Ctenosuchus, a Jurassic crocodile relative from Morocco, had teeth too thin for large prey

Imran's case for a soft-prey diet is an exclusion argument. He said, "The teeth are so fine that we don't think it was catching normal prey. If it caught a big fish, and the fish wriggled, those delicate teeth would get smashed. We think the animal was probably eating soft prey like shrimp, worms and tiny fish, so nothing that would break its teeth." ^[9] The account lists very small fish, crustaceans such as fairy shrimp and other small aquatic creatures as the likely food ^[8].

The thing this doesn't tell you is what the animal actually swallowed. The reading rests on tooth form. Imran speaks of "one of the fossils" ^[5], so the species rests on more than one specimen, but the phys.org account does not say how many, or whether any trace of a meal survives.

The comparisons add a second inference. The teeth resemble those of mesosaurs and of certain long-snouted pterosaurs that lived millions of years later ^[10]. The jaws and teeth also share many similarities with those of the Jurassic pterosaurs Gnathosaurus and Ctenochasma ^[11]. The account frames this as different groups evolving similar feeding adaptations independently ^[12]. It stops short of calling Ctenosuchus a filter feeder, and Imran's own list is soft prey. The pterosaurs' filter feeding is itself something, in the account's words, that "scientists believe" ^[11].

We think the exclusion of large, struggling prey is the sturdy part of the reading, because thin teeth cannot take that load. Which soft prey it took, fairy shrimp or worms or tiny fish, is a finer call than tooth shape is likely to settle. The wider feeding-roles claim, which the account makes for Jurassic crocodile relatives as a group ^[13], depends on the first of those holding.