A toad swallowing a beetle looks like the end of a hunt. In a Japanese laboratory, however, that moment sometimes marked the beginning of an escape that took more than an hour to complete.

In a 2018 Biology Letters study, Shinji Sugiura and Takuya Sato offered the bombardier beetle Pheropsophus jessoensis to two toad species, Bufo japonicus and Bufo torrenticola. Every toad swallowed its beetle. But 43% subsequently vomited it out, after delays of 12 to 107 minutes, and every expelled beetle was alive and active.

Swallowing was not always a successful finish.

A defense that still worked inside

The researchers concluded that the insects discharged hot defensive chemicals inside their captors, provoking vomiting. Size mattered: larger beetles were more successful, while smaller toads were more likely to throw them up. These were laboratory outcomes, not a measured survival rate for all encounters in the wild.

That distinction changes how to read the spectacular footage. A demonstration can establish that an escape route exists without establishing how frequently it is used across a landscape. Watching a predator catch something answers one question. Following what happens afterward can answer a different one.

How an insect makes a hot chemical jet

The machinery behind bombardier defenses has been examined separately. In research reported by MIT in 2015, investigators used high-speed X-ray imaging to observe the internal workings of living bombardier beetles as they sprayed. The camera captured 2,000 frames per second.

Chemical precursors meet in a reaction chamber, generating an irritating secretion, heat and pressure. The resulting jet can approach the boiling point of water. It emerges in pulses rather than one uninterrupted stream.

The imaging revealed a valve and flexible membrane between the storage and reaction chambers. Rising pressure closes the passage; after discharge lowers the pressure, it opens again. That cycle permits another pulse. The researchers suggested that the pauses could help protect the chamber by allowing some cooling between bursts.

This was a separate investigation of the spray apparatus, not an X-ray movie of the Japanese toads’ stomachs. Together, the studies address different scales of the same biological problem: how a chemical defense is generated, and what happens when a predator encounters it.

Some frogs reject the meal much earlier

Another 2018 experiment by Sugiura tested the same beetle species against black-spotted pond frogs, Pelophylax nigromaculatus. Of 28 frogs offered live beetles, 26 rejected them before swallowing. Nineteen stopped after tongue contact, and seven spat the insects out of their mouths.

Only two swallowed their prey. One digested its beetle; the other vomited a living beetle 18 minutes later.

The comparison with dead beetles was revealing. Twenty-four of 28 frogs rejected those too, even though dead insects could not fire defensive bursts. Sugiura concluded that bombing was not essential to prevent swallowing in this particular predator. Surface chemicals or other characteristics detected on contact could contribute, although the precise deterrent required further investigation.

These results should not be added to the toad experiment’s percentage. They involved another predator and a different question. They show why describing an animal as chemically defended is only a starting point: researchers also need to identify when the defense matters and which attacker it deters.

The discovery encouraged another search

In a later Kobe University interview, Sugiura explained that the bombardier finding encouraged him to consider whether other swallowed insects might escape. His work with the small aquatic beetle Regimbartia attenuata revealed a different route through a frog’s body.

Initially, he found a beetle back in its container after having confirmed that the frog had swallowed it. Subsequent filming showed it emerging from the frog’s cloacal opening. In that case, the insect escaped through the other end of the digestive tract.

Experiments restricting the aquatic beetles’ leg movement prevented escape. Sugiura therefore suspected movement played a role, while acknowledging uncertainty about exactly how the insects made their way out. This was a different beetle with a different escape mechanism, not another example of bombardier chemistry.

Why the end of the observation matters

The shared lesson is methodological: an observation stopped too early can misclassify the outcome. A tongue strike, entry into a mouth and disappearance down a throat are visible events, but they need not be interchangeable measures of predation success.

Following the full sequence also makes comparisons more useful. Rejecting prey on contact, spitting it out and returning it from the stomach represent different opportunities for a defense to act. Combining them into a single description of an animal being eaten would hide those differences.

It is also useful to separate a count from an explanation. Recording an escape establishes the event; identifying the mechanism requires comparisons that alter the suspected defense or test a different predator. Those additional steps turn an arresting animal encounter into an experimental question.

For the beetle, the decisive boundary was getting back outside.