Hacker News Reader

Comment by NathanielBaking | Hacker News Reader

As a beekeeper for several years I have posited that bees have evolved to die after stinging because the act of stinging can cause pathogens from the creature being stung to adhere to the stinging bee. This would create an easy vector into the hive where it would then spread eventually infecting the queen. It would be interesting if someone with some game theory experience could model this behavior.

View on HN

I’m not sure game theory experience is the missing bit here. What are the chances of a bee getting infected by stinging an intruder? What are the chances of a bee not stinging an intruder and getting infected? (By just doing normal bee business, and during a hive intrusion) What are the chances that the infection will spread from that bee to a next inside the hive? What are the consequences of an infection?

Weather or not what you are saying makes sense depends on the answer to these questions. But these questions are not in the field of game theory. These are questions of bee epidemiology really.

Also, what are the consequences to the hive of losing a bee? It's a pretty small energy investment to make a new one. The reproductive success of the bee is the reproductive success of the hive - it doesn't have any way to pass on its individual genes.

I know a reasonable amount of game theory and I think you're looking at the wrong field for insight. The stung animal isn't making a strategic choice to carry pathogens around which might "payback" the some hypothetical bee hive which may or may not sting them. From the bees perspective the choice of strategy is independent of any other actors strategic decisions. Its just a normal optimization problem, not game-theoretic coupled optimization problems.

Good point. I took a chance with my Great Courses only knowledge of game theory.

Another point people are pointing out is that bees don't usually die after stinging other bee. Other bees are the easiest vectors for the infections I have had to deal with. I wonder what evolutionary advantage of not returning to the hive grants when a bee stings a larger animal (mammal, bird, etc).

The mating process of the drone male bees (with suicidal semelparity) maybe has a more obvious (but similar?) answer in that they have served their role and want to maximize their reproductive effectiveness.

Certainly attaching your venom pump to a mammal will help discourage it now and in the future. Given average bee lifetimes are 3-4 weeks anyway, it's also not much of a sacrifice for the hive.

The genetic similarly of bees to their queen is higher (75%) than most species, so kinship theory may point to less individualistic behavior as well.

https://www.lakeforest.edu/news/the-emerging-study-of-kinshi...

Not to their queen, but to each other.

Maybe the bee dying is just the side effect of what it's really trying to do. Maybe if the the bees stingers did not have barbs they would just annoy a bear a little and he would happly suffer a few stings to get the honey. But with barbs the stinger stays in the mammal and irritates the skin for longer periods of time. If course this means the bee has its stinger ripped off and dies but his stinger still causes damage and has deterent affect. Also I imagine a bee that did not die when stinging still stood a good chance of dying due to slap from hand or paw anyway. Or if a bear is raiding hive thoese bees are died anyway if they don't stop him so you might as well inflict maximum damage

A stinger could, in principle, evolve to becomd separable and regenerable, like a lizard tail.

Having been stung by bees, one really wants to get the stinger out. If one doesn't have fingers, that could be very annoying to the stingee, and certainly a further deterrent.

I can vouch for the veracity of bumblebees. I disturbed a nest and was chased by a couple of bees. I took 6 hits from one bee. 1st one felt like a rock hit me. Each after was similarly unpleasant. And after the 2nd each one was while I was in full sprint. I eventually escaped by jumping in a car.

The pathogen itself is being optimized to spread though. I think game theory still applies to natural selection in some sense.

Not in this case as the pathogen isn’t going to react to the evolutionary defences of the bee by adjusting its infectiousness.

On evolutionary timescales, it does.

Of course it is. The animals producing the pathogen will adapt to the bees

>> Not in this case as the pathogen isn’t going to react to the evolutionary defences of the bee by adjusting its infectiousness

> Of course it is

That is incorrect. Diseases might evolve to be more infectious. This is not a certainty.

There are many directions a disease can change and selecting to be more likely for a specific species is dependent on an N-vector calculus, with the most impactful cofactors being the length of time they (an attacking bee and a pathogen and the bee victim) coexist in the same environments and the utility of such an adaptation and the nature of the disease (ie incubation time, effects, etc) over time. It's likely the nature of a specific disease and the victim are going to vary more over time, making a favorable situation to prevent any attacking bee to simply not return.

The pathogens may or may mutate but this is separate to bee mutations. They don’t respond to each other.

In this case the actor isn't the bee but the bee's genetic programming, which only propagates via the queen and hive. If the programming allows a bee to cary a fatal disease back to the hive, the program ceases to propagate and may cease to exist. This provides strong selection pressure against fouling the hive.

Since only the Queen reproduces in a hive and all the bees are its sterile offsprings, IMHO the whole hive is equivalent to the Queen evolutionary-speaking. Individual bees are not independent actors. So I think there is no game theory here and this is no different than a giraffe growing a longer neck.

As a fellow beekeeper, I like the cut of your jib - seems like you've done some interesting thinking on this point and it certainly seems like an idea that could merit further research. That said, my personal answer to this is in line with what the author concludes is a possible answer: "Perhaps because they're disposable parts of a larger super-organism which has evolved by multi-level selection" IE: basically everything the hive does, the hive does together and individual bee matter very little. In fact, upon reflection it might make sense to view an individual bee as more akin to a cell in mammals as opposed to an individual organism, IE: cells can die or be regenerated, but the overall body continues. Given that bee lifespans are measured in weeks, even a bee that does not sacrifice itself stinging in defense of the hive will have a very short existence - and be replaced by thousands of others in short order if the queen is healthy.

Interesting thought. My thought is that it wouldn't necessarily take many generations to evolve this trait. A single generation of bees dying after stinging would have a huge advantage over such a hypothetical disease.

Wondering: 1. Do all bee species die after stinging? 2. Do bees always die? What's the mortality rate?

There may be plenty of diseases out there living off bees and other hosts, that are pervasive enough to provide sustained evolutionary pressure to facilitate this development.

We are not speculating on much here. Some others have mentioned the unit economics of losing single bees vs. Losing the whole hive are really excellent.

The bees often don’t die immediately, and could very well return to the hive. Also, except in accidents, in nature this death generally happens at the hive, when eg a bear is eating it.

> I have posited that bees have evolved to die after stinging because the act of stinging can cause pathogens from the creature being stung to adhere to the stinging bee.

OK.

But why do bees die after stinging, but wasps don't.

Bees can sting other insects just fine. Also these Bees can’t reproduce, so that seems like an obvious difference.

From the hives perspective it’s a question of effectiveness vs the utility of individual bees remaining lifespan. Being even slightly more effective at discouraging mammals from raiding a hive for honey is presumably worth the loss of individual bees.

Wasps on the other hand lack the wealth of a bee hive so presumably different tradeoffs are worthwhile.

I have been looking for this comment!

Bees need to defend themselves against other insects. The stinger doesn't detach when the threat isn't great enough to warrant suicide.

I don't know this fo ra fact. Actually I'd never even heard this before. I'd surmise the act of pulling a stinger out of another insect may be less than the amount required to pull the stinger form the bee's body, thus preserving the bee.

Because wasps don't live in large colonies. Thus losing one wasp would be more damaging than the risk of infection.

Also wasp nests aren't as hot and humid as bee hives

Not every path has to converge.

Do mammals hunt wasps for their honey?

Not honey but proteins.

Bees have a barbed stinger that they can't pull out. Wasps don't have a barbed stinger.

I think another possible evolutionary reason may be preventing rogue drones.

A rogue drone that can sting repeatedly is a huge risk to a hive since security screening is only enforced at entrance and the rogue one(s) are inside the hive. This way, with the one-shot stingers, the worst possible damage is the loss of another bee, and the problem drone has taken care of itself.

What's your take on that hypothesis?

Do you mean rogue "worker bee" rather than "drone"? Drones don't have stingers.

But as lanrei comments, "When bees sting other insects they can sting them multiple times, like a wasp."

During "heatballing", may bees not also contract pathogens?

Pathogens are pretty species specific.

It's quite unlikely a bee could get infected by whatever a mammal is carrying.

And it would be easier to evolve "needle cleaning" (flushing acid over it for example).

I don't think game theory is involved as there is no 'conscious' decision involved. Your theory is purely based on evolution through natural selection: the death of a bee after it has strung would increase the survival chances of the hive enough that, over time that (initially random) trait would be favoured.

Game theory doesn't have to operate at the level of individuals making conscious decisions. Consider the concept of the Evolutionary Stable Strategy:

https://cs.mcgill.ca/~rwest/wikispeedia/wpcd/wp/e/Evolutiona....

OK, thanks. Even so, IMHO this is 'simple' evolution through natural selection because a hive is equivalent to a single individual, not a population, evolutionary-speaking since a single individual reproduces and all the bees are also the offsprings of a single individual. So ultimately, IMHO the characteristics and behaviour of the hive only depend on the Queen and its survival. Which can also explain in part why sacrificing individual bees might be a successful strategy.

Reply on news.ycombinator.com