Why Animals Construct Helical Burrows:
Construction vs. Post-Construction Benefits
J. Sean Doodya*, Shivam Shuklaa and Stephen T. Hasiotisb
aDept. of Integrative Biology, University of South Florida, St. Petersburg Campus, 140 7th Ave. South, St. Petersburg, Florida 33701 USA
bDept. of Geology, University of Kansas, 1475 Jayhawk Blvd, Lindley Hall, rm 215, Lawrence, Kansas 66045 USA
*Corresponding author: jsdoody@usf.edu
Abstract
The extended phenotype of helical burrowing behavior in animals has evolved independently many times since first appearing after the Cambrian explosion (~540 million years ago). A number of hypotheses have been proposed to explain the evolution of helical burrowing in certain taxa, but no study has searched for a general explanation encompassing all taxa. We reviewed helical burrowing in both extant and extinct animals and from the trace fossil record and compiled from the literature 10 possible hypotheses for why animals construct helical burrows, including our own ideas. Of these, six were post-construction hypotheses—-benefits to the creator or offspring, realized after burrow construction—-and four were construction hypotheses reflecting direct benefits to the creator during construction. We examined the fit of these hypotheses to a total of 21 extant taxa and ichnotaxa representing 59–184 spp . Only two hypotheses—-antipredator, biomechanical advantage—-could not be rejected for any species (possible in 100% of spp. ), but six of the hypotheses could not be rejected for most species (possible in 86–100 % of spp. ): microclimate buffer, reduced falling sediment (soil), anticrowding, vertical patch, and the above two hypotheses. Four of these six were construction hypotheses, raising the possibility that helical burrowing might have evolved without providing post-construction benefits. Our analysis showed that increased drainage, deposit feeding, microbial farming, and offspring escape could not explain helical burrowing behavior in the majority of taxa (5–48%). Overall, the evidence does not support a general explanation for the evolution of helical burrowing in animals. The function and evolution of the helix as an extended phenotype would seem, at least in some cases, to provide different advantages for different taxa. Although direct tests of many of the hypotheses would be difficult, we nevertheless offer ways to test some of the hypotheses for selected taxa.
Key words: behavior; extended phenotype; costs-benefit; spiral burrow; helix; ichnotaxa
Introduction
An extended phenotype, when referring to a single species, includes some architecture or entity (e.g., beaver dam, termite mound) in which the phenotype is the fitness of the construction for survival and reproduction (Dawkins, 1982, 2004). Scientific interest in extended phenotypes has been widespread and sustained, encompassing diverse areas ranging from parasite manipulation of hosts (Hughes and Libersat, 2019) to relationships between genomes and phenotypes (Hunter, 2018) to human sexual selection theory (Luoto, 2019).
Burrow architectures are classic extended phenotypes that show great diversity and complexity and can reflect important fitness-related traits (Hansell, 2005). In a classic example, the old-field mouse,Peromyscus polionotus , constructs complex burrows with a long entrance tunnel that leads into a nest cavity and a secondary escape tunnel, while its sister species, the deer mouse (P. maniculatus ) builds shorter, single-tunnel burrows (Weber et al., 2013). The complex burrowing behavior of P. polionotus is derived, has a strong genetic component, and its putative adaptive function is to facilitate escape in an open, exposed habitat (Wolfe and Esher, 1977; Weber and Hoekstra, 2009; Weber et al., 2013).
A diversity of terrestrial and aquatic animals excavates mysterious helical burrows that comprise multiple, symmetrical spirals descending into a medium (i.e., substrate, sediment). The first of these kinds of burrows appeared with the Cambrian explosion ~540 million years ago (e.g., Goldring and Jensen, 1996; Hasiotis, 2012; Sappenfield et al., 2012; Zhang et al., 2015) and many others are known only from fossils, including the remarkable 3-meter-deep burrows assigned to the ichnotaxon Daimonelix that were constructed by the terrestrial beaver Palaeocaster from the Miocene (Barbour, 1892; Martin and Bennett, 1977). Various forms of Daimonelix are now known to have been constructed by a variety of terrestrial vertebrates since the Late Permian, approximately 260 million years ago (Smith, 1987; Fischer and Hasiotis, 2018; Raisanen and Hasiotis, 2018). Living examples of species that construct helical burrows include terrestrial taxa, such as some pocket gophers, monitor lizards, and scorpions, and marine forms such as some shrimp and some polychaetes (e.g., Powell, 1977; Koch, 1978; Löwemark and Schäfer, 2003; Hasiotis and Bourke 2006; Netto et al., 2007; Wilkins and Roberts, 2007; Doody et al., 2015).
The relentless, independent evolution of helical burrowing behavior across disparate unrelated taxa dating back hundreds of millions of years attests to its apparent utility. Yet, the reason(s) for the evolution of the helix from more simple burrows remains largely speculative and unresolved. One problem is that little is known of the natural history of the tracemakers of the fossil burrows or trace fossils (e.g., Bromley, 1996; Hasiotis 2007). Another problem is no study has considered all taxa in the search for a general explanation, although a diversity of functions is certainly possible and is sometimes suggested. To illustrate, collectively, helical burrows have been proposed to have evolved to buffer microclimate of the burrow from harsher outside conditions (e.g., temperature, aridity, salinity), promote drainage during flooding, thwart predators, reduce burrow interference with conspecifics, increase surface area to expose more sediment for deposit-feeding, or promote bacterial farming (e.g., Martin and Bennett, 1977; Koch, 1978; Dworschak & Rodriguez, 1997; Myer, 1999; Netto et al., 2007; de Gibert et al., 2012; Doody et al., 2015; Raisanen and Hasiotis, 2018; Muñiz and Belaústegui, 2019). Moreover, some have hypothesized that helical burrows could serve multiple functions (behavioral category polychresichnia; Hasiotis, 2003) (Koch, 1978; Netto et al., 2007; Carvalho and Baucon, 2010; deGibert et al., 2012; Raisanen and Hasiotis, 2018).
These hypotheses offered to explain helical burrowing behavior generally invoke adaptation in the form of ‘post-construction’ benefits to the creator. Indeed, an adaptive function(s) of helical burrows seems plausible, given its multiple origins and the increased effort required to create a helix compared to a simpler (straight) burrow of the same incline (volumetric calculations by Meyer, 1999). Much less attention has been given to ‘construction’ costs-benefits of helical burrows, or those that provide no benefit to the occupant after burrow construction, but rather are restricted to cost-benefits of burrowing behavior itself. In an exception, White (2001) estimated the total cost of helical burrow construction in scorpions by calculating the net cost of soil transport and the costs of the animal moving itself and soil horizontally, and vertically against gravity. Helical burrowing scorpions minimized both (i) the energy used during burrow excavation by descending as steeply as possible, and (ii) the energy required for burrow maintenance, by constructing an entrance run that is shallower than the angle of repose (rest) of dune sand (White, 2001).
Recently, two species of monitor lizards were found to excavate deep, helical burrows for the sole purpose of nesting and the authors discussed the fit of some post-construction (adaptive) hypotheses for the function of the helix in these lizards (Doody et al., 2014, 2015, 2018a, b, 2021). The communal nests of the yellow-spotted monitor (Varanus panoptes ) and Gould’s monitor (V. gouldii ) are by far the deepest extant vertebrate ground nests known (averaging 2–3 m deep and reaching 4 m deep). The burrows are soil-filled, and consist of an incline to a depth > 1m, followed by 2–7 tight descending spirals that terminate in a slightly enlarged nest chamber (Fig.1a,b). Mothers excavate the burrows, lay their eggs, and then abandon the burrows. Unlike scorpions, the lizards do not transport the soil out from the burrows—-they remain soil-filled and the lizards ‘swim’ through the excavated soil after laying eggs. Thus, White’s (2001) calculations and conclusions for scorpions and potentially other animals do not apply to the lizards. Although a few of the post-construction hypotheses might apply to the lizard burrows (e.g., the helix potentially reducing egg predation), ‘construction’ benefits may provide a more parsimonious explanation than ‘post-construction’ benefits; e.g., the helix preventing soil from falling back into the burrow as it’s removed; falling soil might make burrow creation impossible or extremely difficult in a steep and straight inclined burrow.
Herein we generate and review the major ‘construction’ and ‘post-construction’ hypotheses for why animals evolve the extended phenotype of helical burrows. We address 10 hypotheses including those extracted from the literature and our own. We ask if any of the hypotheses could be general for all taxa. If not, we ask the opposite question: Why did helical burrowing evolve for different reasons in different taxa? To address these two overarching questions, we examined the fit of each hypothesis to each of 21 taxa representing 77–188 spp., based on natural history, behavior, and deductive reasoning, from published sources. We outline potential future tests of hypotheses for selected taxa.
Materials and Methods
We surveyed the scientific literature for evidence of helical burrows in invertebrates and vertebrates using two-way searches on Google Scholar and Google (general web search). We excluded species for which there was only one (or less) spiral turn (e.g., Linsenmair, 1967; Basan and Frey, 1977; Hembree, 2009; Carvalho and Baucon, 2010; Kinlaw and Grasmueck, 2012; Hembree, 2014; Mikus and Uchman, 2013; Vazirianzadeh et al., 2017; Paul et al., 2019). We also excluded studies describing burrows that were weakly helical, weakly sinusoidal or ‘loosely spiraling’ (e.g., Finlayson 1935 in Johnson, 1989; Koch, 1978; Coelho et al., 2000; Kinlaw and Grasmueck, 2012). Although there is likely a continuum of sinuosity, we found a somewhat dichotomous grouping of burrows that were slightly curved vs. repeatedly or regularly spiral or helical. We thus only included species with burrows described as ‘tortuous’ or ‘possessing regularly descending spiral coils,’ or a ‘helix’ (e.g., Powell, 1977; Koch, 1978). Although the behavior of constructing weakly helical burrows could be important in understanding the evolution of helical burrowing behavior (indeed Urodacus scorpions do both), the number of taxa exhibiting weakly spiral burrowing are too numerous to consider here. Moreover, the degree of burrow sinuosity has not been quantified for most taxa, making interspecific comparisons difficult. We included papers on both extant and extinct helical burrows produced by animals, although the producers of helical burrows—-trace fossils reported as ichnotaxa or described in open nomenclature—-in the fossil record are often unknown. For trace fossils, several tracemakers from different species as well as different phyla can produce a similar type of trace fossil morphology (ichnotaxon), for example Daimonelix(Raisanen and Hasiotis, 2018). Thus, our data rows in Table 1 often reflected more than one species. We also do not consider burrows that occurred in one horizontal plane, including sinusoidal traces (e.g.,Sinusichnus ; Belaústegui et al., 2014; Soares et al., 2020) and the spiral feeding traces that occur in soft sediment in one surface plane (e.g., the polychaete Paraonis fulgens ; Risk and Tunnicliffe, 1978).
We compiled hypotheses offered for the function of the helix, against which we could assign the likelihood that the hypothesis fit a particular taxon or ichnotaxon. For this we used ‘possible’, ‘not likely’ or ‘n/a’ (not applicable); ‘n/a’ indicated that there was virtually no chance of a fit, based on deductive reasoning. For example, the hatchling escape hypothesis developed for lizards, which proposes that the helix loosens the soil to facilitate hatchlings escaping the burrow through meters of resistant soil, would not be applicable to aquatic species with open burrows. We subjectively assigned ‘not likely’ if a good fit was unlikely, and we explained our reasoning in the text. Unlike with ‘n/a’, ‘not likely’ could change with the addition of new information or if our reasoning or that of others, was less informed. The assignment of ‘possible’ indicated a good fit or potential fit, based on the available evidence, context, and our reasoning or that of other authors. In many cases, however, designation of ‘possible’ reflects that difficulty in testing hypotheses—-for example, directly testing the antipredator hypothesis for most extinct species is not possible. Thus, we could not conclusively claim that a hypothesis was general even if it scored ‘100% possible’ for all taxa. However, by eliminating some taxa for each hypothesis (by assigning ‘n/a or ‘not likely’) we could potentially conclude that some or all of the hypotheses could be general for all taxa.
Results
Table 1 shows the results of the fit of hypotheses for the function of helical burrows to 21 taxa representing 59–184 species. The wide range in the potential number of species reflects both our lack of knowledge of the burrow types in extant and extinct conspecifics and the uncertainty of the species richness of ichnotaxa.
Of the 10 hypotheses, six are post-construction hypotheses and four are construction hypotheses. Of the 21 taxa, 12 (57%) are extant, eight (38%) are ichnotaxa and one includes both (5%; Table 1).
Two of the hypotheses, antipredator and biomechanical advantage, were designated as ‘possible’ for all taxa (score of 100%; Table 1). Other high-scoring hypotheses were the anticrowding (95%), vertical patch (95%), falling sediment (soil) (95%), and microclimate buffer hypotheses (86%). Two hypotheses, deposit feeding and increase drainage, received moderate scores (both 48%), mainly due to the fit of each to only terrestrial or only aquatic animals (Table 1). The remaining two hypotheses, microbial farming and offspring escape, received low scores (24% and 5%, respectively).
Although the hypothesis sample sizes precluded statistical comparison, the mean score for construction hypotheses (91.5 ± 2.99% SE; N=4) was higher than the mean score for post-construction hypotheses (49.3 ± 14.06% SE; N=6).
Discussion
Our review revealed that six of the 10 hypotheses for why animal construct helical burrows could not be confidently rejected for most of the taxa (86–100% possible; Table 1). These hypotheses range from ‘indirectly testable’ (falling sediment, vertical patch, biomechanical) to ‘extremely difficult or impossible to test’ (antipredator, anticrowding). Interestingly, four of those six hypotheses are ‘construction’ hypotheses, raising the possibility that helical burrowing could save on energy costs associated with constructing a helix without implicating post-construction adaptive benefits. Our analysis also eliminated some hypotheses as general explanations for the behavior; hypotheses involving increased drainage, deposit feeding, microbial farming, and hatchling escape could not explain helical burrowing behavior in the majority of the animals (5–48%, Table 1). The function and evolution of the helix as an extended phenotype remain unknown but would seem, in some cases, to provide different advantages for different taxa. In the following sections we discuss the fit of each hypothesis to selected taxa.