterized the courting behaviors and also done a developmental study of its appearance—just the kind of stuff I've been interested in lately. Here's the abstract:

Zebrafish, Danio rerio, is introduced here as a useful organism for investigating teleost courtship display and its development. Pair matings of adult zebrafish confirmed that the courtship behavior sequence fell into three general phases: initiatory, receptive/appetitive, and spawning. The developmental onset of identifiable courtship behaviors was also studied. Interestingly, the progression of behaviors in a typical bout of adult courtship was found to recapitulate the ontogeny of the courtship components as they were first expressed in juvenile fish. This finding suggests that the systems controlling the maturation and sequential expression of behaviors stimulated the expression of the later ones. This idea is consistent with the finding that courtship displays of immature fish were stimulated if they were housed with adults for a short period of time. This characterization of courtship display in zebrafish, including its development, opens the door to a forward genetic analysis of vertebrate reproductive behavior.

Some of the particularly useful parts of the paper were the identification and labeling of the behaviors:

The five male behaviors were: 1) Chase (following or swimming alongside female); 2) Tail-Nose (touching the females side or tail with nose or head); 3) Encircle (circling around or in front of female); 4) Zig-Zag (tail sweep and circle along females body, resembling a figure eight);and 5) Quiver (rapid tail oscillation against females side). The female behaviors were: 1) Approach (abrupt swimming movement like Present expressed independently of any male courtship behaviors); 2) Escort (swimming alongside male or remaining still while being courted); 3) Present (halting in front of male exposing side or swimming up and down in front of male); 4) Lead (returning at least three times to one location in the tank); 5) Egg-Lay (release of eggs with a twitch of the body, also known as oviposition). Egg-Lay can be expressed repeatedly throughout the courtship episode, with the female releasing 5-20 eggs at a time.

I've seen all those! The paper isolated the fish pairwise to make these observations, so I may be seeing some differences, though; this list sounds relatively static, but when I've seen the Quiver, for instance, the fish are also racing from one end of the tank to the other, and there may be an additional male doing an Encircle or Zig-Zag at the same time. These dry descriptions also don't do justice to the romance, or the lovely way a plump female can shimmy, or the frantic desperation of the males.

OK, I have been spending way too much time watching fish.

Darrow and Harris have an excellent summary diagram to illustrate what the interactions between the fish look like, although it didn't seem to reproduce very well (and the pdf version of the journal article is just awful). Take a look.

Ethogram and adult courtship and relationship to the development of courtship. (A) Ethogram of zebrafish courtship display. M and F next to the fish designate which fish is male or female. Arrows between the boxes designate a general progression based on the ranking tests for behavior sequence. (B) Two columns show representative individual courtship displays. Beneath the behavior abbreviations are the mean onset times in minutes. (C) Schematic diagram shows a model of the relationship between components of the courtship display and stages of development when parts of the display undergo maturation.

Zebrafish are common tropical freshwater fish, and lots of people have them in their home aquariums. If you do, you should try getting up sometime before dawn and watching them go at it. What you'll typically see is that, when you flip on the lights, they'll have been sleeping and will just be moving slowly around the tank. They'll begin to school up, and start bumping and nuzzling each other; often, a pair (a slender male and a plump female) will get playful, with frequent fast darts and chasing, and pretty soon the whole school is tumbling along in a wild rumpus, with an unmistakably greater level of activity than you usually see during the day. It's great fun. Download the Darrow and Harris paper and try to follow along. Get the kids up early for an educational experience!

Darrow KO, Harris WA (2004) Characterization and Development of Courtship in Zebrafish, Danio rerio. Zebrafish 1(1):40-45.