(This article was originally published in Cichlid News Magazine, Oct-97 pp. 30-31, It is reproduced here with the permission of author Ronald Coleman).
A fantastic picture of Perissodus microlepsis female with fry in Kipili, Lake Tanganyika. Photo by Ad Konings.
The cichlid mouth is an extraordinary construct; indeed the adaptability of the cichlid mouth may be one of the driving forces behind diversity in the family Cichlidae. The key to the mouth is the structure and musculature of the lower pharyngeal jaw the tooth-bearing bones located deep in a cichlid's throat. While many groups of fishes have teeth all over the inside of their mouths for holding prey and even for partially grinding it up, the flexibility of the cichlid pharyngeal jaw provides a complete processing center. This in turn frees up the regular jaws to take on other specializations, things like unusual feeding modes including substrate-sifting, algae-scraping, eye-biting, and scale-eating, in addition to specialized breeding modes like mouth-brooding. A couple of recent studies add to our understanding and appreciation of the cichlid mouth.
Cichlid aquarists familiar with Central American cichlids have long recognized that there is one type of cichlid that you simple cannot keep with plants, namely the eartheaters or substrate sifters. These cichlids are constantly digging. They get their food by taking in mouthfuls of substrate, sifting through it, swallowing the good, and spitting out the bad. On a recent trip to Costa Rica I watched an adult male Amphilophus rostratus quietly munch his way through a "seam" of buried leaves and debris that became exposed after a storm. The male chomped on a mouthful of substrate, processed it, then moved on to the next mouthful. An area where such a fish has been feeding is full of telltale pock-marks on the bottom, and this provides a useful clue for locating these sometimes elusive fish. You may also recognize this behavior if you have kept any of the Thorichthys genus, such as the firemouth cichlid, Thorichthys meeki. And that is where the puzzle begins.
Many people, myself included, have assumed that the Amphilophus substrate sifters such as Amphilophus rostratus, Amphilophus longimanus, and Amphilophus robertsoni must be closely related to the Thorichthys genus. Recent molecular work by Kevin Roe (University of Alabama), Don Conkel, and Charles Lydeard have shown this not to be the case. Their research uses the molecular composition of mitochondrial DNA to determine relationships among species. As with most molecular analyses, the results are not 100% conclusive, but several striking patterns emerge. Principally, the group traditionally called Thorichthys was found to be monophyletic, i.e., composed of closest relatives. This means that all the Thorichthys are derived from a single common ancestor they are a natural evolutionary group, all of which are substrate sifters to a greater or lesser degree.
The Amphilophus group turned out to be much more complex. Some members are substrate sifters as mentioned above, but some are not, such as the Midas cichlid, Amphilophus citrinellus. Results from Roe et al. (1997) reveal that not only are the substrate sifters only distantly related to the non-sifters, they in fact do not belong in Amphilophus at all.
Things get even more interesting. Despite the similarities in feeding style between the two groups of substrate sifters (the ex-Amphilophus sifters and the members of Thorichthys), these two groups are not closely related. For example, the Herichthys, which includes the Texas cichlid, Herichthys cyanoguttatus, and Paraheraps, e.g., Paratheraps bifasciatus, are more closely related to Thorichthys species than are members of the ex Amphilophus.
This can only mean that substrate sifting has evolved multiple times in Central American cichlids, in addition to evolving independently in South American geophagines (e.g., Satanoperca leucosticta) and in some African lineages.
Nomenclatural warning: if this analysis holds up to scrutiny, there could be a new generic name for the substrate sifters formerly in Amphilophus. Roe et al. (1997) recommend resurrecting the genus Astatheros, so, for example, Amphilophus rostratus would become Astatheros rostratus.
Muchio Hori (Wakayama Medical College, Japan) is also interested in the mouths of cichlids, but for a very different reason. Hori (1993) studied the Lake Tanganyika scale-eating cichlid, Perissodus microlepis, in the wild. In the lake there are at least seven species of cichlids which survive by feeding off the scales of living fishes, a mode of feeding called lepidophagy. To do so, the scale-eater has to approach the target by stealth and then dash in to grab a mouthful of scales. Perissodus microlepis does this from behind to avoid being seen until it is too late. As a consequence, the attacker is forced to hit the target at an oblique angle, either from the "back-left" or the "back-right."
With most any other family of fish the story would end there, but not so with the cichlids. Hori and his colleagues conducted surveys and experiments to show that the scale-eaters have evolved a wonderful adaptation for attacking from the rear: they are either left or right-handed!
First, Hori noted that the mouth of this fish is not symmetrical when viewed from the front; instead, some individuals have mouths bent leftward while those of others bend to the right. Next, Hori tested whether each fish's behavior matched its mouth. By using prey fish as lures, he watched the Perissodus attack. Sure enough those with a "left-handed" mouth always attacked the right flank of the prey, whereas those with a "right-handed" mouth targeted the left side. In both cases, the twisted mouth allows the Perissodus a better angle of attack.
Just to be sure, Hori looked in the stomachs of right-handed attackers and, as expected, found scales from only the left side of prey fish. Apparently it is possible to tell the side of origin of a scale from its structure. Furthermore, the "handedness" of the Perissodus is passed from parent to offspring and appears at an early age, even while the fry are still under parental care.
But there's more, as the saying goes. Despite their impressive mouths, Perissodus are not terribly successful hunters. Only about one in five attacks yield scales. Why is this? Because the prey fish are watching. By looking at the scarring pattern on prey fish, Hori showed that they are watching disproportionately to the side most vulnerable to the local Perissodus. For example, if the local population is skewed towards left-hand attacks, the prey fish watch their right side disproportionately. Consequently, the rarer right-hand attackers in the population gain an advantage and are more successful at feeding. Feeding success ultimately translates into greater breeding success, and so the composition of the local population slowly shifts toward an increase in the frequency of right-handed Perissodus. Eventually the right-handers become dominant and the prey fish switch to guarding their left flanks. Then, the left-handed Perissodus get the upper hand. And so the game continues, generation after generation, with prey fish constantly looking over their shoulders for those amazing cichlid mouths.
- Hori, M. 1993. Frequency-dependent natural selection in the handedness of scale-eating cichlid fish. Science 260:216-219.
- Roe, K. J., Conkel, D., and C. Lydeard. 1997. Molecular systematics of Middle American cichlid fishes and the evolution of trophictypes in "Cichlasoma (Amphilophus)" and "C. (Thorichthys)". Molecular Phylogenetics and Evolution 7:366-376.