Pollination systems in the genus Aloe are diverse, with numerous recent studies identifying unique and often unexpected mutualisms (Arena, Symes & Witkowski 2013; Botes, Johnson & Cowling 2009a; Botes, Wragg & Johnson 2009b; Hargreaves, Harder & Johnson 2008; Johnson 2004). While the perception remains that many Aloe species are pollinated by birds, relatively few studies have quantified the contributions of different pollinator guilds. Furthermore, in cases where a specific pollinator guild (e.g. birds) is identified as the most important contributor to pollination, fewer studies have addressed the specific roles of each species in that guild.

Many Aloe species can potentially hybridise and given the diversity of the genus it is important to understand how the species integrity of co-occurring taxa is maintained. Botes, Johnson & Cowling (2008) proposed that the problem of Aloe species coexistence may be explained by a greater diversity of bird pollination systems in the genus than previously documented. In five co-occurring Aloe species that flower together during winter there was a partitioning of pollinators, with short-billed generalist nectar feeders pollinating species with large amounts of dilute nectar in short corolla tubes, and long-billed specialist nectar feeders, that is, sunbirds, pollinating species with small amounts of concentrated nectar in long corolla tubes (Botes et al. 2008).

Many species-specific studies have provided valuable data regarding the widely distributed Afrotropical genus Aloe, where the diversity of species, with respect to (1) nectar characteristics, including volume and concentration, (2) flowering season and duration, (3) inflorescence, raceme and flower arrangement, (4) floral morphology, including structure, colour and position, (5) distribution and habitat preferences and (6) population size, is important in determining plant–pollinator associations (Botes et al. 2008, 2009a, 2009b; Hargreaves, Johnson & Nol 2004; Hargreaves, Harder & Johnson 2012; Hoffman 1988; Johnson 2004; Pailler, Warren & Labat 2002; Ratsirarson 1995; Stokes & Yeaton 1995; Wilson et al. 2009). While species-specific predictions of pollinators can be made on the basis of pollination syndromes, it is clear that determination of pollinator assemblages requires thorough field study (Fenster et al. 2004; Geerts & Pauw 2009; Johnson & Steiner 2000). An important tool in this regard is the application of camera traps, which in recent years have provided important and detailed information on plant–pollinator interactions (Melidonis & Peter 2015; Steenhuisen et al. 2015; Zoeller et al. 2016).

Aloe reitzii Reynolds (1937) var. reitzii (Asphodelaceae) is a summer-flowering (late January–March) succulent with a restricted distribution in montane grassland around Belfast and Dullstroom, northeastern South Africa (Glen & Hardy 2000; Mucina & Rutherford 2006; Reynolds 1969; Van Wyk & Smith 2003) (Figure 1a and b). It grows on rocky areas, but is by no means confined to them (Glen & Hardy 2000; Reynolds 1969). It is known from 12 to 15 locations, with most populations confined to a small area (Raimondo et al. 2005). Plants reach a height of up to 1.0 m and produce multiple racemes that can reach 2.5 m in height (Glen & Hardy 2000; Reynolds 1969; Van Wyk & Smith 2003). The immature unopened flowers are dark red, becoming orange to pale yellow when mature (Van Wyk & Smith 2003), and are incurved and relatively long (32 mm–50 mm), suggesting pollination by birds (Figure 1a and b) (Glen & Hardy 2000; Reynolds 1969; Symes, Human & Nicolson 2009; Van Wyk & Smith 2003).

The objective of this study was to investigate potential pollinators of A. reitzii var. reitzii as inferred by visitation events captured by remote camera traps, as well as by field observations.

While this study considers visitation data from only three days of a flowering season, camera traps and field observations provided information on the most likely pollinators of A. reitzii var. reitzii. In the grassland environment in which A. reitzii var. reitzii occurs there are a limited number of grassland bird species (Hockey, Dean & Ryan 2005), including specialist (e.g. sunbirds) and generalist avian nectarivores. Only a few potential bird-pollinators that are likely to visit flowering plants occur within A. reitzii var. reitzii grassland habitat and those recorded for the period of study are not unexpected (Hockey et al. 2005), given that flower and inflorescence structure in A. reitzii var. reitzii strongly suggests pollination by birds. Furthermore, nectar characteristics suggest that A. reitzii var. reitzii is more likely to be pollinated by sunbirds. Generalist avian nectarivores are predicted to prefer flowering plants with dilute (8%–12% w/w) and large volumes (40 μL–100 μL) of nectar, while specialist passerine nectarivores (i.e. sunbirds) prefer nectar of higher concentration (15%–25% w/w) with lower volumes (10 μL–30 μL) (Johnson & Nicolson 2008). With values for A. reitzii var. reitzii at the lower end of the concentrations predicted for specialist nectarivores, and volumes extending into the range expected for generalist nectarivores, it is not surprising that the majority of floral visits were by Cape Weavers, which are generalist nectarivores and are adept at feeding on nectar (Craig 2014; Johnson & Nicolson 2008). Malachite Sunbirds were recorded on as many study plants as Cape Weavers, suggesting that these two species may both be important pollinators of A. reitzii var. reitzii. Malachite Sunbirds are ubiquitous throughout southern Africa and have been identified as primary pollinators of an entire guild of plant species – the ‘Malachite Syndrome’ (Geerts & Pauw 2009).

Many of the study plants received significant visitation from invertebrates. While it is possible that several of these were nectar robbers (Richardson 2004), honeybees were recorded in this study and have been shown as important pollinators of other Aloe species, (Botes et al. 2009a; Symes et al. 2009). They made up by far the greatest proportion of recorded insect visitors so, like in A. greatheadii var. davyana, they may be important pollinators when they occur in large numbers (Symes et al. 2009). Interestingly, observations of a wide variety of insects visiting study flowers contrasts findings in winter-flowering Aloe species which, unless they are visited by honeybees, are less likely to be visited by insects (Botes et al. 2009a; Hoffman 1988; Kuiper et al. 2015; Payne, Symes & Witkowski 2016; Symes et al. 2009).

Aloe peglerae, an endemic grassland Aloe of the Magaliesberg Mountain Range, is pollinated mainly by Cape Rock Thrush Monticola rupestris, a diurnal generalist nectarivore, with > 60% of pollination made by this species (Arena et al. 2013; Payne et al. 2016). However, pollination is also supplemented by nocturnal mammals such as Namaqua Rock Mouse Micaelamys namaquensis and Eastern Rock Sengi Elephantulus myurus (Payne et al. 2016). Anecdotal reports from the landowner at Site 2 suggest that sengi (Elephantulus sp.) have been observed visiting flowers during the day (B. Struwig, pers. comm.). However, no photographs were obtained of small mammals visiting flowers during this study. Thus, while visits by small mammals, such as sengi (Wester 2010), may be more common in rocky areas at our study sites, this mammal–plant association remains to be investigated.

In conclusion, A. reitzii var. reitzii is visited primarily by generalist and specialist avian nectarivores, as well as by invertebrates. Although the latter cannot be confirmed as legitimate pollinators, the fact that (1) birds and insects have been shown to be successful pollinators of numerous Aloe species and (2) fruit set at the site is abundant each year (landowners, pers. obs.) suggests that these organisms are important role players in the pollination of A. reitzii var. reitzii. A generalist pollination system might contribute to fitness of A. reitzii var. reitzii, a range-restricted, summer-flowering endemic occurring in fire-dominated grasslands; this remains to be investigated further. Furthermore, this study shows the usefulness of remote camera traps in testing predictive floral syndromes in the diverse genus Aloe.