How Cheetah Really Hunt
Researchers at the Royal Veterinary College have captured the first detailed information on the hunting dynamics of the wild cheetah (Acinonyx jubatus) in its natural habitat. Using an innovative GPS and motion sensing collar they designed, Professor Alan Wilson and his team were able to record remarkable speeds of up to 58mph.
To date, measurements of cheetah locomotion mechanics have only been made on captive animals chasing a lure in a straight line, with few studies eliciting speeds faster than racing greyhounds. For wild cheetahs, estimates of speed have only ever been made from direct observation or film, in open habitat and during daylight hours.
Wilson’s team developed a tracking collar equipped with a GPS module and accelerometers, magnetometers, gyroscopes. These sensors recorded precise position and velocity data of the animal’s movements.
Collar software monitored the accelerometers creating activity summaries for each brief hunting event. Overall, researchers recorded data from 367 runs by three female and two male adult cheetahs over 17 months (Figure 1).
Figure 1. A cheetah wearing a GPS collar makes a sharp turn (From Wilson et al, 2013)
Data revealed that wild cheetah runs started with a period of acceleration, either from stationary or slow movement (presumably stalking) up to high speed. The cheetahs then decelerated and maneuvered before prey capture. About one-third of runs involved more than one period of sustained acceleration. In successful hunts, there was often a burst of acceleration after the speed returned to zero, indicating that the cheetah was subduing the prey – in this case mainly Impala, which made up 75% of their diet.
The average run distance was 173m. The longest runs recorded by each cheetah ranged from 407 to 559 m and the mean run frequency was 1.3 times per day, so, even if some hunts were missed, high speed locomotion only accounted for a small fraction of the 6,040-m average daily total distance covered by the cheetahs (Figure 2).
Figure 2. Some of the hunt statistics from the study. a, Top speed, averaged over a stride, reached in each run color-coded for outcome. b, Distance covered in each run. c, Top speed in each run coded for terrain type. d, Peak acceleration and deceleration recorded in each run. (From Wilson et al. 2013)
The team was also able to identify factors that make up a successful hunt. Successful hunts involved greater deceleration on average, but there was no significant difference in peak acceleration, distance travelled, number of turns, or total turn angle. Rather, outcome was determined in the final stages of a hunt (rather than hunts being abandoned early to save energy or reduce risk of injury), and the higher deceleration values likely reflect prey captures.
The acceleration power for the cheetahs was double that for racing greyhounds and more than three times higher than polo horses in competition. Interestingly, grip and maneuverability, rather than top speed, were shown to be key to hunting success. Hunts involved considerable maneuvering, with maximum lateral (centripetal) accelerations often exceeding 13ms-2 at speeds less than 17ms-1 (polo horses achieve 6ms-2).
According to Professor Alan Wilson, “Although the cheetah is recognized as the fastest land animal, very little is known about other aspects of its notable athleticism, particularly when hunting in the wild. Our technology allowed us to capture what to our knowledge is the first detailed locomotor information on the hunting dynamics of a large cursorial predator in its natural habitat and as a result we were able to record some of the highest measured values for lateral and forward acceleration, deceleration and body mass.
Source: Modified from materials provided by the Royal Veterinary College, London.
Reference
Wilson, A. M., J. C. Lowe, K. Roskilly, P. E. Hudson, K. A. Golabek & J. W. McNutt (2013) Locomotion dynamics of hunting in wild cheetahs. Nature, 498:185-189. (doi:10.1038/nature12295)
To date, measurements of cheetah locomotion mechanics have only been made on captive animals chasing a lure in a straight line, with few studies eliciting speeds faster than racing greyhounds. For wild cheetahs, estimates of speed have only ever been made from direct observation or film, in open habitat and during daylight hours.
Wilson’s team developed a tracking collar equipped with a GPS module and accelerometers, magnetometers, gyroscopes. These sensors recorded precise position and velocity data of the animal’s movements.
Collar software monitored the accelerometers creating activity summaries for each brief hunting event. Overall, researchers recorded data from 367 runs by three female and two male adult cheetahs over 17 months (Figure 1).
Figure 1. A cheetah wearing a GPS collar makes a sharp turn (From Wilson et al, 2013)
Data revealed that wild cheetah runs started with a period of acceleration, either from stationary or slow movement (presumably stalking) up to high speed. The cheetahs then decelerated and maneuvered before prey capture. About one-third of runs involved more than one period of sustained acceleration. In successful hunts, there was often a burst of acceleration after the speed returned to zero, indicating that the cheetah was subduing the prey – in this case mainly Impala, which made up 75% of their diet.
The average run distance was 173m. The longest runs recorded by each cheetah ranged from 407 to 559 m and the mean run frequency was 1.3 times per day, so, even if some hunts were missed, high speed locomotion only accounted for a small fraction of the 6,040-m average daily total distance covered by the cheetahs (Figure 2).
Figure 2. Some of the hunt statistics from the study. a, Top speed, averaged over a stride, reached in each run color-coded for outcome. b, Distance covered in each run. c, Top speed in each run coded for terrain type. d, Peak acceleration and deceleration recorded in each run. (From Wilson et al. 2013)
The team was also able to identify factors that make up a successful hunt. Successful hunts involved greater deceleration on average, but there was no significant difference in peak acceleration, distance travelled, number of turns, or total turn angle. Rather, outcome was determined in the final stages of a hunt (rather than hunts being abandoned early to save energy or reduce risk of injury), and the higher deceleration values likely reflect prey captures.
The acceleration power for the cheetahs was double that for racing greyhounds and more than three times higher than polo horses in competition. Interestingly, grip and maneuverability, rather than top speed, were shown to be key to hunting success. Hunts involved considerable maneuvering, with maximum lateral (centripetal) accelerations often exceeding 13ms-2 at speeds less than 17ms-1 (polo horses achieve 6ms-2).
According to Professor Alan Wilson, “Although the cheetah is recognized as the fastest land animal, very little is known about other aspects of its notable athleticism, particularly when hunting in the wild. Our technology allowed us to capture what to our knowledge is the first detailed locomotor information on the hunting dynamics of a large cursorial predator in its natural habitat and as a result we were able to record some of the highest measured values for lateral and forward acceleration, deceleration and body mass.
Source: Modified from materials provided by the Royal Veterinary College, London.
Reference
Wilson, A. M., J. C. Lowe, K. Roskilly, P. E. Hudson, K. A. Golabek & J. W. McNutt (2013) Locomotion dynamics of hunting in wild cheetahs. Nature, 498:185-189. (doi:10.1038/nature12295)
Bats Mop up Nectar with Specialized Tongue
Nectar-feeding bats and busy janitors have at least two things in common: They want to wipe up as much liquid as they can as fast as they can, and they have specific equipment for the job. A study in the Proceedings of the National Academy of Sciences describes the previously undiscovered technology employed by the bat Glossophaga soricina: a tongue tip that uses blood flow to erect scores of little hair-like structures exactly at the right time to slurp up extra nectar from within a flower (Figure 1).
Figure 1. (A) Tongue papilla erection (produced by saline injection) in a G. soricina tongue. The papillaebecome erect and the tongue tip lengthens when saline is injected into the vascular spaces. (B and C) Scanning electron micrographs of the injected tongue in dorsal (B) and lateral (C) views.(Scale bars: 1 mm.) (from Harper et al. 2013)
The bat’s “hemodynamic nectar mop,” as the paper dubs the tongue tip, features speed and reliability that industrial designers might envy, said lead author Cally Harper, a graduate student in the Department of Ecology and Evolutionary Biology at Brown University. As a matter of what nature can evolve, she said, the tongue tip is surprisingly clever.
“Typically, hydraulic structures in nature tend to be slow like the tube-feet in starfish,” Harper said. “But these bat tongues are extremely rapid because the vascular system that erects the hair-like papillae is embedded within a muscular hydrostat, which is a fancy term for muscular, constant-volume structures like tongues, elephant trunks and squid tentacles.”
In other words, the bat’s cylindrical tongue has a mesh of muscle fibers that contract so that the tongue becomes thinner but longer (extending farther into the flower). The discovery reported in the paper is that the same muscle contraction simultaneously squeezes blood into the tiny hair-like papillae.
As blood is displaced to the tongue tip, the papillae flare out perpendicular to the axis of the tongue. In their erect state, they not only add exposed surface area, but also width, allowing the tongue to function as a highly effective nectar gathering device (Figure 2).
Figure 2. Blood flow and papilla erection in actively feeding G. soricina. (Upper) Tracings from frames from a color high-speed movie. (Lower) The tongue is shown in pink, the vascular sinuses and papillary veins in red, and the sugar water in light gray. (From Harper et al. 2013)
The entire extension and retraction of the tongue tip occurs within an eighth of a second. Hovering requires a lot of energy, so nectar-feeding bats must get a lot of calories quickly for it to be worthwhile.
Scientists knew about the papillae before this paper, but had always thought they were as passive as the strings on a floor mop. Recent insights by other scientists into the mechanics of hummingbird tongues prompted Harper to take a closer look at the shape of the tongue tip in bats and how it is involved in gathering nectar.
Movie 1. A high speed movie of G. soricina feeding. (From Harper et al. 2013)
Lighting for high-speed video was difficult. “The one thing bats don’t like is a lot of light.”In detailed anatomical studies, Harper was able to observe clear vascular connections between the main arteries and veins of the tongue and the papillae. In experiments she could get them to erect by pumping in saline.But the color videos of bats feeding on nectar, while challenging to create, Harper said, were especially convincing.
“That was one of my favorite parts of the study — the Aha moment,” she said. “We shot color high-speed video of the bats gathering nectar, which is challenging to obtain because color cameras require a lot of light and the one thing that bats don’t like is a lot of light.”
But along with professors and senior co-authors Beth Brainerd and Sharon Swartz, Harper figured out how to focus a lot of light right where the tongue tip would be without shining any of that light into the bats’ eyes. What Harper could then see is that when the papillae extend, they turn from a light pink to a bright red as they fill with blood. “That was really the icing on the cake as far as nailing this vascular hypothesis,” Harper said. Harper said she does not know for sure whether other nectar-feeding bats also have blood-activated papillae on their similar-looking tongues.
Other species such as hummingbirds and bees employ different rapid means of morphing their tongues for improved nectar feeding. Any or all of these highly evolved designs, the authors speculate, could give people technological inspiration.
Source: Modified from materials provided by Brown University and PNAS.
Reference
Cally J. Harper, Sharon M. Swartz, and Elizabeth L. Brainerd. Specialized bat tongue is a hemodynamic nectar mop. PNAS, May 6, 2013 DOI: 10.1073/pnas.1222726110
Figure 1. (A) Tongue papilla erection (produced by saline injection) in a G. soricina tongue. The papillaebecome erect and the tongue tip lengthens when saline is injected into the vascular spaces. (B and C) Scanning electron micrographs of the injected tongue in dorsal (B) and lateral (C) views.(Scale bars: 1 mm.) (from Harper et al. 2013)
The bat’s “hemodynamic nectar mop,” as the paper dubs the tongue tip, features speed and reliability that industrial designers might envy, said lead author Cally Harper, a graduate student in the Department of Ecology and Evolutionary Biology at Brown University. As a matter of what nature can evolve, she said, the tongue tip is surprisingly clever.
“Typically, hydraulic structures in nature tend to be slow like the tube-feet in starfish,” Harper said. “But these bat tongues are extremely rapid because the vascular system that erects the hair-like papillae is embedded within a muscular hydrostat, which is a fancy term for muscular, constant-volume structures like tongues, elephant trunks and squid tentacles.”
In other words, the bat’s cylindrical tongue has a mesh of muscle fibers that contract so that the tongue becomes thinner but longer (extending farther into the flower). The discovery reported in the paper is that the same muscle contraction simultaneously squeezes blood into the tiny hair-like papillae.
As blood is displaced to the tongue tip, the papillae flare out perpendicular to the axis of the tongue. In their erect state, they not only add exposed surface area, but also width, allowing the tongue to function as a highly effective nectar gathering device (Figure 2).
Figure 2. Blood flow and papilla erection in actively feeding G. soricina. (Upper) Tracings from frames from a color high-speed movie. (Lower) The tongue is shown in pink, the vascular sinuses and papillary veins in red, and the sugar water in light gray. (From Harper et al. 2013)
The entire extension and retraction of the tongue tip occurs within an eighth of a second. Hovering requires a lot of energy, so nectar-feeding bats must get a lot of calories quickly for it to be worthwhile.
Scientists knew about the papillae before this paper, but had always thought they were as passive as the strings on a floor mop. Recent insights by other scientists into the mechanics of hummingbird tongues prompted Harper to take a closer look at the shape of the tongue tip in bats and how it is involved in gathering nectar.
Movie 1. A high speed movie of G. soricina feeding. (From Harper et al. 2013)
Lighting for high-speed video was difficult. “The one thing bats don’t like is a lot of light.”In detailed anatomical studies, Harper was able to observe clear vascular connections between the main arteries and veins of the tongue and the papillae. In experiments she could get them to erect by pumping in saline.But the color videos of bats feeding on nectar, while challenging to create, Harper said, were especially convincing.
“That was one of my favorite parts of the study — the Aha moment,” she said. “We shot color high-speed video of the bats gathering nectar, which is challenging to obtain because color cameras require a lot of light and the one thing that bats don’t like is a lot of light.”
But along with professors and senior co-authors Beth Brainerd and Sharon Swartz, Harper figured out how to focus a lot of light right where the tongue tip would be without shining any of that light into the bats’ eyes. What Harper could then see is that when the papillae extend, they turn from a light pink to a bright red as they fill with blood. “That was really the icing on the cake as far as nailing this vascular hypothesis,” Harper said. Harper said she does not know for sure whether other nectar-feeding bats also have blood-activated papillae on their similar-looking tongues.
Other species such as hummingbirds and bees employ different rapid means of morphing their tongues for improved nectar feeding. Any or all of these highly evolved designs, the authors speculate, could give people technological inspiration.
Source: Modified from materials provided by Brown University and PNAS.
Reference
Cally J. Harper, Sharon M. Swartz, and Elizabeth L. Brainerd. Specialized bat tongue is a hemodynamic nectar mop. PNAS, May 6, 2013 DOI: 10.1073/pnas.1222726110
Bats See in 3D Even in the Dark
Animals navigate and orient themselves to survive – to find food and shelter or avoid predators, for example. Research conducted by Dr. Nachum Ulanovsky and research student Michael Yartsev of the Weizmann Institute’s Neurobiology Department, published in Science, reveals for the first time how three-dimensional, volumetric, space is perceived in mammalian brains. The research was conducted using a unique, miniaturized neural-telemetry system developed especially for this task, which enabled the measurement of single brain cells during flight.
The question of how animals orient themselves in space has been extensively studied, but until now experiments were only conducted in two-dimensional settings. These have found, for instance, that orientation relies on “place cells” – neurons located in the hippocampus, a part of the brain involved in memory, especially spatial memory. Each place cell is responsible for a spatial area, and it sends an electrical signal when the animal is located in that area. Together, the place cells produce full representations of whole spatial environments. Unlike the laboratory experiments, however, the navigation of many animals in the real world, including humans, is carried out in three dimensions. But attempts to expand the scope of experiments from two to three dimensions had encountered difficulties.
Ulanovsky chose to study the Egyptian fruit bat (Figure 1), a very common bat species in Israel. Because these are relatively large, the researchers were able to attach a wireless measuring device containing electrodes that measure the activity of individual neurons in the bat’s brain.
Figure 1. An Egyptian fruit bat. (from Dr. Yossi Yovel in the lab of Dr. Nachum Ulanovsky, Weizmann Institute of Science)
The next challenge the scientists faced was adapting the behavior of their bats to the needs of the experiment. Bats naturally fly toward their destination – for example, a fruit tree – in a straight line. In other words, their normal flight patterns are one-dimensional, while the experiment required their flights to fill a three-dimensional space.
The solution was to be found in a previous study in Ulanovsky’s group, which tracked wild fruit bats using miniature GPS devices. One of the discoveries was that when bats arrive at a fruit tree, they fly around it, utilizing the full volume of space surrounding the tree. To simulate this behavior in the laboratory – an artificial cave equipped with an array of bat-monitoring devices – the team installed an artificial “tree” made of metal bars and cups filled with fruit.
Measuring the activity of hippocampus neurons in the bats’ brains revealed that the representation of three-dimensional space is similar to that in two dimensions: Each place cell is responsible for identifying a particular spatial area in the “cave” and sends an electrical signal when the bat is located in that area. Together, the population of place cells provides full coverage of the cave – left and right, up and down.
A closer examination of the areas for which individual place cells are responsible provided an answer to a highly-debated question: Does the brain perceive the three dimensions of space as “equal,” that is, does it sense the height axis in the same way as that of length or width? The findings suggest that each place cell responds to a spherical volume of space, i.e., the perception of all three dimensions is uniform. The researchers note that for those non-flying animals that essentially move in flat space, the different axes might not be perceived at the same resolution. It may be that such animals are naturally more sensitive to changes along the length and width axes than that of height. This question is of particular interest when it comes to humans because on the one hand, humans evolved from apes that moved in three-dimensional space when swinging from branch to branch, but on the other hand, modern, ground-dwelling humans generally navigate in two-dimensional space.
The findings provide new insights into some basic functions of the brain: navigation, spatial memory and spatial perception. To a large extent, this is due to the development of innovative technology that allowed the first glimpse into the brain of a flying animal. Ulanovsky believes that this trend, in which research is becoming more “natural,” is the future wave of neuroscience.
Source: Modified from materials provided by Weizmann Institute of Science.
Reference
M. M. Yartsev, N. Ulanovsky. Representation of Three-Dimensional Space in the Hippocampus of Flying Bats. Science, 2013; 340 (6130): 367 DOI: 10.1126/science.1235338
The question of how animals orient themselves in space has been extensively studied, but until now experiments were only conducted in two-dimensional settings. These have found, for instance, that orientation relies on “place cells” – neurons located in the hippocampus, a part of the brain involved in memory, especially spatial memory. Each place cell is responsible for a spatial area, and it sends an electrical signal when the animal is located in that area. Together, the place cells produce full representations of whole spatial environments. Unlike the laboratory experiments, however, the navigation of many animals in the real world, including humans, is carried out in three dimensions. But attempts to expand the scope of experiments from two to three dimensions had encountered difficulties.
Ulanovsky chose to study the Egyptian fruit bat (Figure 1), a very common bat species in Israel. Because these are relatively large, the researchers were able to attach a wireless measuring device containing electrodes that measure the activity of individual neurons in the bat’s brain.
Figure 1. An Egyptian fruit bat. (from Dr. Yossi Yovel in the lab of Dr. Nachum Ulanovsky, Weizmann Institute of Science)
The next challenge the scientists faced was adapting the behavior of their bats to the needs of the experiment. Bats naturally fly toward their destination – for example, a fruit tree – in a straight line. In other words, their normal flight patterns are one-dimensional, while the experiment required their flights to fill a three-dimensional space.
The solution was to be found in a previous study in Ulanovsky’s group, which tracked wild fruit bats using miniature GPS devices. One of the discoveries was that when bats arrive at a fruit tree, they fly around it, utilizing the full volume of space surrounding the tree. To simulate this behavior in the laboratory – an artificial cave equipped with an array of bat-monitoring devices – the team installed an artificial “tree” made of metal bars and cups filled with fruit.
Measuring the activity of hippocampus neurons in the bats’ brains revealed that the representation of three-dimensional space is similar to that in two dimensions: Each place cell is responsible for identifying a particular spatial area in the “cave” and sends an electrical signal when the bat is located in that area. Together, the population of place cells provides full coverage of the cave – left and right, up and down.
A closer examination of the areas for which individual place cells are responsible provided an answer to a highly-debated question: Does the brain perceive the three dimensions of space as “equal,” that is, does it sense the height axis in the same way as that of length or width? The findings suggest that each place cell responds to a spherical volume of space, i.e., the perception of all three dimensions is uniform. The researchers note that for those non-flying animals that essentially move in flat space, the different axes might not be perceived at the same resolution. It may be that such animals are naturally more sensitive to changes along the length and width axes than that of height. This question is of particular interest when it comes to humans because on the one hand, humans evolved from apes that moved in three-dimensional space when swinging from branch to branch, but on the other hand, modern, ground-dwelling humans generally navigate in two-dimensional space.
The findings provide new insights into some basic functions of the brain: navigation, spatial memory and spatial perception. To a large extent, this is due to the development of innovative technology that allowed the first glimpse into the brain of a flying animal. Ulanovsky believes that this trend, in which research is becoming more “natural,” is the future wave of neuroscience.
Source: Modified from materials provided by Weizmann Institute of Science.
Reference
M. M. Yartsev, N. Ulanovsky. Representation of Three-Dimensional Space in the Hippocampus of Flying Bats. Science, 2013; 340 (6130): 367 DOI: 10.1126/science.1235338
Announcement
30 Mar 2013
I apologize for the reduced frequency of blog posts in the last few months. I am working on a new 6th edition of Mammalogy (by Vaughan, Ryan, & Czaplewski) which will be out in early January 2014.
If any reader would like to submit a post for this blog summarizing a recent publication on mammals, please contact me.
If any reader would like to submit a post for this blog summarizing a recent publication on mammals, please contact me.
Ancestral Tarsiers Had Tri-color Vision
The prevailing hypothesis concerning the evolution of the anthropoid visual system is that ancestral haplorhines were nocturnal. Later their descendants invaded a diurnal niche, with the evolution of highly acute, three-color vision. Now, a new study (Melin et al. 2013) challenges this view, suggesting instead that stem haplorhines already possessed three-color vision before they move into a fully diurnal niche.
In order to better understand how primate vision evolved, Amanda Melin and her colleagues examined tarsier eyes. Tarsiers are tiny nocturnal primates (Figure 1) that branched off from anthropoid primates (monkeys, apes and humans) early on (approximately 56 million years ago).
Figure 1. A Philippine tarsier (Tarsius syrichta). (from Flickr/ Roberto Verzo)
When tarsier genes that encode opsins, a class of color photopigments used in color vision, were analyzed, the results suggested that the last common ancestor of crown tarsiers possessed high-acuity trichromatic color vision (Figure 2). This three-color vision is the same type found in living monkeys and apes. Although this type of vision would normally denote a daytime lifestyle, fossil tarsiers had large eyes, suggesting they were nocturnal.
Figure 2. The enlarged orbits of fossil tarsiers (middle Miocene) predates or is coeval with trichromatic vision in the last common ancestor of crown tarsiers. Colored squares depict opsins and their spectral sensitivities. This ancestral combination of hyper-enlarged eyes and trichromatic vision suggests an activity pattern that includes dim (mesopic) light levels (indicated in grey). Living tarsiers are active mainly under dark (scotopic) light levels (indicated in black). Palaeogeographic maps depict Sundaland during the Late Eocene (35 Ma) and Early Miocene (20 Ma) (from Melin et al. 2013)
Why would ancestral tarsiers have trichromatic vision if they were completely nocturnal? One possibility is that early tarsiers were originally adapted to dim light, such as occurs at twilight or on moonlit nights. Dim lighting would be dark enough to favor large eyes, but bright enough to support three-cone color vision. This suggests that the three-color vision developed before primates became diurnal. Over time tarsiers shifted to a nocturnal niche while anthropoid primates evolved to take advantage of the daytime. The findings give researchers a better understanding of how vision evolved in the earliest of our ancestors.
Reference
A. D. Melin, Y. Matsushita, G. L. Moritz, N. J. Dominy, S. Kawamura. Inferred L/M cone opsin polymorphism of ancestral tarsiers sheds dim light on the origin of anthropoid primates. Proceedings of the Royal Society B: Biological Sciences, 2013; 280 (1759): 20130189 DOI: 10.1098/rspb.2013.0189
In order to better understand how primate vision evolved, Amanda Melin and her colleagues examined tarsier eyes. Tarsiers are tiny nocturnal primates (Figure 1) that branched off from anthropoid primates (monkeys, apes and humans) early on (approximately 56 million years ago).
Figure 1. A Philippine tarsier (Tarsius syrichta). (from Flickr/ Roberto Verzo)
When tarsier genes that encode opsins, a class of color photopigments used in color vision, were analyzed, the results suggested that the last common ancestor of crown tarsiers possessed high-acuity trichromatic color vision (Figure 2). This three-color vision is the same type found in living monkeys and apes. Although this type of vision would normally denote a daytime lifestyle, fossil tarsiers had large eyes, suggesting they were nocturnal.
Figure 2. The enlarged orbits of fossil tarsiers (middle Miocene) predates or is coeval with trichromatic vision in the last common ancestor of crown tarsiers. Colored squares depict opsins and their spectral sensitivities. This ancestral combination of hyper-enlarged eyes and trichromatic vision suggests an activity pattern that includes dim (mesopic) light levels (indicated in grey). Living tarsiers are active mainly under dark (scotopic) light levels (indicated in black). Palaeogeographic maps depict Sundaland during the Late Eocene (35 Ma) and Early Miocene (20 Ma) (from Melin et al. 2013)
Why would ancestral tarsiers have trichromatic vision if they were completely nocturnal? One possibility is that early tarsiers were originally adapted to dim light, such as occurs at twilight or on moonlit nights. Dim lighting would be dark enough to favor large eyes, but bright enough to support three-cone color vision. This suggests that the three-color vision developed before primates became diurnal. Over time tarsiers shifted to a nocturnal niche while anthropoid primates evolved to take advantage of the daytime. The findings give researchers a better understanding of how vision evolved in the earliest of our ancestors.
Reference
A. D. Melin, Y. Matsushita, G. L. Moritz, N. J. Dominy, S. Kawamura. Inferred L/M cone opsin polymorphism of ancestral tarsiers sheds dim light on the origin of anthropoid primates. Proceedings of the Royal Society B: Biological Sciences, 2013; 280 (1759): 20130189 DOI: 10.1098/rspb.2013.0189