Cannibalism in seahorses is very uncommon. It does occasionally occur among seahorses with pelagic fry, but such episodes are normally rare exceptions. Hippocampus erectus is not typically known for this, but it does take place in very isolated incidents. I have never personally witnessed a case of cannibalism in H. erectus. Hippocampus erectus fry undergo an abbreviated pelagic phase and ever once in a while there are reports of cannibalism, but 99.99 % of the time erectus make model parents.
However, some species are more prone to this aberrant behavior than others. For example, unlike most seahorses, captive-bred Hippocampus abdominalis are confirmed cannibals under certain circumstances. This unusual behavior does not appear to be a consequence of dietary deficiencies or overcrowding. Rather, it seems to be due to the simple fact that cultured Pots are accustomed to being fed; rather than hunting for live prey along the bottom, they expect their prey to be introduced from the surface and come down to them like manna from heaven. They are thus conditioned to take anything above them in the water column within the right size range that drifts past, and unfortunately this includes surface-hugging pelagic fry.
As a case in point, David Warland experienced cannibalism of fry by young Potbelly adults 18-24 months old maintained at a stocking density of 30 seahorses per 800 liters (211 gallons). Since that’s one seahorse for every 26.6 liters (7 gallons), he does not believe over stocking was a factor in the cannibalism, nor did hunger play a role since the H. abdominalis were fed to capacity 4-5 times daily during daylight hours (sunup to sundown).
Surprisingly, when cannibalism does occur, it is often the female that’s the culprit. That is the case with H. comes, one of the few seahorses species that is known to cannibalize its young on occasion (Neil Garrick-Maidment, pers. comm.). When this happens with Tigertails, it is typically the accompanying female who begins predating the fry as her mate gives birth (Lesley Holland, pers. comm.). The male may then follow suit as well and the entire brood can quickly be destroyed (Lesley Holland, pers. comm.). Once a pair of Tigertails turns cannibal, this unnatural behavior is likely to become habitual, putting their future offspring at risk.
In all my years, I have only once heard reports of any such problems from seahorses with benthic fry, and in that case it was the least likely culprits you could imagine which were at fault — female dwarf seahorses (H. zosterae). Now, cannibalism is of course unknown in H. zosterae, since the adult dwarfs are much too small to swallow even newborn fry, but apparently infanticide is not. This is what Joanne Heuter, a very accomplished breeder of dwarf seahorses, reports in that regard:
"The adult females will sometimes kill the newborns by snapping the back of their necks with their heads. I don’t know why, maybe competition for food, your guess is as good as mine. This is the only act of aggression I have ever noted in the dwarf seahorses…. I keep the babies away from the main tank for at least three weeks after birth."
Cannibalism is really quite rare in seahorses, so the hobbyist need not be overly concerned by this possibility, although there are a few species such as H. abdominalis and H. comes that seem to be a bit more predisposed to this bizarre behavior. Seahorse keepers, and Potbelly and Tigertail breeders in particular, should simply be aware that cannibalism could occur and be on guard lest it becomes a problem.
When it comes to Mustangs and Sunbursts, cannibalism is an extreme rarity. So if you have a pregnant male and are expecting a brood shortly, don’t worry — when the big moment arrives, the odds are overwhelmingly in your favor that Mr. Mom will be a regular Dr. Spock when it comes to his offspring, not a Hannibal Lector!
However, it is not at all uncommon for the filtration system to filter out newborn seahorses along with other suspended particles. Because of the pelagic nature of erectus fry, when a large brood is delivered an aquarium, the result is often a writhing mass of newborn sea horses, hopelessly tangled together at the top of the tank. This dangerous situation develops because a newborn’s first instincts are to head to the surface to fill its air bladder and then to anchor itself to something solid. In the vastness of the ocean this is not a problem, since strong currents rapidly disperse the young, but in the confines of an aquarium, the first hitching post it finds will very likely be the tail or snout of one of its siblings (Giwojna, Jan. 1997). The same mistake is apt to be repeated by the rest of the pelagic fry, as they cluster at the surface, until the entire spawn is snarled together tail-to-tail, head-to-tail, tail-to-snout and so on (Giwojna, Jan. 1997). This is a very common experience when raising seahorses such as Hippocampus erectus, H. reidi, and H. ingens, which produce large broods of pelagic fry. As you can imagine, the surface hugging fry can easily be drawn into overflow box is or the intakes of power filters, so you might want to check your prefilter and filter box for any trapped newborns.
Although the parents are extremely unlikely to cannibalize their young, there are a number of other predators you should be aware of that can take a toll on the fry. For example, any fish with mouths large enough to swallow the newborns are apt to do so. Likewise, decorative shrimp of all kinds — fire shrimp, cleaner shrimp, peppermint shrimp, etc. — will actively prey on seahorse fry and can rapidly decimate a brood. Snails are just fine but crabs of all kinds can pose a risk to the newborns after the fry pass through their pelagic phase and begin to orient to the substrate. Scarlet reef hermit crabs (Paguristes cadenati) and micro-hermit crabs that stay tiny and micro stars (miniature brittle starfish with a leg span no greater than perhaps a dime) are usually okay, but larger hermit crabs and serpent starfish may become opportunistic predators of baby seahorses.
In short, if your pregnant male gave birth at night, and there were only two fry remaining in the morning, it’s far more likely that the newborns were eaten by the aquarium filter or non-seahorse tankmates, rather than by their parents. If so, you will need to take precautions to prevent that from happening the next time your male is due to deliver.
Best of luck with your seahorses and their future progeny!