Videos used in creating ethograms for Callistoctopus furvus

Callistoctopus furvus (the “eastern”or “sand octopus”) is a relatively unstudied species of medium-large nocturnal octopus inhabiting shallow coastal areas of the tropical western Atlantic. To aid in field identification, promote further study of this species and enhance the understanding of skin patterning and behaviour in nocturnal shallow-water octopuses, we composed an ethogram of C. furvus from photographs and 3.11 h of video collected of wild individuals in the natural environment. We catalogued two colours, 17 static chromatic components, six groups of skin patterns (divided into 16 subpatterns), three textures, 10 postures and 11 behaviours. Photographs and videos of C. furvus from other geographic areas could be classified using this ethogram, supporting its validity for the study of C. furvus throughout its geographic range. We also described eight features that can be used to differentiate C. furvus from co-occurring octopods through observation alone: strict nocturnality, brick red incorporated into most skin patterns and rows of regularly spaced white spots that are always at least faintly visible on the arms, a polka-dotted deimatic pattern, large “U”-shaped papillae on the mantle, six distinct mantle shapes, a shallow arm web, the execution of very few “parachute attacks” and a tendency to enter existing holes or bury itself in response to disturbance. Notable features include the use of at least one “dynamic”chromatic component, a putative masquerade as algae, and potential mimicry of trumpetfish or cornetfish. Greater observation and study of C. furvus and other octopods from a diversity of habitats, latitudes and temporal niches will enhance our understanding of how these factors have shaped the use of colours body patterning and behaviour of this unique group of molluscs.

Methods
Study site
This study was conducted at 10 sites less than 1.5 km offshore of the southern and southwestern coasts of South Caicos (21.7 ◦N, 71.8 ◦W) in the Turks and Caicos Islands. These islands lie at the southeastern end of the Lucayan archipelago on shallow (< 20 m) carbonate banks. Salinity ranges from 36.4 to 36.5 ppt (Jury, 2013) and sea surface temperatures range from 22◦C to 28◦C according to season (Logan & Sealey, 2013). The predominant substrates at these sites are patch reefs, limestone bedrock, sandy plains and seagrass (O’Brien et al., 2023).

Video collection
Video was collected sporadically between 28 February 2020 and 4 May 2023 around South Caicos by teams of two to six snorkelers or five to nine divers. Between 19:30 and 23:30 h, these teams travelled by boat to dive at sites offshore (3–18 m) or snorkelled from shore (0–4 m). They searched seagrass, sand, algal plain, patch reefs and bedrock haphazardly for both Octopus briareus and Callistoctopus furvus. When one was located, the nearest observer to the octopus began recording using a Nikon CoolPix W300 (resolution 1080 ×1920 res), GoPro Hero 8 (1080 ×1920 res), SeaLife DC1400 (1280 ×720 res) or SeaLife Micro 3.0 camera (1080 ×1920 res), illuminated by the 25% setting of a 2000–3000 lumen SeaDragon photo/video light (2000F, 2500F or 3000F). The videographer conducted a continuous focal-follow and recording of the octopus until it disappeared from view, the camera or dive light battery died or time and/or air constraints forced the end of the excursion. Over the course of 4 years, a total of 3.11 h of footage of 27–35 individuals estimated to be between 7 and 15 cm mantle length (ML) was collected. Since octopuses were not measured or tagged to avoid interfering with natural behaviour, estimations of the number of distinct individuals sampled are based on identifying features such as wounds, relative size (based on visual comparison with objects in the surrounding environment) and the location of sighting. Some individuals were filmed once; others were filmed multiple times over a single or multiple evenings. Videos lasted for a median [interquartile range (IQR)] of 51 s (89 s), with the minimum time that any particular individual was filmed being 5 s and the maximum being 7 min 11 s.

Video analysis
To derive an ethogram of C. furvus, we utilized the framework and terminology of Packard & Sanders (1971) and Packard & Hochberg (1977), and synthesized by Mather & Alupay (2016) and Hanlon & Messenger (2018), summarized in Figure 1 below. Briefly, the overall appearance, termed “body pattern”, is composed of four levels of constituent parts that graduate in size: elements combine to compose units, units make up components and the combination of components on all body regions is what creates the overall body pattern. The constituent parts include aspects of body texture, posture, behaviour and the overall arrangement of colour and iridescence of the skin, hereafter referred to as “skin pattern”to distinguish it from “body pattern”. The first author reviewed all footage (3 h and 6.60 min) three or more times and isolated each unique skin pattern, texture, posture and behaviour expr essed until no new ones could be isolated. Skin patterns were then described and named, and screenshots of each (three or more for each pattern) were collected and used to identify, name and describe the individual colours and chromatic components that composed them. Textures, postures and behaviours were also described and named. Identification of textural, postural and locomotor components was not attempted due to difficulty in dis- cerning such fine details under the uncontrolled nocturnal filming conditions and the frequent obstruction of body regions by other body parts. Whenever possible, terminology from previous octopus behavioural studies (e.g. Hanlon & Wolterding, 1989; Leite & Mather, 2008; Mather & Alupay, 2016) was used. Where no precedent existed, we devised our own names based on colour and shape or resemblance to known objects or patterns. According to standard convention, chromatic components were written with the first letter of the first word capitalized, while skin pattern, texture, posture and behaviour names were indicated by capitalizing the first letter of each word.

Octopuses express chromatic components, skin patterns and textures on a continuum of intensity, from small and/or barely visible to large and/or highly contrasting, and this intensity of expression can vary across different body parts (Packard & Hochberg, 1977). To account for this, we have listed the chromatic components observed on eight regions of the body that were noted to differ from contiguous regions: the dorsal mantle, funnel, head, arms bases (the area between the head and the arms where the arms diver ge from the rest of the body), arms and web, eye bulbs, eye and ventral arm area. Because it was rarely visible, we did not attempt to categorize the ventral mantle. Most chromatic components are “static”, meaning that they consist of relatively fixed displays of colours and patterns that are displayed for seconds to minutes, but some can be “dynamic”, incorporating flashing or strobing through the coordinated action of chromatophores, producing expressions that occur over fractions of a second (How et al., 2017). To validate the ethogram, the second author independently analysed the same video sequences to confirm that the appearance of the octopus could be classified according to the categories established by the first author in all of the footage. Videos were analysed second by second in this manner from the first second (0) to the last second of the video, resulting in a total of 11,201 frames. Any frame that did not match an existing description was described and added to the ethogram.

Collection and analysis of internet media
In order to determine if the chromatic components, skin patterns, textures, postures and behaviours identified in our examina tion occur species-wide or are unique to the Turks and Caicos population, both authors independently classified components, skin patterns, textures, postures and behaviours in photographs and videos of C. furvus sourced from the internet. Photographs were collected via several means. Eleven photos were sent to us by colleagues from Brazil (J. Voix and Dr M. de Jesus), while eight others were published in Jesus et al. (2021, 2022). Ten videos taken in South Florida and the Virgin Islands (2021–2024) were obtained with permission from photographer Dr R. Fonseca. Twelve photographs taken around Lee Stocking Island in the Bahamas (2009) were obtained with permission from photographer Dr P. Lyons. Forty-five images were found using Google image search (accessed 2 October 2020 and 27 October 2023) with the search terms “Callistoctopus furvus” “Callistoctopus macropus”, “Octopus macropus” (C. furvus is commonly misidentified with these names) and “white-spotted octopus” and using images that were labelled as from the western Atlantic. We also procured 14 photographs from the wildlife observation website iNaturalist (3 October 2023) using the search term “Callistoctopus macropus” (“Callistoctopus furvus” redirects to “Callistoctopus macropus”) and selecting all of the photographs taken in the western Atlantic. A single video was found on YouTube (2 November 2023) using the search terms “Callistoctopus furvus” “Callistoctopus macropus”, “Octopus macropus” and “white spotted octopus” and selecting any recorded in the western Atlantic. We also solicited media via FASTMOLL, an email listserv for cephalopod researchers, and received six photos and four videos (02:10) from Dr P. Sanchez from the Gulf of Mexico.

Ethical note
Permission to conduct these observations was granted by the Turks and Caicos Department of Environment and Coastal Resources (DECR; permits D2-001,2023-03-30-17 and 2024-02-13-05). Our minimally invasive method (filming) limited the impact of the study on the animals involved by recording individuals in their natural environment and never capturing or handling them. Our presence and lights disturbed the surrounding wildlife to some degree, but the fact that we never spent longer than an hour at any site limited this impact.