Reproduction of the pancake tortoise in the wild has remained poorly known over the decades. This study fills the knowledge gap by investigating the reproductive ecology of the species in its natural habitats in Kenya. Data were collected using the time-constrained search-and-seize method, group-level scan sampling, radio tracking, and camera-trapping. Our observations revealed a total of 19 mating events that occurred between December and April, predominantly during the rainy seasons. Mating duration averaged 15.2±2.4 minutes. Male-male aggression was observed during the mating season, with the dominant male keeping off all other males and mating with the resident females. Nest preparation events occurred between April and June. Nine nesting events were recorded, but in only five of them was an egg deposited. The nests were dug in loose soil, appeared almost circular, and measured 8.4±0.9 cm wide and 7.0±0.1 cm deep (n=7). The average clutch size was one egg. The eggs were white, hard-shelled, oval, and elongated measuring 4.4±0.4 cm long, 2.7±0.04 cm wide and weighed 17±0.6 g (n=3). Incubation period lasted 177± 5.7 days (n=2), and hatching coincided with the onset of the short rainy season in November, aligned with the availability of abundant food for the tortoises. The hatchlings were minimally wider than long, having an average straight carapace length of 4.5±0.6 cm, an average width of 4.5±0.4 cm, and on average weighed 14.7±3.7g, (N=9). They remained near the nest site for 4-5 days before relocating to tiny rock crevices away from the adult tortoise crevices. Egg and hatchling predation was high, with four of the total of six eggs, observed after they were laid, being destroyed. Three juveniles, out of nine, were also eaten by predators. These findings contribute useful information for the formulation of effective conservation and management strategies for this critically endangered species.

Study area. — The study was carried out in Kitui County, Kenya, between February 2020 and November 2022. The areas surveyed occur within latitudes 00^o - 01^o south and longitudes 038^o - 039^o east and at an elevation ranging from 510 -775 m above sea level. In this region, populations of pancake tortoise are distributed discontinuously in a north-south orientation following the Precambrian basement system that has characteristic exfoliating granitic rock outcrops and kopjes with suitable crevices (Malonza 2003, Spawls et al. 2018). Soils here are generally sandy. The vegetation type of the region is mainly Acacia-Commiphora bushland, dominated by Commiphora, Acacia, Grewia, and Combretum (Wood and Mackay1997). The area experiences two short-lived rainy seasons between October and December, and March and June. Rains are erratic, and thus this region experiences frequent droughts. Rivers and streams are seasonal and usually exist as dry sandy river beds. Human residents often practice subsistence cultivation and pastoralism.

Time-constrained search-and-seize method. — Between February and July 2020, rock crevices inhabited by pancake tortoises were identified through the time-constrained search and seize method as described by Karns (1986) and Sutherland (1996). Whenever possible, the tortoises therein were extracted (using a hooked stick), counted, and their sexes determined. Females have short stumpy tails and males have a characteristic thicker and more elongated tails (Malonza 2003). Using a Vernier callipers, the straight carapace length (SCL), width (over the broadest part of the tortoise), and height (Ht) for each tortoise accessed were measured. Tortoises were also weighed and scute rings counted (SRC) in order to estimate their age (Germano 1994, Lagarde et al. 2001). Counting of the growth annuli was done mainly on the front costal scutes, which were less abraded.  During this time, any observation of reproductive behaviour was recorded.

Group-Level Scan Sampling. — From July 2020 to December 2022, group-level scan sampling (Van Belle 2017) was used to study the behaviour of pancake tortoises occupying 12 rock crevices (stations) in the Wingemei area in Nuu. In total, 39 tortoises were initially present in the crevices, and each of these was marked with a unique number on the carapace using a nontoxic-paint pen to enable us distinguish between different individuals. Monitoring visits were conducted regularly to the stations both during the dry and wet seasons, and on each visit the number of tortoises present in or near each crevice was recorded. Any breeding behaviour of the tortoises observed during each visit (i.e. pre-copulatory aggression, courting, mating, digging of nest, egg laying, and hatching) was recorded. The crevices were visited on a rotational basis, usually between 08:00 h -18:00 h. Thrice night visits to the crevices (between 19:00 h - 23:30 h) were done to observe the tortoise behaviour during these times. On two occasions, nesting females were also observed from evening into the night (16:00 - 21:30 h), using spotlights.

During the nesting period, when nest digging behaviour was observed, researchers maintained a safe distance [8.3±3.2 m (SD), dependent on the physical habitat configuration] from the tortoises to avoid distracting them. All the nests located had their depth, width, and distance from the adult tortoise rock crevice measured. The soils at the nest sites were dug out to know the soil type and compaction. The plants overshadowing each nesting site were identified, and an approximation of their vertical sunlight shielding done by laying crown cover grids (0.5 m x 0.5 m) at the base of the plants, then estimating the percentage of the total grid area receiving shadow when the sun was overhead.

Whenever an egg was laid, date and time were recorded. The soil covering it was carefully removed, the depth measured, and the orientation of the egg in the nest noted. Taking care not to break or turn it, the egg was carefully handled, and its length and breadth were measured using Vernier callipers. Its mass was weighed using a Pesola® (100g) micro-line 20100 balance (www.ecotone.pl). The colour and shape of the eggs were also noted. Finally, the egg was gently restored to its original position and orientation, and the soil and leaf litter, respectively, returned to cover the nest. The nests were geo-referenced and monitored (through daily visits) throughout the incubation period. After hatching, the hatchling length, width, and height were measured using Vernier calliper and its mass weighed using the Pesola® micro-line 20100 balance. Their colour, shape, and behaviour (e.g. where they hid, number of days they stayed at the nest site before finding a new crevice) were also recorded.

Radio tracking. — To enable us to closely follow and monitor the behaviour of individual pancake tortoises, 14 adult tortoises selected at random from different crevices in Wingemi area were fitted with 9g R1-2B radio transmitters (Holohil Systems Ltd, Ontario, Canada) and their movement was tracked using FCC ID: GZ383TTRX1000S radio receiver. The transmitters were fitted on the front costal (pleural) scute using clear hardened epoxy resin with the transmitter tail positioned slightly facing sideways in females to avoid obstruction during mating following guidelines in Goodlett et al. (1998). The radio-tracking process involved us visiting the tortoise crevices and observing if the radio-tagged tortoises were present. If not, then we switched on the radio-receiver and tracked the tortoise to its exact location. It was now possible to observe what the tortoise was doing, for instance, foraging, mating, nesting or perhaps if it shifted to a new crevice.

Camera-trapping. — Four E1C EREAGLE (7.45mm lens) camera traps were deployed close to four of the pancake tortoise inhabited rock crevices (selected at random) and set in continuous recording video mode to collect tortoise behaviour near the crevices (Agha et al. 2015). Cameras enabled us to collect data on the time when the pancake tortoises exited and returned to their crevices. In cases where there was mating within the detection field, the camera recordings helped us to understand the time/seasons and frequency of mating by the tortoises. Furthermore, the camera traps helped us to gather data on other behaviour of the tortoises (e.g., fighting) in a 24-hour cycle non-invasively. After the eggs were laid, two of the camera traps were transferred from the adult tortoise camera stations and placed near and overlooking two of the nests (in continuous recording video mode) to record any activity around the nests (especially by predators) and to maximize the chances of capturing the hatching events (date and time of hatching, weather during hatching, activity by the hatchling after hatching). One camera was set on a tree overhanging a nest on 16 May 2022, and another was supported using a wooden pole in a close-up view above the second nest on 3 June 2022.

References:

Agha, M., Augustine, B., Lovich, J. E., Delaney, D., Sinervo, B., Murphy, M. O., Ennen, J. R., Briggs, J. R., Cooper, R. and Price, S. J. 2015. Using motion-sensor camera technology to infer seasonal activity and thermal niche of the desert tortoise (Gopherus agassizii). – J. Therm. Biol. 49: 119–126. http://www.sciencedirect.com/science/article/pii/S0306456515000273#

Germano, D. J. 1994. Growth and age at maturity of North-American tortoises in relation to regional climates. – Can. J. Zool. 72: 918–931.

Goodlett, T., Goodlett, G. and Hamilton, P. 1998. Review of Radio Transmitter Attachment Techniques for Turtle Research and Recommendations for Improvement. – Herpetol. Rev. 50(26): 9

Karns, D. R. 1986. Field herpetology: Methods for the study of Amphibians and Reptiles in Minnesota. James Ford Bell Museum of Natural History, University of Minnesota. –Occasional paper 18: 1–88.

Lagarde, F., Bonnet, X. Henen, B. T. Corbin, J., Nagy, K. A. and Naulleau, G. 2001. Sexual size dimorphism in steppe tortoises (Testudo horsfieldi): growth, maturity, and individual variation. – Can. J. Zool. 79: 1433–1441.

Malonza, P. 2003. Ecology and distributions of the pancake tortoise, Malacochersus tornieri in Kenya. – J. East Afr. Nat. 92: 81–96.

Spawls, S., Howell, K., Hinkel, H. and Menegon M. 2018. A field Guide to East African reptiles. – Bloomsbury Publishing.

Sutherland, W. J. (Ed). 1996. Ecological census techniques. A handbook. – Cambridge University Press, Cambridge.

Van Belle, S. 2017. Observational Methods. In: Bezanson, M., Mackinnon, K. C., Riley, E., Campbell, C. J., Nekaris, K., Estrada, A., DiFiore, A. F., Ross, S., Joes-Engel, L. E., Thierry, B., Sussman, R. W., Sanz, C., Loudon, J., Elton S. and Feuntes, A. (Eds.). – The International Encyclopedia of Primatology, Wiley and Sons. Inc.  https://doi.10.1002/9781119179313.wbprim0385.

Wood, R. C. and MacKay, A. 1997. The distribution and status of pancake tortoises, Malacochersus tornieri, in Kenya. In: Van Abbema, J. (ed.) – Proceedings: Conservation, Restoration, and Management of Tortoises and Turtles pp. 314–321.