Kin Selection for Alarm Calling Behavior in Rodent Species

If a squirrel makes an alarm call and there are no relatives to hear it, does it make a sound?

"I would lay down my life for two brothers or eight cousins."

J.B.S. Haldane

In the presence of predators, animals such as the Belding’s ground squirrel (Spermophilus beldingi) will often emit distinct types of vocalizations called “alarm calls” by naturalists (Dunford, 1977, 782). Since the documentation of these alarm calls, biologists have theorized about their adaptive value and what selective pressures gave rise to them. Two camps have emerged in the literature: those who argue that the alarm-calls are an example of kin selection, and others such who argue that they can be explained via individual selection. In this paper I will explore the evolution of alarm calls using the Belding’s ground squirrel as a case study, and then look at alarm calls among rodents in general. I will argue that while alarm calling may have been selected for via kin selection in certain instances, in most cases it evolved because it conferred individual fitness benefits to the caller, and therefore can be explained without reference to higher level selection.

Sherman (1977) argues that these alarm calls are an example of altruism because they are used to warn relatives of approaching predators but end up endangering the caller, thus lowering his or her individual fitness. His evidence for this conclusion comes from field studies of squirrels’ responses to terrestrial predators such as coyotes, bobcats and weasels. Because of the social structure of these squirrels—females tend to stay and breed near their birthplace until death, whereas males disperse from the area where they were born—female ground squirrels in the same area tend to have a high degree of genetic relatedness, whereas males do not (1248). Thus, Sherman predicted that if kin selection were the mechanism behind these alarm calls, then females would be more likely than males to emit these calls. And indeed, that is what he found: “when a predatory mammal appears, old, reproductive, resident females with living kin are the most likely to call, while males are the most consistent noncallers” (1250). Furthermore, those males who successfully breed, produce progeny, and stay at that location tend to call more than males who were not reproductively successful. Finally, he found that the less living close relatives a female ground squirrel has, the less likely she is to give alarm calls (1249).

This discrimination based on relatedness suggests that kin selection is the mechanism behind alarm calling; therefore, invoking Hamilton’s rule seems appropriate in this case study. In fact, Hamilton actually mentions alarm calls among birds in his famous 1964 paper: “The alarm call of a bird probably involves a small extra risk to the individual making it by rendering it more noticeable to the approaching predator but the consequent reduction of risk to a nearby bird previously unaware of danger must be much greater” (15). The reasons the birds and squirrels take this risk, according to Hamilton and Sherman, is because doing so confers an inclusive fitness benefit to the caller: though they may be endangering themselves by giving the call, doing so warns their relatives—whom they share a lot of genes with—thereby increasing those relatives’ fitness, which indirectly benefits the caller. In principe, this behavior could have evolved via Hamilton’s rule: if squirrels who alarm-call tend to have more living relatives who could pass on their genes, this alarm calling behavior would be propagated.

Hamilton’s rule working in this case depends on the alarm call being a costly behavior for the caller; however, later work by Sherman (1985) suggests that in some cases alarm calling may not endanger the caller but actually confer an individual fitness benefit upon it. For example, in his study of aerial predation on Belding ground squirrels, Sherman found that giving an alarm call was positively associated with escaping from the predating hawk. One might wonder if we are conflating correlation and causation here, as alarm calls could be a consequence rather than a cause of escape behavior. But in fact, Sherman found that of all squirrels who were spotted by a predator and tried to escape, those who gave alarm calls were more likely to survive than those who did not. This makes one wonder whether kin selection is the only mechanism behind the evolution of alarm calling in Belding ground squirrels.

Further explanation by Sherman sheds light on this question and suggests that another mechanism may be at play. First, note that there exist two main types of alarm calls—trills in response to the presence of a terrestrial predators, as were mentioned by Sherman (1977), and whistles, which are given when an aerial predator is spotted and are discussed by Sherman (1985). Sherman showed that while the frequency of the former depends on the relatedness of nearby individuals, whistling tendencies “were not affected by sex, age, or kinship” (Sherman, 1985, 320). Furthermore, while trills tend to attract a predator to the caller, resulting in callers being killed more often than noncallers, whistling may be associated with minimal risk or result in the caller being less likely to be killed (320). Thus, Sherman argued that these whistles were used to “indicate to attacking raptors that they have been seen, thus discouraging their pursuit” and to “startle the predators,” benefitting the caller, not his or her kin (321). This dichotomy between trills in response to terrestrial predators and whistles in response to aerial predators muddies the waters, and suggests that perhaps the two types of calls evolved via two different selective mechanisms, kin selection and individual selection.

In order to see which of these two hypotheses played a greater role in the evolution of alarm calling more generally, Shelley and Blumstein (2005) examined the relationship between sociality, alarm calling, and diurnality among 209 rodent species. They argued that alarm calling in order to communicate with predators—either to let them know they have been detected or to deter them—is inherently risky, and we would expect rodents to try to minimize the amount of risk they face when performing this behavior. One way to do this is only whistle when one is able to see the predator. Shelley and Blumstein cite earlier field work showing that prey will often “visually assess the relative risk of predation and only call when they are not subjected to excessive risk” (169). Therefore, the authors predicted that if alarm calling evolved because of the individual fitness benefits it conferred (via communication to predators), then alarm calling species should tend to be diurnal rather than nocturnal, because diurnal species would be much better at visual assessment (169). Conversely, the authors argue that “if alarm calling evolved via nepotism, we would expect to see social species (particularly those living near kin) more likely to call than are nonsocial species” (169).

The authors found that while sociality was highly associated with alarm calling, diurnality “accounted for three to six times as much variation in whether or not a species alarm called than did sociality” (173). The authors also used a phylogenetic tree to reconstruct the evolution of alarm calling among the 209 species. The tree suggested that in all cases “the evolution of diurnality preceded the evolution of alarm calling” (173). These findings, coupled with the authors’ hypothesis that diurnality is critical to communicating with predators because of the necessity of light for visual assessment, suggest that among the 209 rodent species examined, alarm calling originally evolved not to warn kin, but because it offered an individual fitness benefit to the caller. This is not to say that alarm calling is never used to communicate to kin, or that it could not have been selected for via kin selection; however, it does suggest that alarm calling is used as a means of improving one’s own fitness, not the fitness of one’s relatives. So why do we see Belding ground squirrels alerting relatives by alarm calling? The authors conclude that “while alarm calling may have evolved as a means of communicating to predators…in certain species, alarm calling may have also independently evolved to communicate to conspecifics to warn them of danger” (173).

Sherman’s (1985) paper on aerial predation and the paper by Shelley and Blumstein both force us to reevaluate a prima facie case of kin selection. Yes, it seems that one type of alarm call made by Belding ground squirrels could have evolved via kin selection, and Hamilton’s rule would assist in making sense of when these alarm calls would occur. However, what may problematize the application of Hamilton’s rule here is that alarm calling can actually provide a benefit to the caller, which is not accounted for in the equation. If we set individual fitness benefits aside, the selection for this type of nepotistic alarm calling is an example of what Samir Okasha calls MLS1—multi-level selection one—a type of group selection where a phenotype is selected for not because it provides a direct fitness benefit to its holder but because it increases the average fitness of the population, and by extension the caller’s inclusive fitness (Okasha, 1020). However, it seems that alarm call in response to aerial predators, and alarm calls among rodent species more generally, can be accounted for using only individual fitness.


Christopher Dunford, “Kin Selection for Ground Squirrel Alarm Calls,” The American Naturalist 111, no. 980 (1977): 782ꟷ785.

W.D. Hamilton, “The Genetical Evolution of Social Behavior. I,” Journal of Theoretical Biology 7 (1964): 1ꟷ16.

Samir Okasha, “Multilevel Selection and the Major Transitions in Evolution,” Philosophy of Science 72, no. 5 (2005): 1013ꟷ1025.

Erin L. Shelley and Daniel T. Blumstein, “The Evolution of Vocal Alarm Communication in Rodents,” Behavioral Ecology 16, no. 1 (2004): 169ꟷ177.

Paul W. Sherman, “Nepotism and the Evolution of Alarm Calls,” Science 197, no. 4310 (1977): 1246ꟷ1253.

Paul W. Sherman, “Alarm Calls of Belding’s Ground Squirrels to Aerial Predators: Nepotism or Self-Preservation?” Behavioral Ecology and Sociobiology 17, no. 4 (1985): 313ꟷ323.