[SciShow intro plays]

Hank: In November 1664, the flagship of French King Louis XIV, La Lune, sank in the Mediterranean Sea. Hundreds of people, as well as thousands of now-historical artifacts, went down with it. But last month, one of La Lune’s lost treasures -- a grapefruit-sized vase -- was retrieved, and not by a human. The vase was fished out by a humanoid robot diver called OceanOne.

At 100 meters below the waves, the wreck of La Lune is pretty hostile to human visitors. At that depth, a scuba diver would have just ten minutes to collect samples before spending two hours in a slow, staggered rise back up to the surface, to make sure they didn’t get the bends. Robots do not have that problem.

Now, using machines to help with deep ocean research isn’t new - but they tend to be big and box-shaped - perhaps hard for a human controller to identify with. OceanOne was designed to combine the robustness of a robot with a human’s intuition. The scuba-bot’s human-like hands, with its nimble fingers and opposable thumbs, are fantastic tools for the slow, fiddly work involved in taking underwater samples. Plus, we have an intuitive sense for how they work, so it’s easier for humans to control OceanOne’s movements.

The robot also gives haptic -- or touch -- feedback, which gives researchers more information about the objects they’re handling -- like how heavy those objects are. OceanOne’s maiden voyage -- on April 15 -- was controlled by Oussama Khatib, a computer science professor from Stanford and the head researcher on the OceanOne project. Using the video from OceanOne’s cameras and the human-like joints of its hands and wrists, Khatib could carefully reach the robot’s fingers inside the vase to grasp it.

Meanwhile OceanOne’s artificial intelligence fired multi-directional thrusters, keeping itself stable. Force sensors on the bot’s wrists relayed tactile feedback back to his joystick controls, so he could grip just tight enough to extract the vase from the wreckage without shattering or dropping it. OceanOne was originally designed for studying coral deep under the Red Sea, and that’s still the goal. And eventually, there might be whole fleets of these humanoid scuba bots, scouring the seafloor for science.

So, that’s how we’re using robots to re-capture long-lost artifacts. And some plants are also in the capturing business -- the insect-capturing business. Venus flytraps, for example. You’ve probably heard of them before: they’re carnivorous plants with snap-traps that ensnare bugs, then digest them to help supplement the nutrients the plants get from their swampy soil habitat.

And in a study published this week in the journal Genome Research, a team of scientists from Germany and Saudi Arabia used the science of genetics to uncover two new aspects of the flytrap’s bug-catching: How the plant evolved its trap, and its ability to digest insects. Biologists have thought for a while that the trap’s main structure is made of modified leaves. But without genetic evidence, it’s been tough to confirm. Plus, y'know, normal leaves don’t digest and absorb insects!

So the researchers were trying to figure out whether something completely new evolved here, or if it was just co-opted from another part of the plant. To find out, they looked at RNA profiles – snapshots of which genes each plant tissue is actively using at a given time. Turns out the main trap structures have profiles very similar to those of leaf bases – so we can tick that evidence box. The glands inside of the trap, though, expressed genes much more similar to the Venus flytrap’s roots -- which actually makes sense, since both parts of the plant absorb nutrients. So rather than evolving as a whole separate thing, the trap might have developed as a kind of leaf-root hybrid organ -- combining the abilities of leaves and roots to achieve something neither tissue could do alone.

The researchers also looked into how the plants might have developed the ability to eat insects at all. They tracked RNA profiles during insect digestion, to see if they matched up with any behaviors of other, better-understood plants. Of course, most plants don’t respond to bugs touching them by locking them down and digesting them. But most plants do have anti-insect responses – they’re just a bit more subtle. When an insect like a caterpillar starts munching on a plant, the plant can release jasmonic acid, the plant’s so-called “touch hormone”.

In response to these chemical signals, the plant’s cells release bad-tasting -- or even toxic -- substances, in the hope of driving the herbivore away. The Venus flytrap uses a modified version of that jasmonic acid signal. Instead of releasing compounds to make itself taste bad, jasmonic acid stimulates the flytrap to unleash a cocktail of killer digestive enzymes into its green stomach. In doing so, the plant has basically weaponized more common defense tools for a nutritional bug-boost.

So, as unique as they look, the Venus flytrap’s claws and chemistry are evolutionarily repurposed tools. These traits are examples of exaptation, where rather than developing a new feature from scratch, they’re repurposing traits that are already in place. And in this case, that repurposing led to something unique, amazing and yeah, deadly, as well.

Thanks for watching this episode of SciShow News, which was brought to you by all of our patrons on Patreon. If you want to help support this show, just go to patreon.com/scishow. And don’t forget to go to youtube.com/scishow and subscribe!