The Colorado River in southwestern USA includes a couple of unusual cyprinid and catostomid fishes with extremely large humps on their heads.
It's long been thought that these evolved to help the fish swim in very fast-flowing water, but a new study has suggested that the humps probably came about for very different reasons, too.
The Humpback chub, Gila cypha, a cyprinid, grows to around 40cm/15" and has a prominent nuchal hump which all of the textbooks say helps the fish hug the bottom in the swift flow.
Similarly, the Razorback sucker, Xyrauchen texanus, a member of the catastomid family which reaches about 90cm/36", also has a big hump which is also believed to be an adaptation for life in the very fast moving Colorado.
Now, Donald Portz and Harold Tyus of the University of Colorado at Boulder, are suggesting that the main purpose of their humps might not necessarily be for a hydrodynamic advantage alone.
In a paper due to be published in the journal Environmental Biology of Fishes in November, the pair describe how they tested the hydrodynamics of the two fishes' body shapes using special casts of whole specimens with and without humps in an experimental flow tank.
They found that the presence of a hump did increase drag coefficients, but also came with a high energetic cost, because more energy is required to swim with a big hump stuck to the top of your head and lots of energy is required to produce the tissue that makes it up.
They argue that the humps may have formed through convergent evolution due to predation pressures by a predatory member of the carp family called the Colorado pikeminnow.
Despite its small-sounding name, the Colorado pikeminnow, Ptychocheilus lucius is no tiddler, and specimens can reach up to 180cm/72" in length.
It's the top predator in the Colorado and is the only fish big enough to see Humpback chub or Razorback suckers as a potential meal.
Portz and Tysus calculated from the gape size of the pikeminnow that it could consume about 55% of the Razorbacks and 71% of the smaller Humpback chub. However, if the hump was removed from the model, the level of predation could rise to 73% and 83% respectively, showing that the presence of the hump had an advantage in preventing the fish being eaten by pikeminnows quite as easily.
For more details see the paper: Portz, D. and H. Tysus (2004) - Fish humps in two Colorado River fishes: a morphological response to cyprinid predation. Environmental Biology of Fishes. 71. (3) 233-245.