The other day, I was asked a simple question that I knew the answer to, right off the top of my head, and since I'm nothing but lazy and lovin' the easy stuff, I thought I'd expand on it a bit here. The question was, "How do flounder get to be that way, with their eyes all on one side of the head?" And the answer is…pedantic and longwinded, but not too difficult.

The Pleuronectiformes, or flatfish, are a successful teleost order with about 500 known species, some of which are important commercially and are very tasty. The key to their success is their asymmetry: adults are camouflaged ambush predators who lurk on the sea bottom, taking advantage of their flat shape to rest cryptically and snap up small organisms that wander nearby. They lie on their sides, and have peculiarly lop-sided heads in which one eye has drifted to the other side, so both eyes are peering out from either the left or right side (which side is consistent and characteristic for a particular species, although there is at least one species with random assignment of handedness to individuals, and mutant strains are known in others that reverse the handedness.)

But they don't start out that way. The embryos and larvae of flatfish are symmetrical in external form, and the larvae feed by swimming about in the water column and catching planktonic prey. Here are some drawings of the early stages of the development of the summer flounder, Paralichthys dentatus, and you can see how ordinary they look (you can compare these to Chuck Kimmel's Danio staging series, for instance; flounder develop much more slowly than zebrafish, and the larvae are very distinct to my eye, but the general process is similar.)

Scale bar = 500µm. Prehatching Stages 0-11 of Paralichthys dentatus (drawn from photographs of live specimens). Stages 0-8 are animal pole views; Stages 7a, 9-11 are side views. (The perivitelline space is not visible in the animal pole view of Stage 0.) The chorion is shown in Stages 9, 10a, and 10 late. Stage numbers correspond to those in Table 2 except for Stage 7a, a side view of 128 cells in a double-layered dome; Stages 10a and 10 late depict the landmark features of Stage 10 seen in different orientations. GR, germ ring (medium gray); RA, raised embryonic axis (dark gray). Dashed circles represent lipid droplets. Prehatching Stages 12-20 of Paralichthys dentatus (drawn from photographs of live specimens). Stage number designations correspond to those in Table 2 except for 13a, which is a ventral view of Stage 13 illustrating eye rudiments (dark gray) and notochord (mediumgray). Stage 16 is shown in anterodorsal view. At Stage 20 xanthophores are circular and melanophores are dendritic. Dashed circles represent lipid droplets.

The weird stuff all happens later, after the larva has feasted and grown for a while. When the fish is about a month old, and over the course of the next several weeks, one eye begins migrating upwards towards the top of the head and over, until it is adjacent to the other eye.

Hatched larval stages of Paralichthys dentatus. Scale bars = 100µm. Camera lucida drawings from fixed specimens; pigment cell morphology and distribution are not representative of live larvae. Although fin rays begin to develop by late Stage C/23, they are not readily visible until Stage F/26. In Stages F/26-I/29 the position of the migrating right eye is shown in gray.

As you might guess, this process involves extensive remodeling of the skull. Bones soften and degenerate, epithelial and connective tissue thickens and pushes the eye socket around, and just in general many bones, including those of the jaw, end up oddly skewed.

Development of the jaw apparatus in hatched larvae of Paralichthys dentatus. Scale bars = 100µm. Camera lucida drawings from cleared and stained specimens (see Materials and Methods). Blue, anterior elements of first and second pharyngeal arches; red, posterior elements of first and second pharyngeal arches; B, branchial cartilages; Cb, ceratobranchials; H, hyoid; Hs, hyosymplectic; Ih, interhyal; M, Meckel's cartilage; Q, quadrate.

Here's a better view of both sides of the adult flatfish skull:

Camera lucida drawings of a cleared and stained skull of Pleuronichthys verticalis. (A) The blind side of the head; (B) the ocular side. Note that the ocular side has been reversed horizontally in order to facilitate comparisons with the blind side.

By the way, here are a few frames from a nifty movie of a flatfish using that queer skull to catch its breakfast:

It's just lying there flat on the ground, with its two eyes popping up and looking forward. When something yummy floats into the neighborhood, it snaps its jaws open and flares its operculum, and sucks it right into the mouth within a few milliseconds.

It's not just the skull that gets jiggered around; many other changes occur in the animal at this point. Pigmentation forms on one side and not the other, there are changes in the bony structure of the fin rays, some fins regress and others become larger, and there are internal changes to the gut. It sounds stunning and radical, but here's the thing: most fish, maybe all teleosts, go through a period called larval metamorphosis during which similar changes occur. For a while, I dabbled with studying the homologous period in the larval zebrafish. Unlike a flounder, the changes in a zebrafish are subtle, and unless you look closely, it just seems to be part of the continuum of growth. At about 3 weeks of age, they begin extensive calcification of the cartilaginous skeleton, adopt the adult pigmentation pattern, and what interested me most, undergo changes to the nervous system—in particular, I noticed an expansion of the dorsal root ganglia. Flatfish have coupled the normal suite of changes that occur in teleosts to genes that are differentially expressed on different sides of the body, and carry the changes to a more dramatic degree.

The other vertebrates we think of as undergoing dramatic morphological changes are the frogs, and actually, frog metamorphosis is almost certainly derived from teleost larval metamorphosis. They use the same trigger! The signal that initiates metamorphosis in frogs and fish is a surge of thyroid hormones, and thyroid hormones are also important in us mammals as a regulator of bone turnover. Metamorphosis can be blocked in flatfish by exposing them to an antagonist of thyroid hormone activity, thiourea, which has a number of dramatic effects.

Suppressing thyroid hormones with thiourea results in:

• inhibition of adult pigmentation
• inhibition of Pb formation (a unique bone in the flatfish skull)
• inhibition of eye migration
• inhibition of dorsal fin resorption
• inhibition of formation of radial bones in the fin rays

But the fish still continues to grow, so you end up with a larger version of the juvenile form.

So, significant parts of this process are simply modifications of normal changes that occur in vast numbers of vertebrates as they develop out of their larval stages into adulthood. The one unique thing in the flatfish is the linkage of the effects to asymmetry, and that, unfortunately, is still a bit of a mystery. We know some of the story: we know some genes central to the molecular identity of left and right, like Nodal and Pitx2, and we know some of the molecular biology of thyroid hormone action. What's missing right now is the molecular factor that ties those two processes together.

Gibb AC (1995) Kinematics of prey capture in a flatfish, Pleuronichthys verticalis. J Exp Biol 198:1173-1183.

Martinez GM, Bolker JA (2003) Embryonic and Larval Staging of Summer Flounder (Paralichthys dentatus) J Morph 255:162-176.

Okada N, Tanaka M, Tagawa M (2003) Bone development during metamorphosis of the Japanese flounder (Paralichthys olivaceus): differential responses to thyroid hormone. Proceedings of the 26th Annual Larval Fish Conference, Browman HI, Skiftesvik AB, eds. Published by the Institute of Marine Research, Postboks 1870 Nordnes, N-5817, Bergen, Norway.