The Jamaican flower bat Phyllonycteris aphylla (Chiroptera: Phyllostomidae) is endemic to Jamaica and categorized as Critically Endangered on the IUCN Red List (Koenig & Dávalos, Reference Koenig and Dávalos2015). Kurta & Rodríguez-Durán (Reference Kurta and Rodríguez-Durán2023) estimated there were 500 individuals (up from 250; Koenig & Dávalos, Reference Koenig and Dávalos2015) at Stony Hill Cave, which was purchased by the Natural Resources Conservation Authority in partnership with Bat Conservation International (Donahue, Reference Donahue2019). Despite its significance as a nocturnal pollinator, little is known about the ecology and natural history of this species, and it has twice been declared extinct. After the species was first described in 1898 (Miller, Reference Miller1898), only fossilized specimens were collected until the species was rediscovered at St. Clair Cave in 1957 (Goodwin, Reference Goodwin1970). Capture records from St. Clair, Riverhead, Mount Plenty, Marta Tick, Wallingford and Oxford Caves all indicated the presence of P. aphylla there from the 1960s to the 1980s although they have not been captured at these sites since (Pregill et al., Reference Pregill, Crombie, Steadman, Gordon, Davis and Hilgartner1991; Genoways et al., Reference Genoways, Baker, Bickham and Phillips2005). Genoways et al. (Reference Genoways, Baker, Bickham and Phillips2005) found that P. aphylla was relatively common in certain localities between 1974 and 1985, but there are no published records of P. aphylla after 1985, and the species was believed extinct until its rediscovery at Stony Hill Cave in 2010 (Koenig & Dávalos, Reference Koenig and Dávalos2015).

What is known about the ecology and life history of P. aphylla is limited and anecdotal. It appears to form smaller colonies than heterospecifics, as observed by Goodwin (Reference Goodwin1970) in St. Clair Cave and based on our present knowledge from Stony Hill Cave. The data available suggests this species feeds on fruits, nectar, pollen and insects (Nowak, Reference Nowak1999). It is believed to be a cave-obligate rooster, and based on prevailing descriptions seems to be limited to so-called hot caves (Genoways et al., Reference Genoways, Baker, Bickham and Phillips2005), which are defined by high bat densities that result in a high temperature and low air quality within particular chambers; often these chambers have small entrances and abundant detritivores consuming guano in passages below the roost, creating additional heat and consuming the limited oxygen (Rodríguez-Durán, Reference Rodríguez-Durán1995; Ladle et al., Reference Ladle, Firmino, Malhado and Rodríguez-Durán2012).

In March–April 2023, we surveyed bats at seven caves throughout Jamaica. During this survey we encountered reproductive male and female P. aphylla at two sites where there were no previously documented captures of this species. Here we describe our observations at these two sites, and call for further research on this important species. Over 11 netting nights between 22 March and 2 April 2023, we surveyed caves at Windsor, Green Grotto, St. Clair, Mosley Hall, Riverhead, Mount Plenty and Rock Spring (Fig. 1). All sites except for Rock Spring have been previously surveyed for bats, and there are records of captured P. aphylla from Windsor, St. Clair, Riverhead and Mount Plenty (Genoways et al., Reference Genoways, Baker, Bickham and Phillips2005). At each site, we deployed 1–2 harp traps (Faunatech Austbat, Mount Taylor, Australia) for 2–5 hours. Captured bats were placed in individual cloth bags for subsequent processing and data collection, which involved measuring mass and forearm length, and collecting hair, faeces and wing tissue samples from some individuals. All individuals were released at the point of capture.

Fig. 1 Caves on Jamaica visited or known to have Jamaican flower bat Phyllonycteris aphylla roosts historically and/or currently. Caves surveyed between 22 March and 3 April 2023 are indicated by squares and those not visited as a part of this study are indicated as triangles. One cave surveyed during this study (Mosley Hall) did not have historical records of P. aphylla and none were captured there, this has therefore been omitted from this map. There is no confirmed location for Wallingford which has a historical record for the species (Genoways et al., Reference Genoways, Baker, Bickham and Phillips2005). Because of permit restrictions, we are unable to provide precise locations of caves.

On the evenings of 24 March and 2 April 2023, we surveyed Green Grotto Caves, near Discovery Bay, St. Ann parish. This cave is both a historical site of national importance and a major tourist attraction. Subfossil remains of P. aphylla were collected in 1919–1920 from this cave complex (Koopman & Williams, Reference Koopman and Williams1951), although there are no published records of P. aphylla being captured (Genoways et al., Reference Genoways, Baker, Bickham and Phillips2005). The principal bat roosting area (known as Bat College) was believed to hold thousands of bats, primarily Erophylla sezekorni, and potentially P. aphylla (D.L. Whyte, pers. comm., 2022). Phyllonycteris aphylla were captured there by H. U. Schnitzler sometime between 2000 and 2003, although no specimens were collected and no record was published (Koenig, Reference Koenig2018); acoustic monitoring also supported their likely presence (Koenig, Reference Koenig2018). Fenton et al. (Reference Fenton, Fenton, Koenig and Whyte2020) photographically surveyed the Lake-Side Bat Exit, believed to be connected to Bat College. Among hundreds of bats photographed, they observed four male P. aphylla using this exit (Fenton et al., Reference Fenton, Fenton, Koenig and Whyte2020). No female P. aphylla were recorded. On both nights of trapping we set a single harp trap perpendicular to the Lake-Side Bat Exit at c. 17.45, and captured bats until 22.30. We captured several hundred bats of seven species, including 20 P. aphylla (14 on 24 March 23, 6 on 2 April 23). Among these 20 individuals, we captured five pregnant females and two adult females that were not obviously pregnant at the time of capture. These individuals were processed and an additional four (including another pregnant female) were captured at 22.30 on 2 April (Table 1). These records are the first published captures of P. aphylla at Green Grotto Caves, as well as the first confirmation of reproductive females at this site.

Table 1 Capture records of 24 Phyllonycteris aphylla netted at Lake-Side Bat Exit, Green Grotto Caves, St. Ann, Jamaica. Capture times indicate when bats were extracted from harp traps. The final four bats were released at the trap without processing. All individuals were adults.

Rock Spring Caverns, St. Mary parish, has not been previously surveyed for bats, although it was geologically surveyed and mapped by a team from Leeds University, UK, in 1963 and is well known to local people and cavers (Fincham, Reference Fincham1997). Exploration by the Jamaican Caves Organisation from 2020 onwards documented 11 holes that appear to give access to the caverns, seven of which were numbered by the original expedition and four subsequently named; e.g. Big Hole and Scary Hole (Fig. 2). Assessments revealed that bats roost throughout most of the cave at varying densities, and based on knowledge of caves throughout the island, it is believed to be one of the largest communities of roosting bats in Jamaica (R.S. Stewart, pers. comm., 2023). On the evening of 1 April 2023, we visited Rock Spring Caverns at Big Hole and set two harp traps at the south-facing edge at c. 18.00. Bats began to emerge at c. 18.20 and through to 20.00 we captured c. 200 individuals of six species; among these were 14 P. aphylla (Table 2). By 20.30 the harp traps had captured c. 100 additional bats, the majority being P. aphylla. We assessed age-class, sex and reproductive status of each P. aphylla before it was released. Among this later group we captured 26 non-reproductive adult males and 23 pregnant females; there were an additional three adult females that were not obviously pregnant. Assuming no individuals were recaptures (we were unable to mark these bats because of permit restrictions), we captured a total of 66 mature P. aphylla. Based on the IUCN Red List assessment, this is likely the largest known population of the species (Koenig & Dávalos, Reference Koenig and Dávalos2015).

Fig. 2 The internal layout of Rock Spring Caverns, St. Mary Parish, Jamaica (Fig. 1). This map was made as part of a 1963 Leeds University expedition and adapted from Fincham (Reference Fincham1997). Temperatures throughout the cave are believed to be stable (23.1–23.6 °C; external ambient 30.0 °C). (1) Constriction < 1 m², periodically submerged; (2) Bat numbers variable, distance below surface c. 30 m; (3) Large roost area, distance below surface c. 35 m; (4) Bats throughout, fewer than location 3, distance below surface c. 40 m; (5) Bats throughout, distance below surface c. 50 m; (6) Big Hole, bats roosting throughout, distance below surface c. 60 m; (7) Deep guano deposits distance below surface c. 30 m; (8) Scary Hole, bats roosting throughout, distance below surface c. 60 m; (9) Constriction c. 1 m² at bottom of Pebble Chute, periodically submerged, distance below surface c. 60 m.

Table 2 Capture records of 14 P. aphylla trapped at Big Hole, Rock Spring Caverns, St. Mary, Jamaica. All bats were captured in two harp traps and were extracted at approximately the same time ± 15 min. These records do not include 52 individual P. aphylla captured at 20:30 in the harp traps on the same day; there were 26 males and 26 females (23 pregnant). All individuals were adults.

¹ Not weighed.

The capture records presented here substantially alter what was previously known about this species. Not only do they increase the number of sites and reproductive individuals known, but they also the assumption that P. aphylla is only found in hot caves. Although Stony Hill Cave is hot (ambient temperatures of ≤ 38 °C), both Green Grotto and Rock Spring contain flowing or pooled water that moderates temperatures within these caves, and both are described as cool (i.e. below external ambient temperatures) with regard to their overall temperature profile (R.S. Stewart, pers. comm., 2023). At Rock Spring Caverns, although some roosting areas are perceived to be warmer than other areas, no surveyed locations in the cave are hot (23.1–23.6 °C in passages and roost spaces; ambient 30.0 °C). Likewise, at Green Grotto Cave, although some unexplored areas may have temperatures above ambient, the principal bat roosting areas are cooler than external temperatures (27.5–29.2 °C; ambient 35.6 °C; D.L. Whyte, unpubl. data, 2023). McFarlane (Reference McFarlane1986) suggested that P. aphylla may also be restricted to caves associated with abandoned river resurgences, however this is not the case at Stony Hill, Green Grotto or Rock Spring. There are no known uniting geological or other physical features between current and previously known roosts. Thus, we believe there is no known limitation for a cave to host a P. aphylla colony and more caves should be surveyed to examine this.

Future work on P. aphylla should focus on locating additional roosts and safeguarding the three known populations. The primary threats to this species include disturbance from guano harvesters and predation by feral cats (J. Flanders, pers. comm., 2023; M.B. Fenton & D.L. Whyte, unpubl. data, 2019). However, it is unclear why we did not record the species at previously known roosts such as St. Clair Cave; we do not know whether this is a result of abiotic factors, such as changes in cave microclimate, or disturbance by people or feral cats. Monitoring of known roosts is required to ensure the survival of this Critically Endangered species, as is improving protection and establishing protected areas around known roosts to reduce disturbance, as Bat Conservation International has done at Stony Hill Cave (Meierhofer et al., Reference Meierhofer, Johnson, Perez-Jimenez, Ito, Webela and Wiantoro2023). Further surveys of Rock Spring Caverns are required to identify chambers of importance for P. aphylla, and deployment of temperature loggers to identify microclimates favourable to roosting within this cave system. We encourage the Urban Development Corporation and National Environment and Planning Agency to create a bat monitoring programme for the Bat College population. Our future work at Rock Spring Cave will focus on monitoring bats at other exits (we only trapped at one of the 11 known entrances) to refine the estimate of population size, and identify areas of ecological importance near the cave to inform management decisions. We feel these new records provide hope for the future of this Critically Endangered endemic species.