"My colleague Francesco Santini [from Pisa] visited natural history museums to take pictures of the species that interested us. It's something that's coming back in the study of biodiversity. Five, ten years ago, we saw museums as a place where organisms were put in alcohol and forgotten on the shelf. But that's no longer the case; many researchers are returning to these museums to take photos (or other images) of organisms, in order to get them out of their jars and study their morphology. Museums provide us with access to an enormous bank of species", Dr. Frédérich explains. 

A method of collection that in no way resembles Hollywood's adventurer image of the scientist, but one that has led to a database containing 384 specimens from 178 species of carangoid fishes, including 24 fossilised species. "We needed a group of fish that had a decent amount of well-preserved fossils, so that we could take precise morphological measurements."

Fossil of Ceratoichthys pinnatiformis (Blainville, 1818) dating from the Eocene (late Ypresian)fromMonte Bolca (Pesciara cave), left side view. ©Roberto Lazzarin (Collezione Museo Civico di Storia Naturale di Verona). 

No difference in the rate of speciation...

In total, 47 landmarks were chosen to describe the body shape of the specimens in lateral view ), thus capturing the body's elongation (i.e. the ratio between the fish's length and height) and the curvature of the cephalic region. Elongation is directly linked to swimming performance, such as manoeuvrability, acceleration and endurance, characteristics that provide an advantage in the pelagic environment. As for the curvature of the cephalic region, there are significant variations within this group, thus providing a good view of their morphological evolution.

Landamarks used to analyse variation in fish body shape, illustrated in Alepes djebada. (c) CC J.E. Randall: http://pbs.bishopmuseum.org/images/JER/.

Thanks to a combination of molecular evolution models and fossil calibration points, it was possible to produce a time-calibrated phylogeny. Genetics provided phyletic relationships and the fossils imposed time constraints. Evolution in diet (piscivorous or non-piscivorous) and type of habitat (reef or non-reef associated) was reconstituted on this tree. For extinct species, the habitat was deduced from the rocks where the fossils were found.