This story is adapted from Phallacy: Life Lessons from the Animal Penis, by Emily Willingham.

For humans and other vertebrates, the literal response to the question of what makes a penis is that it consists, to varying degrees, of connective tissue, spongy swellable tissue, muscles, and a blood supply. And it’s pretty straightforward to see many vertebrate organs that look like penises, call them that (“and that’s a penis, and that’s a penis!”), and be right. But not always.

If you were on Mars and found an animal with something you suspected was a penis, what features would you use to rule your hypothesis in—or out? How about “something that inserts into a partner’s genitalia during copulation and transmits gametes”? Seems reasonable, although it’s even possible to argue about what copulation is. Some other time. Let’s look at some of the things that animals insert into their partner’s genitalia during copulation and see if they fit this vision of a “penis.”

Phallic Feet

Millipedes are probably best known for all those legs, although it’s not the thousand that the name implies. The millipede record holder for most legs has only 750, and most have far fewer. You can distinguish a member of the eighty thousand or so species of millipedes from its less speciose centipede doppelgängers by how many legs they have per segment, if you want to get that close. Millipedes have two pairs of legs per segment, while centipedes have one.

The millipede leg pairs of interest for our purposes are the eighth ones, which these animals use as intromitta, or gonopods (which basically means “feet that copulate”). These arthropods are not alone among their arthropodish kind in coopting an appendage this way. The genetics of limb building also might contribute to phallus formation in vertebrates. These realities give fresh relevance to tired, aspirational penis-related jokes about “third legs,” although as noted, it’s the eighth pair of legs in millipedes.

The eighth pair of legs isn’t the entire copulatory story for these multipoded animals, at least for the well-studied members of the genus Parafontaria. These species also pull in the services of the second pair of legs, which is where their genital openings are. Rather than getting involved in intromission, though, as you might think genitalia would, this pair of legs near the genitalia simply source sperm for the eighth pair.

An amorous millipede begins courting by first trying to insert his uncharged eighth-leg-pair intromitta into his partner of choice. If she does not reject this trial poke and he’s successful, his second pair of legs will charge up the old eighth pair with sperm. Now fully loaded, he’ll intromit again, this time with the goods. The pair will become immobile and stay coupled for 29 to 215 minutes.

Why would a millipede have any need to do a dry run with its legs-slash-intromitta when life is short, especially for millipedes, and dangers probably lurk all around while he stalls with a test poke? If you’ll recall, millipede species are copious in number. Millipedes are only millipedes, and they can make mistakes about which species is which. In this genus, at least, the test thrust is one way for the suitor to confirm that his target partner is, in fact, the right fit for his intromittent legs. This quick test saves him from wasting a bolus of precious (seriously) sperm on a millipede inamorata from the wrong species. Given how long that second bout can last, it’s probably also a way to make sure the male is investing all of that precious time on the right mating partner.

You probably never pictured yourself viewing millipedes as a pattern of normal sex behavior, but now’s the time. Although employing four pairs of legs to inseminate a female seems outside the bounds of human experience (and it is), the millipede at least inserts the intromittum into the partner’s genitalia. The first time, there is no sperm, but the second time, if there is a second time, the sperm cometh. For millipedes, we’ll have to split the difference on whether they meet our test of “something that inserts into a partner’s genitalia during copulation and transmits gametes.”

The Well-Armatured Intromittum

We can’t say that much for the many species of insect that skip that bit completely and just inject sperm any old place on a mate’s body. Some flatworms also have no choice but to do this because of a lack of a “receptive female aperture,” as one embryologist put it. There are no genitalia to serve as the insertee, so the animal is forced to use a “stylet” at the tip of its “protrusible sperm duct” to pierce the partner’s body just about anywhere. Following the piercing, the injected sperm undergo a great migration through the partner to the eggs.

Although the lack of an obvious opening for sperm deposition explains this sloppy target practice on the part of flatworms, it doesn’t explain all cases of hypodermic sperm delivery. Some species of spiders and insects, which are not especially closely related groups, have evolved this adaptation and use quite similar forms to achieve it. Whatever the pressures are that shape the hypodermic intromitta, they seem to converge again and again on the same structure: something stabby but with a hollow tube to deliver sperm. But it doesn’t deliver the sperm into the genitalia, definitely not meeting the criterion of “something that inserts into a partner’s genitalia during copulation and transmits gametes.” For now, let’s just call these “intromitta” and not get any more specific.