 Okay, well first of all I'd like to thank the organizers for the invitation. It's such a beautiful place for a meeting and the the topic that I want to talk about, billfishes, is of course related here to the location because you can find some of these species in the local waters. And because they're so unusual I want to give you a bit of background information before I actually get into the into the group hunting aspect specifically. Okay let's make sure I use this right. So what these fish do is certainly these sailfish they hunt schooling fish like these sardines you see in this picture and they're about 50 sailfish in this picture going after them. And this is a typical scene during the early part of the hunt and the sailfish will then peel off smaller schools of sardines 100, 200 or so and then go after them and in many cases and kill all of them. But the larger school will often return to the deep and it's those smaller subunits that die. And when you look at group hunting in the literature it's really apparent that a lot of the work on group hunting has been done in organisms where individual recognition is very important where the individuals perform different roles during the group hunt where you often have closed membership to groups you might even call them teams in some cases and you see spatially coordinated attacks. So this type of group hunting often requires complex cognitive abilities and when you look at reviews of group hunting they are mostly about these aspects. And what I want to discuss in this talk are alternatives, simple mechanisms that can be involved in group hunting and bring benefits to individuals when they hunt aggregated prey. So here's some pictures of the the billfishes. I'm sure everybody has seen this one probably on your plate and the swordfish most people are familiar with it. It's actually surprising how little is known about these animals in the wild. None of these species can be kept in captivity and some of them live quite deep like the swordfish. So sports fishermen capture them all the time. People eat them daily but very very little is actually known about their life history and hunting strategies. And some of the even more obscure species are these spearfishes here. Some of those are also here in the Mediterranean, this one over here. And the study that I will be talking about took place near Cancun. So we went out into this region here between Cancun and Cuba going far offshore and looking for these fish. And when you find them what you see is this. So this is in slow motion 240 frames per second. It's a top-down view. The sailfish arrives and interestingly swims at about the same speed as the prey puts the bill very close to them and then they get whacked. So you see scales flying, a lot of tissue also in the water so massive damage to these fish. And then the surprising thing is the sailfish does it again and again and again and the fish don't really seem to learn all that much at least from what we can see here in the process. So on this occasion the sailfish was not successful but it will return. There's nowhere to hide for these fish. I show you another one here. So here you can also see some potential handling with the bill. So often when they hit the fish they will actually put the bill towards it and sort of guide it into the mouth. Yeah finding these fish is very hard but once you found them observing them is actually not that difficult. They're not afraid of us and we were initially afraid of them. They sometimes come closer than you might expect. But they never actually touched us. Yeah one of the first things we did when we made these observations we tried to define states and transition probabilities between states and construct simple first order Markov chains to get an understanding of the dynamic process and I quickly take you through this. So they arrive at the fish school, they approach, they put the bill really close to the fish and why the fish let them do that is something we still find puzzling and then they either whack them really hard a slash or they tap a single fish very lightly but this will destabilize the individual. Usually they make prey contact sometimes there's prey handling and then they either consume the fish and sometimes they also miss them and then they return and try again. So re-approach and then it continues. So here's some of the initial work we did with Paolo Domenici and we looked at the speed and acceleration of the head, the snout and the tip of the bill and the tip of the bill reaches quite amazing accelerations during this slashing process and the calculations from Paolo indicated that these accelerations actually higher than those of the sardines than the escape behavior of the sardine. So once the bill is very close it's actually very difficult for the sardine to escape and in the analysis we looked at fish at sardines that are very close to the bill before they actually get hit and what you might expect is that these fish we call them here target fish that they should increase their tail beat frequency to get out of this zone where they might get hit and we compared them to other fish that are outside this area that will be impacted by the bill and we found no difference in the tail beat frequencies or no difference in the overtaking behavior not during bill contact but post bill contact. Suddenly these fish after getting hit they accelerate so clearly they are capable of swimming faster than they initially do but they don't seem to realize that danger is imminent and I show you this in the video one more time you see it quite nicely at the beginning so the sailfish comes and it puts the bill close here and this would now be the moment to really get out of there I mean these fish shouldn't be here and they could easily overtake the others they can do it but they just don't and the question is why so here you see then what happens so one possibility is that the bill is so thin that it's simply below the looming threshold required for the fish to perform an evasive action so normally when predators like sharks dolphins and so on come with their mouth and try to grab a fish the fish will see this large object coming towards them and this triggers this evasion response and probably the bill is too thin it's long enough so that the fish probably believe they are safe from the sail fish and they don't take it as seriously but we wanted to take a closer look at these bills to understand what the surface properties of them are and this is here an image of taken by a micro CT of the bill tip and as you can see it's covered in lots of nasty micro teeth and they cause this damage to the fish when they make contact interesting feature here is also that they're actually pointing forward away from the mouth of the sailfish in virtually all predators they're facing and towards the mouse to retain the prey yeah and then came a long and painful process of obtaining this and this is where my brother and his team did a lot of work so I had initially I easily assumed that I would take my billfish head and give it to a scanning facility that they scan it for me and then that I a week later I would have the data on the number of these teeth and and their properties when it actually took us two years to get this this is almost as bad as counting wildebeest and it's very tough actually and to design an automated micro teeth recognition software so what you see here is a ring we took out the inside from the bill and we just take sort of one slice if you like take out the inside and just concentrate on the surface properties and then we can label these teeth in different colors so if they are still complete we make them here purple and if they are broken they become turquoise and when they are broken they will fall out leaving these cavities and then new teeth will regrow so we went to the dental hospital and had this whole process explained to us I didn't really know anything about teeth before that yeah and then we can look at two distributions along the rostrum so this is close to the mouth this is the bill tip and we find that the percentage of broken teeth increases towards the front you would expect this because that's where most of the biomechanical forces are at play here the upper similes are the Marlin down here the sailfish we see the same trends but the Marlin shows a lot more broken teeth a higher percentage and this is probably explained because they have by this year they have very little and regrowth whereas in the sailfish we see they're regrowing a lot of teeth and I'll come back to that later on when I show you the Markov chains comparing the different attack behaviors another interesting aspect which Sitzelbuy investigated in our team was the presence of an oil gland that they have at the base of their bill discovered by John Widdler a couple of years ago we found similar structures filled with oil in the other species that we work on can see it here oil filled cavities and we looked at the composition chemical composition of this oil looks like the swordfish down here disappeared that was our reference specimen to reproduce John Widdler's data but as you can see the oil composition and chemically is actually very similar between these species and we didn't find anything exciting in this oil by the way it's just normal fish oil I mean I had high hopes of finding some interesting stuff maybe for bill repair or other things and the function that John Widdler proposed when he discovered the structure is that this oil is excreted dorsally onto the forehead of the swordfish and provides hydrodynamic benefits when they swim because they are among the fastest swimmers in the oceans so here you can see the the oil pores in different species so they're actually sort of really perforated on the dorsal surface yeah and an interesting finding here was that in the swordfish this oil really just comes out on the forehead but in a species like the sailfish it seems to cover the entire bill and the question is of course why and because hydrodynamically this wouldn't really make a big difference we expect the highest friction and to occur up here and not so much and on the bill itself and so what we are planning to do in the future with this system here we want to 3d print these bills and then have some with and without microtease within with this oil and then look in flow chambers at the hydrodynamic properties of these objects and also the interaction with live fish to see and whether there are actually properties and that could hydrodynamically camouflage these and bills make them less noticeable and to the fish and during attack when the bill is inside the school. Another aspect which we investigated and was the the function of the sail and only one of these species has a sail the others don't so this big dorsal fin and it turns out that they always raise it just before they attack and it prevents the head from swinging so you can see here when it's down then head and tail are in anti-face and they can then minimize the swinging of the head when they put up the sail and it comes at a tremendous hydrodynamic cost they slow down they need to compensate for that but it seems to be very important for them to keep the bill really still when they put it into the fish school and then hit the fish with very great precision and this is a work by Stefan Omaras. Yeah this probably might be a bit small for the people at the back and we were wondering I mean why does only one species have this massive sail and and it's absent in all almost all the others and we looked at the attack sequences in different species comparing it to the marlins here and it turns out that some of these other species that also hunt in groups when they approach the fish school they will actually speed up massively so the tailbeat frequency goes up and they plow through the fish school often divided and they do it with an open mouth and they try to to grab fish and they don't actually make that much use of the bill for hitting the fish and and this might also be why and they don't really massively regrow teeth micro teeth on the bill because they're not reusing using the bill as much yeah and an aspect of the speed and obviously the larger the the predators and the faster they can swim the bigger the stride lengths and this can be very important and predator-prey interactions the bigger fish or bigger animals are usually faster and the smaller ones are more maneuverable and and the sailfish were considered the record breakers in the fish world this went back here to estimates from the 1940s and 1960s with estimates of over a hundred kilometers per hour this always sounded a bit dubious to be honest and once we saw how the fish attack their prey and that they swim at about the same speed and that I can snorkel alongside them we were wondering what do they actually need these high speeds for but of course there might be situations that we haven't seen where they go very fast so it was really a matter of measuring it and we took two approaches to this one is a muscle twitch method which gives you a theoretical maximum and the other one is that you put attached accelerometers to these fish in the wild let them hunt for a while and then get your accelerometer back and look at your values this was largely done here by Morton Svensson so we looked at the stride lengths of the fish and that can be attained in one tailbeat done with a higher resolution sonar then you look at the muscle contractions and how that corresponds to tailbeats and then you can work out an estimate of the maximum swim speed without taking into account the drag underwater and you see the different species here we captured all large predators that we could find in the pelagic waters there the sailfish were the fastest among those that we tested but you can see that these values don't even reach 40 kilometers per hour so they were actually a lot slower than than we had expected so we removed this claim from Wikipedia and elsewhere about these 100 kilometers it's actually very interesting when you look at the original claims from the 1940s the Russian article from 1960 was apparently never translated into English and I think a lot of people just trusted and that it presented measurements but it didn't it actually just told the story and the 1940s article was published in country and life and I think it was never really meant to be a sort of serious scientific claim but it wasn't just quoted by journalists you find it in a lot of scientific textbooks yeah the measurements from the accelerometers I don't want to show in detail but they correspond roughly with what we get from the muscle twitch method but they are slightly lower as you might expect because we had it factors in the drag underwater so when these fish attack I showed you already what happens when they engage with the fish but interestingly it's usually one sail fish at a time so they almost queue up for this kind of thing you see one doing it here and the others are waiting so there's a lot of turn taking involved and when they hit the fish they usually injure quite a few about here to an average but only every fifth attack is successful so this means over time a lot of injured fish are building up in these groups it's actually quite a stressing to watch this because I mean you see hundreds of fish that are heavily injured and some are dying as you watch them they get hit so many times and and eventually they get picked off but for the sail fish this means while they are waiting somebody else is injuring fish and as they get more injured they become easier to capture we see this relationship here capture rate as a function of injury level so the waiting time isn't just wasted time for these sailfish but something beneficial is actually happening for them and this increases their efficiency a lot so this is a model by Pavel Romantzchuk which indicates that the efficiency for the individuals is highest when they're in group sizes of about here 10 or 12 individuals and this can potentially provide benefits for groups of sailfish of up to 50 60 or more individuals before they fall below the level of the efficiency of a single individual and this is particularly important if they are under time pressure then efficiency really matters and what we found in the sailfish is that when darkness comes or twilight they abandon the hunt they are very visual and the fish often get stolen by other predators and dolphins and other species come along and take the sardines away also very annoying for us when that happens because the moment you hear the dolphins coming you know your study for that day is over and they have a different technique they actually hit with their tails into the fish schools very much like killer whales and destroy the schools very quickly within seconds yeah and if some if there is this benefit of others injuring fish for you then this of course begs the question are there maybe sailfish around that wait during the initial hunting period for others to injure a lot of fish and then when the sardines are quite weak then they participate and start sort of harvesting sardines when it's easy so it was basically asking the question are some sailfish sitting back during the early attack stages and pavits model indicated that there is a relatively small parameter space where we might expect this and when the attack behavior is energetically quite costly then this could potentially be the case that some individuals sit back another prediction from this is that when if sailfish are monitoring different sardine schools that are under attack and then they might focus on those that already heavily injured so we would expect that in the late attack stages more sailfish and come to those and schools of sardines and once we were able to identify the sailfish individually from their dorsal fins and we could look at some of these aspects here in more detail and we didn't actually see any sailfish and that are sitting back and waiting and we also did not see an increase in sailfish numbers towards the end of the attacks but I have to admit and we haven't monitored that many attacks where we see it from fairly early on to the very end and because it's not that easy to keep up with the fish for the entire attack. This brings me to the topic of letterality so when we got the first images of these builds from the micro CT we noticed that at the tip some of them have these one-sided abrasions and so we were beginning to wonder whether some of these fish are actually right-handed and left-handed when they hit. We started looking at the literature and letterality is very widespread in the animal kingdom we all know it from humans as well they are right-handers and left-handers and there's individual level letterality and population level so humans typical example of the letter about 85 to 90% of humans are right-handed the others are left-handed and this apparently is the case and in human populations throughout the world and to my knowledge it's not entirely clear why most of us are right-handed there are a number of attempts to explain why this is I've seen some quite amusing ones even in the sort of disciplines from the humanities that sort of people developed this because they were holding in combat the shield in the left hand to protect the heart and the sword in the right hand but handedness actually goes back a lot longer than this type of combat nevertheless combat is a very interesting field to look at for handedness anybody who does martial arts will know that the proportion of left-handed individuals in martial arts is up to 40% in fact in almost all sports where it's one-on-one also in tennis and so on because the rare phenotype is then at an advantage so you see more left-handers in those kind of situations which is an indication there are costs and benefits of these kind of things even to the state in humans yeah why specialize at all handedness why in the first place it can really increase task efficiency you can do it a lot faster you can do it more accurate you hesitate less but it can make you predictable I mean if you are fighting with someone somebody boxing and you know this person is right-handed you can you can anticipate certain things it can be a huge advantage and being predictable can then be a problem if you become predictable to your prey so are the the sailfish am lateralized and if so are there adaptive benefits of this yeah for this we needed to identify them we use the shape of the dorsal fin this is work with Ralf Curvers and we looked at the expected degree of laterality here so this is a lateralization index so you take the attacks to the left side minus the right side divided by the total number of attacks so this ranges from minus one to plus one and then in the middle and you are not lateralized those individuals these are the expected distributions here and we see the empirical one here so there seem to be evidence for handedness because we get the spine model distribution capture success clearly increases as laterality goes up so more lateralized individuals seem to be more capable at capturing prey and they are clearly much better with their preferred side than their unpreferred side thing is pretty familiar to most of us if you try to throw something with your for me it would be the left hand and well I can't even open this bottle now and but yeah I mean you will see there's a big difference and if you look at the difference in performance between preferred and unpreferred then this increases and as a function of how lateralized they are so there's clear evidence that these fish are gaining something from being lateralized but unlike in humans and some other organisms left-handers and right-handers were about equally frequent in this population okay and this creates an interesting opportunity for the evolution of group living on the y-axis we have here the group lateralization index so we look for a given group at how many attacks are done to the left and to the right so in a way this indicates how predictable a group is in its attack behavior the higher this value the more predictable it is and we can see that predictability goes down as group size goes up so if left and right-handers are an equal proportion in the population and the fish come together randomly in groups of different sizes then for these groups predictability of attacking behavior should go down that's the prediction and you see the empirical values are the unfilled circles here and you can see that this fits the curve quite well if the dots were predominantly below this curve then this would be an indication that they actively disassort and that left-handers are preferably together with right-handers and vice versa but we didn't find evidence for this but it's a small sample size admittedly but of course a very simple way for these fish for the sailfish to get a benefit from being in a group because they can specialize at the individual level without becoming predictable to their prey so to summarize I showed you one effect here of group hunting making these animals and more efficient by means of this turn-taking and injuring more fish than they actually take and allowing them to specialize without becoming predictable and the latter also makes this interesting prediction that predators more generally might be able to diversify more when they are in their hunting strategies when they are part of groups than when they are hunting alone because this should keep and predictability down and this would be interesting to test for some other species and this brings me to the end of my talk I just quickly want to highlight a few of those people or organizations I haven't mentioned yet because this was certainly one of the bigger studies where I relied on a lot of expertise from other people and other teams so first of all the people who kept us safe in the water our Mexican colleagues then all the teams who did the scanning and the interpretation of what we had scanned and Kevin Boswell from Florida University was brave enough to swim into the open ocean with his sonar thank you very much