Early ant trajectories: spatial behaviour before behaviourism

Abstract

In the beginning of the twentieth century, when Jacques Loeb’s and John Watson’s mechanistic view of life started to dominate animal physiology and behavioural biology, several scientists with different academic backgrounds got engaged in studying the wayfinding behaviour of ants. Largely unaffected by the scientific spirit of the time, they worked independently of each other in different countries: in Algeria, Tunisia, Spain, Switzerland and the United States of America. In the current literature on spatial cognition these early ant researchers—Victor Cornetz, Felix Santschi, Charles Turner and Rudolf Brun—are barely mentioned. Moreover, it is virtually unknown that the great neuroanatomist Santiago Ramón y Cajal had also worked on spatial orientation in ants. This general neglect is certainly due to the fact that nearly all these ant researchers were scientific loners, who did their idiosyncratic investigations outside the realm of comparative physiology, neurobiology and the behavioural sciences of the time, and published their results in French, German, and Spanish at rather inaccessible places. Even though one might argue that much of their work resulted in mainly anecdotal evidence, the conceptual approaches of these early ant researchers preempt much of the present-day discussions on spatial representation in animals.

Notes

¹ For a biographical account written on the occasion of the 50th anniversary of Santschi’s death, see Wehner (1990).

² At the turn of the nineteenth to the twentieth century it was a hotly discussed issue, whether trail laying ants could detect the inbound and outbound direction of a pheromone trail on the basis of the chemical polarity of the trail itself, or even of the individual footprint marks left by the marching ants, or whether they were unable to do so. Albrecht J. T. Bethe, who then was assistant at the Physiological Institute of the University of Strassburg, where a decade earlier Jacques Loeb had developed his tropism concept of animal orientation (Loeb 1890), proposed the former hypothesis (Bethe 1898, 1900, 1902). Adele Marion Fielde, an American missionary and feminist, who later in her life became a scientist, followed Bethe in this respect (Fielde 1901). Erich Wasmann, a Jesuit priest and famous entomologist, favoured the latter hypothesis (Wasmann 1898, 1901). Auguste Forel argued against either version and instead pointed out that mechanosensory information had to be taken into account as well (and in this context introduced his somewhat vague concept of a ‘topochemical sense’: Forel 1902, 1903). For an extensive and critical discussion of Forel’s topochemical sense, see Brun (1916b). Felix Santschi later discovered by careful observation of walking ants that chemical trails did not consist of passively applied chemical footprint marks, as had previously been assumed, but by actively deposited secretions of anal glands (Santschi 1913a). Almost four decades later and unaware of Santschi’s early work, John Carthy from Cambridge University ‘rediscovered’ this phenomenon (Carthy 1950). Finally, two more decades had to pass until the first trail pheromone of ants was chemically identified (in leafcutter ants of the genus Atta; Tomlinson et al. 1971).

³ Rosengren (1971), Harrison et al. (1989) and Jackson et al. (2004).

⁴ Victor Cornetz was a prolific writer. His first observations were published in the very same year in which he had started to work on ants (Cornetz 1909b). In the following year he wrote already an extensive monograph accompanied by an Album and lengthy handwritten notes (Cornetz 1910a, b, c).

⁵ Henri Louis Charles Piéron (1881–1964) received his M.Sc. degree in philosophy and his Ph.D. degree in physiology, before he entered the field of psychology and founded, in 1920, the Institute of Psychology at the University of Paris. His work concentrated on psychophysics and comparative psychology. Much admired by his colleagues he was extremely well read as well as an active writer. While he was still working on the orientation of ants (Piéron 1912), he published his first book in psychology (Piéron 1910). Surprisingly, in his biographical essay about Piéron, Jean Piaget did not mention Piéron’s myrmecological studies at all (Piaget 1966). For the classical account on Piéron’s pedometer hypothesis, see Piéron (1904).

⁶ For leg-manipulation studies, see Wittlinger et al. (2006). Cheung et al. provided mathematically rigorous analyses of how much noise various path integration schemes can tolerate; e.g. Cheung and Vickerstaff (2010) and Cheung (2014).

⁷ Cornetz interchangeably used a number of terms depicting the marked directedness of foraging journeys, e.g. ligne d’equilibre d’aller (Cornetz 1910b), ligne de reference (Cornetz 1911a), l’axe de sinuement (Cornetz 1910d). For the term espace de recherche, see e.g. Cornetz (1910d). Since the 1980s short-term sector fidelity has been recorded in many desert ant species. For recent examples, see Muser et al. (2005) and Sommer et al. (2013).

⁸ The first displacement experiments, in which ants selected homeward courses parallel to their pre-displacement courses, and ran for distances equivalent to their pre-displacement distances, were performed by Piéron (1904). Santschi dubbed this kind of orientation l’orientation virtuelle (Santschi 1913a). By using the terms conservation de l’orientation and la règle de constance Cornetz referred to the observation that in successive foraging runs the ants tended to maintain a rather constant direction (Cornetz 1910d, 1911b). He argued that this direction was determined by a completely internal sense of direction (un sens de perception des directions absolues dans l’espace; Cornetz 1914a).

⁹ In an early letter to Cornetz (dated 27 December 1910) Santschi already wondered about the faculté de direction, which Cornetz had frankly attributed to the ants, and asked what sensory system might be involved. In his later publications he presented clear evidence against Cornetz’ mysterious internal sense of direction, and instead described several ways of navigation based on a variety of external cues. His two most comprehensive accounts can be considered masterpieces of argumentation and descriptive narrative style (Santschi 1913a, 1923b).

¹⁰ The first experiments on light compass behaviour in ants have been performed in the laboratory by John Lubbock in Lasius (Lubbock 1881), by Hugo Viehmeyer in Leptothorax (Viehmeyer 1900) and by Charles Turner in Tapinoma (Turner 1907). Santschi’s classical 1911 paper describes the discovery of the sun compass in ants, and in fact in any animal species (Santschi 1911).

¹¹ Santschi (1911, 1913a). Based on considerations made by Santschi (1911, p. 324) Cornetz coined the term ‘compass eyes’ (les yeux boussole: Cornetz 1911a).

¹² Ramón y Cajal (1921); see also notes 43–46 and the special reference to Santiago Ramón y Cajal’s experiments on the orientation of ants further below in the text.

¹³ For Cornetz’ argument about changing ambient light distributions, see Cornetz (1912, 1913a). The difficulty to account for the daily changes in solar azimuth is mentioned in Cornetz (1913b, c).

¹⁴ Santschi published a short note on his cardboard-cylinder experiments in the very year in which he had performed them (Santschi 1914, appendix), but a detailed description followed only several years later (Santschi 1923b). There he interpreted his observation in a number of ways—for details, see Wehner (1990)—but finally was left with a riddle (Santschi 1923b, p. 159). Karl von Frisch remarked in a footnote that he had learned about Santschi’s crucial experiments only after he had performed his own ones (von Frisch 1949).

¹⁵ The paper that contains Santschi’s witty remark on time compensation of the sun compass makes wonderful reading and should be a must for anyone interested in animal navigation (Santschi 1913a). For Santschi’s rejection of Cornetz’ argument by hinting at the predictability of the sun’s daily movement, see also Santschi (1913b). Final proof came nearly simultaneously from experiments in bees and birds (von Frisch 1950; Kramer and von Saint Paul 1950). Before Karl von Frisch und Gustav Kramer had performed their critical experiments, the hypothesis of time compensation was erroneously refuted by Brun (1914, pp. 176–184) and Wolf (1927).

¹⁶ The travel during which Felix Santschi accompanied Auguste Forel to South America is described at length in Forel (1935, pp. 170–187).

¹⁷ Even though Santschi fully admired Auguste Forel and adopted nearly all of Forel’s scientific, philosophical and social arguments, his taxonomic concepts show that he never became a genuine evolutionary biologist, and in this respect did not follow his master.

¹⁸ Provided with letters of recommendation from the Swiss Ambassador in Paris and the French Minister of War to the commander of the French troops in Tunisia, Général Leclerc, Cornetz travelled at his own risk in the most southern parts of the country well beyond the French military posts in Medenine and Metameur. He stayed with the Berber tribes in the plains and sand-dune fields as well as in the rugged mountain area of the Djebel Maouaia around Guermessa and Douiret. The only geographic literature he had at hand was Largeau (1877) and Rolland (1890). His narrative report contains information not only about the physical geography and topography, but also about the distribution and life styles of the various Bedouin groups of the area (Cornetz 1897).

¹⁹ Until 1929 Georges Bohn’s La Naissance de l’Intelligence (The Emergence of Intelligence; Bohn 1909) was reprinted several times in French, translated into German (1910), but unfortunately never into English. For other historically interesting accounts on that subject, see Romanes (1882), Emery (1893), Wasmann (1897, 1899, 1909), von Buttel-Reepen (1900) and Lukas (1905). Much of the literature published mainly in the two decades around 1900 is discussed by Margaret Floy Washburn, then a professor of philosophy at the well renowned Vassar College north of New York City (Washburn 1908). On the early history of animal psychology, see also Claparède (1909) and Ziegler (1920).

²⁰ Santschi (1913a) and Brun (1916a).

²¹ Cornetz (1909a). Recently, Erik Jonsson has referred to Victor Cornetz’ early observations in his book “Inner Navigation”: Jonsson (2002, pp. 111–116): “Adari Wayfinding in the Sahara”. To the best of my knowledge this is the only time since Rudolf Brun’s and Santiago Ramón y Cajal’s early references that a scientist has dwelled on Victor Cornetz’ work.

²² de Gourmont (1909, pp. 153–158).

²³ Victor Cornetz wrote his final book in response to Pierre Jaccard’s elaborate account on long-distance orientation in humans (Jaccard 1932; Cornetz 1933). See also Cornetz (1909a)

²⁴ The two appraisals of Santschi are from Cornetz (1911a, cit. p. 234; 1914b, cit. p. 9). For the comparison of Santschi with Forel and Wheeler, see Cornetz (1925, cit. p. 139).

²⁵ Loeb (1890). This book already foreshadowed Loeb’s later attempts to consider organisms from a strictly physicochemical viewpoint. For the reference to Albrecht Bethe, see note 2.

²⁶ Loeb (1901, 1912, 1918), Rádl (1903) and Crozier (1929, cit. p. 48).

²⁷ Kühn (1919), Fraenkel (1931), Köhler (1931), Fraenkel and Gunn (1961) and Jander (1963, 1970).

²⁸ References to Jacques Loeb: Cornetz (1911a, p. 236) and Santschi (1913a, p. 384).

²⁹ For the concepts of menotaxis and mnemotaxis, see Kühn ( 1919, pp. 35, 51); an extended treatment has been provided by Fraenkel and Gunn (1961, pp. 106, 312).

³⁰ For Loeb’s mechanistic conception of life, see Pauly (1987). We may add that at present the tide has turned completely. With the ‘cognitive turn’ in full swing, animals are regarded as autonomous agents. That sensory stimuli directly trigger behaviour—Loeb’s central tenet—is now considered the exception rather than the rule, and is thought to occur mainly in emergency cases. The role of external stimuli is relegated to guiding (orienting) ongoing behaviour, and to contributing to the search process within the repertoire of behavioural modules, which the brain—the behavioural organizer—has at its disposal. This view, which Cornetz certainly would have acknowledged, is most forcefully expressed by Heisenberg (1983, 2013). For the application of this way of thinking to homing in honeybees, see Menzel (2014).

³¹ The small-scale search movements are described in Turner’s Ph.D. publication (Turner 1907).

³² Some exemplary references to the Turner loops: Cornetz (1910a), Piéron (1912) and Santschi (1913a).

³³ Wehner and Srinivasan (1981), Müller and Wehner (1994) and Schultheiss and Cheng (2011). The search loops performed around frequently visited feeding sites exhibit similar structural properties to the ones around the ants’ final goal, the nest entrance (Bolek et al. 2012, Schultheiss and Cheng 2013).

³⁴ Watson ( 1907, cit. p. 300) and Rau ( 1923, cit. p. 290).

³⁵ Du Bois ( 1938, cit. p. 309).

³⁶ Du Bois’ (1929) empathic “overwork” comment is certainly based on the generally acknowledged fact that Turner was an indefatigable worker during all his life, who had to perform his research under the most adverse circumstances. Turner retired from Sumner High School in St. Louis because of illness and died one year later in the house of his son in Chicago from myocarditis (Abramson et al. 2003).

³⁷ For biographical accounts on Charles Henry Turner, see Kessler et al. ( 1996 ), Dewsbury ( 2003 ) and Abramson ( 2009 ).

³⁸ Warden and Warner (1927) and Cadwallader (1984). At the turn of the nineteenth to twentieth century the University of Chicago, where Turner did his dissertation work, was among the first three academic institutions in the United States (besides Clark and Harvard), at which departments of animal psychology were established.

³⁹ Colour vision: Turner (1910) and von Frisch (1914; for a remark on Turner’s work, see pp. 79–80). Pattern recognition: Turner (1911). In his paper on colour vision in bees Turner discussed the question of whether brightness differences could have played a role, and presented some indirect evidence that the hue of colour must have been the decisive cue, but he did not perform the finally decisive control experiments.

⁴⁰ Turner (1908a, b; 1910, cit. p. 277; 1923). For recent work on visual spatial homing in ants, see e.g. Wehner and Räber (1979), Wehner et al. (1996) and Wystrach et al. (2011, 2013). Modelling approaches range from the first formulation of the ‘snapshot model’ (Cartwright and Collett 1983) via concepts of ‘image difference functions’ (Zeil et al. 2003) to attempts of mapping the entire navigational information content of natural environments (Stürzl et al. 2015).

⁴¹ Studies on cockroaches (Turner 1912) and hymenopterans (see notes 39, 40). William Morton Wheeler, who several times referred to Turner’s work (see note 42), fully agreed with Turner’s view on insect behaviour. In fact, Wheeler had “intense and heated discussions” with Jacques Loeb, when in the 1890s both were colleagues at the University of Chicago (Parker 1938, cit. p. 221). In his studies on hearing in saturniid and erebid moths Turner emphasized the functional “significance” of a stimulus necessary to elicit an animal’s response (Turner 1914; Turner and Schwarz 1914). Actually, Charles Turner—together with and independently of the Slovenian zoologist Ivan Janez (Johann) Regen, who performed the famous ‘telephone experiment’ in crickets (Regen 1913)—provided the first conclusive behavioural evidence of hearing in insects.

⁴² Turner’s neuroanatomical work: Turner (1891, 1899); see Wheeler (1928, cit. p. 178). It is not clear why Wheeler called Turner one of his students. Wheeler had left the University of Chicago in 1899 for the University of Texas in Austin, and when Turner later started his studies at the University of Chicago in 1906, Wheeler was Curator of Invertebrate Zoology at the American Museum of Natural History in New York (Parker 1938; Cadwallader 1984; Abramson 2009).

⁴³ The pioneering accounts of Santiago Ramón y Cajal and his major collaborator Domingo Sánchez y Sánchez are timeless masterpieces in vertebrate and invertebrate neuroanatomy (Ramón y Cajal 1899–1904; Ramón y Cajal and Sánchez y Sánchez 1915). At this juncture, I invite the reader to consult Nicholas Strausfeld’s brilliant account on the structure, evolution and research history of arthropod brains. In this scholarly and aesthetically appealing piece of art Nick retrieves many of the neuroanatomical treasures created by the great Spanish scientists (Strausfeld 2012).

⁴⁴ The comments are from Santiago Ramon y Cajal’s autobiography (Ramón y Cajal 1937, cit. p. 485). For a published account on Forel’s studies, which Ramón y Cajal later read, see Forel (1886).

⁴⁵ Ramón y Cajal (1937, cit. pp. 505 and 506).

⁴⁶ Santiago Ramón y Cajal summarized his “preliminary” data on spatial vision and orientation in ants in only one publication (Ramón y Cajal 1921). For a recent survey, see Lopera Chaves and Freire Mallo (2009).

⁴⁷ Ramón y Cajal and Sánchez y Sánchez (1915); Ramón y Cajal S (1918). In contrast to the adherents of Loeb’s and Watson’s mechanistic view of behaviour the contemporaneous researchers on ant navigation expressed a strong interest in the structure of insect brains, and—with the exception of Victor Cornetz—they all performed some neuroanatomical studies. The results of this work have remained largely unpublished. In this regard, it is the more regrettable that Santiago Ramón y Cajal never tried his hands at ants. The first one who did so and then succeeded beautifully was Wolfgang Goll, whom his PhD supervisor Otto Pflugfelder had introduced into using the Golgi method (Pflugfelder 1937; Goll 1967). Independently of the two researches at Stuttgart-Hohenheim, the Golgi method was independently ‘rediscovered’ for staining individual neurons in insects by Nicholas Strausfeld, then a graduate student at University College London (Strausfeld 2012 and Fig. 3.15 therein).

⁴⁸ Reference to memory: Ramón y Cajal (1937, cit. p. 589). Wilder Penfield’s quotation is from the foreword to Ramón y Cajal’s last monograph, in which one year before his death Ramón y Cajal provided a brilliant summa summarum of his arguments in favour of the neuron theory, or “neuron doctrine”; see Ramón y Cajal (1933).

⁴⁹ Among biologists Rudolf Brun had been almost forgotten already at the time of his death (Kutter 1970), and in today’s community of psychologists he is completely unknown. Even though Brun was an outspoken Semonian (see notes 51, 52), Daniel Schacter does not mention him in his comprehensive account on Richard Semon (Schacter 2001). For a local appreciation of Brun’s activities as a neurologist, see Minkowski and Müller (1945).

⁵⁰ Rudolf Brun had drawn most of his conclusions from Cornetz’ and Santschi’s work and had described and discussed his own work already in an early monograph (Brun 1914). By later adding the results of new and more systematically performed experiments he refined, but did not revise his concepts expounded in Die Raumorientierung der Ameisen. For his later work, see Brun (1916b, 1925a, 1932). Moreover, two review papers bear witness to Brun’s comprehensive knowledge of the main topics discussed in his days in ‘animal psychology’ (Brun 1917, 1920).

⁵¹ Richard Semon defined a ‘mneme’ in general biological terms as “the enduring though primarily latent modification in the irritable substance produced by a stimulus” (Semon 1904; English edition, cit. p. 12), as some kind of “cell or organic memory”. In Richard Dawkin’s ‘meme’ this unit of memory trace has turned into the cultural equivalent of a gene, into a unit carrying ideas, symbols or practices that can be transmitted culturally from one mind to another, and can self-replicate, mutate, and respond to selective pressures (Dawkins 1957). Surprisingly, when Dawkins introduced the term meme, he did not mention Semon’s mneme at all. The influential geneticist Wilhelm Johannsen, who had coined the term ‘gene’, read Semon’s book with interest, but concluded that even though the author had proposed some “ingenious speculations” Semon had drawn a false analogy between heredity and memory, between genes and mnemes (Johannsen 1913, cit. p. 423). Semon’s concept of the mneme was much appraised by Auguste Forel (Forel 1905). Alfred Kühn adhered to it—‘mnemotaxis’ (Kühn 1919)—but in modern memory research it has scarcely left any trace (Schacter 2001). Etymologically, the term mneme is derived from μνήμη (mnēmē), Greek for memory. Mnemosyne is the Greek goddess of memory.

⁵² Brun (1914, cit. p. 17). In order to appreciate this statement one has to take into account that Brun strictly adhered to Semon’s mnemic (psychological) terminology and described all results of his studies on spatial learning in ants in the framework of Semon’s concepts (Brun 1914, pp. 25–28).

⁵³ Brun (1942). How Brun after having overcome some initial hesitation became a Freudian, and how he tried to reconcile psychoanalysis with his kind of biological thinking, is described in Brun (1926, 1956). See also Aeschlimann (1980).

⁵⁴ For Rudolf Brun’s neuroanatomical studies on some insect species including ants, see Brun (1923, 1925b).

⁵⁵ Two papers in The New Visual Neurosciences might suffice to provide an impression of the current discussion on the structure of the insect’s navigational toolkit (Menzel 2014; Wehner et al. 2014). The main views are expressed most concisely in a debate revolving around the interpretation of particular experiments in visual navigation of honeybees (Cheeseman et al. 2014 and rebuttals).

⁵⁶ Szymanski (1911), Bohn (1912) and Santschi (1913a).

⁵⁷ For detailed studies on path integration in ants, see e.g. Müller and Wehner (1988), Collett and Collett (2000) and Wehner et al. (2002); see also note 61.

⁵⁸ Cornetz (1929). This is Victor Cornetz’ last paper on spatial orientation of ants. There and in former papers he only occasionally addressed the question of what we now call path integration, but deliberately dismissed of this possibility. He concluded that the ants were incapable of trigonométrie automatique and instead behaved according to la règle du contre-pied (Cornetz 1914a, p. 118).

⁵⁹ As can be deduced from his published records, Rudolf Brun had clear evidence that Formica and Polyergus foragers referred to distant landmark cues, when they “closed the polygon” (Brun 1914, pp. 195–196, 215; Brun 1916b, pp. 300–302). Nearly a century had to pass until systematic studies showed that in cluttered, landmark-rich environments—similar to those in which Brun had performed his experiments—landmark guidance could almost completely override path integration, so that under these conditions the latter cannot be properly studied (Fukushi 2001; Beugnon et al. 2005; Narendra et al. 2013).

⁶⁰ Brun (1914, cit. p. 192). The contemporaneous neglect of path integration as a possible means of navigation is the more astounding as in marine navigation ‘dead reckoning’ had been used for centuries (Taylor 1950), and as in an early study on pigeon homing Édouard Claparède, Professor of Psychology at the University of Geneva, had already alluded to path integration in describing a bird’s ability to directly return to the home loft after displacement along a twisted detour route (Claparède 1903). For even earlier, though vague discussions of whether homing pigeons could use information obtained by route integration during their outbound journeys, see Darwin (1873), Exner (1883) and Reynaud (1898).

⁶¹ In research on animal orientation the concept of path integration was first formalized by Horst and Marie-Luise Mittelstaedt (Mittelstaedt 1963; Mittelstaedt and Mittelstaedt 1973). As this kind of navigation results in the computation of a goal vector, it has also been treated as ‘vector navigation’ (Wehner 1982; Collett and Collett 2000). Full theoretical analyses of path integration have recently been provided by Allen Cheung and Robert Vickerstaff, see e.g., Cheung and Vickerstaff (2010) and Vickerstaff and Cheung (2010). In forced detour paradigms similar to the one shown in Fig. 9, Cataglyphis ants always update their home vector correctly (Schmidt et al. 1992; Wehner 2003 and Fig. 2b therein).

⁶² Thorndike (1911), Watson (1913), Pavlov (1928) and Skinner (1938).

⁶³ Bulletin de la Société Entomologique Suisse 18:286–289 (1940). Entomologist’s Record and Journal of Variation 53:56, 99–100 (1941). Arbeiten über morphologische und taxonomische Entomologie aus Berlin-Dahlem 8:6–7 (1941). Revista de la Sociedad Entomologica Argentina 13:344 (1947). For a biographical account on Felix Santschi, see note ¹.

⁶⁴ L’Année Psychologique 38:979 (1937).