About 300 million years ago, the skies of the late Palaeozoic teemed with massive insects. Meganeuropsis permiana, a predatory species similar to a modern dragonfly, had a wingspan topping 70 centimetres and weighed roughly 100 grams. Scientists studying these prehistoric giants wondered why insects no longer reach such sizes. Three decades ago they proposed an answer, the so-called “oxygen constraint hypothesis.”

For many years the prevailing view was that dragonfly-sized insects required an atmosphere richer in oxygen to survive because insect respiratory systems are less efficient than those of mammals, birds, or reptiles. As atmospheric oxygen fell, the theory went, the conditions could no longer sustain gigantic bugs. “It’s a simple, elegant explanation,” said Edward Snelling, a professor of veterinary science at the University of Pretoria. “But it’s wrong.”

Insect breathing

Unlike mammals, insects lack a central pair of lungs and a closed circulatory network that transports oxygenated blood to tissues. “They respire through an internal network of tubes called the tracheal system,” Snelling explained.

Air gets into an insect’s body through small openings in the exoskeleton called spiracles. From there it moves into larger tubes, the tracheae, which branch repeatedly into very fine, blind-ending tracheoles. Those tracheoles extend into the tissues, and mitochondria in adjacent cells often sit close to them.

Insects can actively move air through the larger tracheae by flexing their bodies, but this active ventilation does not reach the tiny tracheoles. At that final stage, oxygen must cross into tissue by passive diffusion.

Diffusion is slow, which is the central problem. The oxygen constraint hypothesis argued that as an insect grows, oxygen must travel farther to reach inner tissues.

“As the insects get bigger and bigger, the challenge of diffusion becomes greater,” Snelling said.

To keep muscles from suffocating, a larger insect would need much wider or many more tracheoles to sustain oxygen delivery, implying a structural limit. If an insect became too large, the volume of air tubes required to oxygenate its muscles would occupy too much space, crowding the very muscle fibers they were meant to feed and greatly reducing flight capability.

The findings differed by 18 standard deviations from the predictions of Bell’s theorem, strongly indicating that a superposition of temporal order is an intrinsic feature of quantum mechanics.

Yet the experiment is at a stage similar to early entanglement work: numerous loopholes remain. For example, most photons are lost during the procedure (only about 1 percent of those injected emerge on the other side to be measured). It is technically possible that those losses were biased toward a subset of photons that would otherwise display correlations compatible with hidden-variable explanations.

The team also didn’t separate the equipment by sufficient distances to exclude influences traveling slower than light, and indefinite causal-order experiments introduce a few other peculiarities. Still, the study points toward follow-up experiments that could seal these gaps, and past progress shows such loopholes can be closed.

Usually, when reporting on something this strange, all we can do is marvel at how odd reality is compared with our expectations. This instance is different: the physics involved is already known to have many practical uses.

“The [device used in this work] could also be of interest for applications, as it has been shown to outperform causally ordered processes across a wide range of tasks such as channel discrimination, promise problems, communication complexity, noise mitigation, various thermodynamic applications, quantum metrology, quantum key distribution, entanglement generation, and distillation, among others,” the authors note.

Put another way, being uncertain about the order of time may actually prove useful.

* I wouldn’t have known this work existed if I hadn’t read an excellent summary of it on the American Physical Society news site.

PRX Quantum, 2026. DOI: 10.1103/5t2y-ddmt  (About DOIs).

Lose a life and that neat gun disappears — unless you happened to save beforehand.

Credit:
id Software

Difficulty tuning also needs some work. The easiest option, “Can I Play, Daddy” is comically simple for anyone with modern FPS experience: plenty of item pickups and enemies that attack very slowly and deal almost no damage when they do hit. The default “Bring ‘em On!” setting, by contrast, can feel borderline unfair at times, with foes able to strip away half your health from just a few stray shots.

Although a handful of new enemies trickle in once you get past the shareware levels, I didn’t find any of them particularly compelling. After running through dozens of levels I was itching for a genuinely new weapon instead of just “the old weapon with a higher rate of fire.”

Then there are the broader design choices that feel odd by modern standards. Like many arcade-style games of the era, Wolfenstein 3D tracks a numerical score for each playthrough. You’re also given a limited number of lives, and dying strips you of your weapons (earning enough points can grant extra lives).

However, the ability to save at any moment makes those mechanics largely irrelevant for players willing to save-scum the toughest encounters. And while Wolfenstein 3D carries over damage between levels, completing a full episode forces you to start a brand-new game, with no explicit continuity between episodes.

Look ma — one hand

The biggest surprise on my recent replay of Wolfenstein 3D was how well the game adapts to mouse controls. In 1992 I probably barely knew how to use a mouse, let alone aim a virtual gun with it. This time around, I was pleased to discover the entire game can be played effectively one-handed, without touching the keyboard at all.