The Search for Nepenthes: the Use of Species Distribution Modelling to Find High Suitability Areas in the Philippines. Charles K

Home , Martin Cheek, Nepenthes alata, Nepenthes maxima, Nepenthes ramos, New Nepenthes

The Search for Nepenthes: The Use of Species Distribution Modelling to Find High Suitability Areas in the Philippines.

Charles King

2020

Dissertation submitted for the degree of Master of Science in Plant and Fungal Taxonomy, Diversity and Conservation awarded by Queen Mary, University of London. https://doi.org/10.34885/181

Abstract

The Philippines long been known as a biodiversity hotspot and recent studies have indicated that it contains the highest number of Nepenthes species, surpassing previous species rich areas of Borneo and Sumatra. Despite the Philippines importance for Nepenthes species, deforestation across the archipelago with an estimated 12,300 km² of tree cover lost from 2001-2019, threatening Nepenthes populations. It is therefore important to highlight areas of suitability where Nepenthes populations and new species may be found. Within this study a species distribution model was used in conjunction with BIoclim variables and Global Forest Cover to create a map of suitability areas for Nepenthes. Results from this study indicate that there are 3 prioritised areas for future exploration when considering future deforestation threats and accessibility. Here we present priority areas for future exploration in the context of suitability, deforestation threats, and accessibility.

Introduction

Nepenthes is a monotypic genus belonging to Nepenthaceae, a carnivorous plant family renowned for their unique morphology and carnivorous adaptations (Bohn and Federle, 2004). Nepenthes can survive in nutrient poor environments through the use of a fluid-filled pitcher which is attached to the leaf blade by a tendril (Murphy et al., 2020). This pitcher organ is found universally throughout Nepenthaceae, although their arrangements for trapping nutrients, and their pitcher morphology can vary greatly between species (Murphy et al., 2020). Of the extant Nepenthes species, it is estimated that 160-180 species are accepted (Murphy et al., 2020; Plants of the World Online, 2020). New Nepenthes species are frequently discovered in the Philippines, including N. maximoides, which was described as part of this project (King & Cheek, in press see Appendix 1).

Nepenthes has a paleotropical distribution mostly across Malesia (Murphy et al., 2020). Previously, Sumatra and Borneo were thought to be most species rich, both with 30 species (Cheek and Jebb, 2001). Recently, the Philippines has been shown to be richer with 59 species (King and Cheek, in press). As the world’s current hotspot for Nepenthes diversity, the Philippines is now thought to have the highest potential for finding new Nepenthes species (King and Cheek, in press). However,

Recent figures estimate that from 2001-2019, the Philippines lost 12,300 km² of tree cover, the majority of which was lost from natural forests (Global Forest Watch, 2020). Historically, habitat destruction can be observed on the larger islands of Luzon and Mindanao with much of the lowland forests now lost to agricultural and industrial practices (Sohmer and Davis, 2007, Cheek and Jebb, 2013). The deforestation occurring across most of the Philippine archipelago has caused the endangerment and extinction of a variety plant species (Hughes, 2017; Galindon et al, 2018; Lillo et al, 2020). Several Nepenthes species that have only recently been identified and described have gone extinct, such as N. maximoides, collected in 1911, and N. extincta, collected in 1978 (Cheek and Jebb, 2013; King and Cheek, in press). Currently, these species are classified as Critically Endangered, but they have not been seen in situ since their original collection, thus are likely extinct (Cheek and Jebb, 2013; King and Cheek, in press). This suggests that there may have been species that have become extinct before they have been able to be identified and protected. The risk of extinction of Nepenthes species currently undiscovered, alongside the reduction of suitable habitat, has highlighted the need to identify key areas for future exploration and conservation.

Species distribution modelling (SDM) can be used to highlight priority areas for current populations and for future species discovery (Williams et al., 2009). SDM uses a combination of environmental variables, previous collection sites and expert knowledge or machine learning to create a map of suitability tailored to a chosen organism (Beaumont, Hughes and Poulsen, 2005). Previous studies have successfully used SDM to highlight areas for conservation, exploration and to predict expected distribution of a variety of species (Williams et al., 2009). The environmental variables used can be tailored to each species being investigated and multiple variables per species can be used; layering these variables creates predicted patches of low to high suitability in which populations of the species may occur (Williams et al., 2009). For the purposes of this paper, the SDM will use climatic data with collection points and expert knowledge to select climate variables. As Nepenthes generally require particular abiotic factors within their habitats, expert knowledge can be used to choose variables which best represent these factors (Pecchi et al., 2019). Previous studies have used SDM to identify locations of exploration for both endangered tree species and fungal species within the Philippines (Garcia et al., 2013; Almadrones-Reyes and Dagamac, 2018). These studies demonstrate how predictive techniques can provide information on distribution, and how this can be used for conservation efforts in the region (Garcia et al., 2013).

© The Author. All rights reserved. 3 SDM has not yet been undertaken for Nepenthes in the Philippines, despite the impact it could have for conservation of the genus. This paper will explore areas of suitability for a variety of Nepenthes species using climatic variables, forest cover, and the specimens from K herbarium identifed by Jebb and Cheek and their subsequent papers (Cheek and Jebb 2013a-g).

Material and methods

N. maximoides paper

Following the guidelines from the IUCN (2012), a conservation assessment of N. maximoides was written. Herbarium material was examined and drawn using a Leica Wild M8 dissecting binocular microscope (Further details and scopes used in Appendix 1). To expose characters required for identification and characterisation of the species, the specimen was partly unmounted. This was due to some critical features being obstructed in the original mounting. The map of mountain distributions was made using SimpleMappr (https://www.simplemappr.net). Details of the collection location were gathered by analysing other herbarium specimens collected by H.M. Curran (JStor Global Plants, 2020)

Nepenthes Species Distribution Modelling

Georeferenced for species records were taken either from herbarium specimens at RBG Kew (K), citations of herbarium specimens from their monographs (Cheek & Jebb, 2013), or the secondary source Philippineplants (2020). Where specimen labels did not have a georeference, the location given on the specimen was observed on Google Earth and given coordinates on nearby high-altitude areas (>1000m) with appropriate forest cover near to settlements. These criteria for Nepenthes populations were given based on previous habitats where Nepenthes have been found and previous collections (King & Cheek, in press; Gronemeyer et al., 2016)

A total of 71 Nepenthes specimens were selected to provide this study with previous collection points (full list of species used can be found in Appendix 2). Within this study, four Bioclim variables were used to provide climate data for the Philippines (Beaumont, Hughes and Poulsen, 2005; Worldclim, 2020). All 19 variables provided by Bioclim were examined for correlations against previous collections (Worldclim, 2020). From these variables, the 4 chosen best cover Nepenthes niche and has the least correlation when compared to the overall Philippine climate (Beaumont, Hughes and Poulsen, 2005). These variables are Annual Temperature, Temperature Seasonality, Annual Precipitation, and Precipitation of the Driest Quarter (Worldclim, 2020). To create a map of suitability, a combination of the packages raster, rdgal, ggplot2 and tidyr within R-Project (Appendix

© The Author. All rights reserved. 4 4) were used in conjunction with Bioclim to gather data on each of previous collections (R Core Team, 2014; Worldclim, 2020). Due to the occurrence of outliers and extreme values, the upper and lower 5% of the histograms were removed (Limpert and Stahel, 2011). As this study examines the general suitability various Nepenthes specimens with a low sample size, the removal of more than 5% of the points would render the data unrepresentative of the populations in question. Once given the threshold each Bioclim variable was given a presence or absence value of 1 or 0 and then combined with addition to create a suitability map, with values ranging from 1-4 (low to high suitability) (Poirazidis et al., 2019). A forest cover layer was applied overlaying this map to remove areas below 60% forest cover. The threshold of 60% was applied following the guidance of Hill et al., 2019, who suggests that this best represents tropical undisturbed forests. As collection points are historic, the use of recent forest cover data will outline areas which have been lost since.

Figure 1: Histograms of climate data against species occurrence points.

Species distributions, when related to mean annual temperature (Bio1), indicate that the highest frequency of Nepenthes are found between 22°C-27°C, with temperatures lower than 19’C being suitable for only a few species. The distribution of species against temperature seasonality (standard deviation ×100) (Bio4) indicates that Nepenthes species prefer a lower range of temperature seasonality between 4-10°C. Above 10°C, there is a significant decrease in the occurrence frequency

© The Author. All rights reserved. 6 of Nepenthes. The highest occurrence frequency for annual precipitation (Bio12) is around 2500- 3000mm per year with a reduction of species occurrence above 3000mm per year. Distributions for precipitation of the driest quarter (Bio17) indicate that the highest Nepenthes frequencies are found at 600mm per quarter with higher amounts of rainfall (400-600cm) having the highest frequencies Nepenthes species.

© The Author. All rights reserved. 7 Figure 2: Map of the Philippines layered with annual mean temperature, temperature seasonality, annual precipitation, precipitation of driest quarter (Worldclim, 2020). Suitability scale is defined by the number of climate variables suitable for Nepenthes found in an pixel. Where 1 would be low suitability (one variable), 2-3 would be moderate (two or three variables),and 4 would be high suitability (all variables present). Blank sections of Figure 2 refer to areas where forest cover is below 60% (Global Forest Change, 2020). Points from previous Nepenthes collections marked in blue (Nepenthes species used can be found in Appendix 2).

Figure 2 shows that a majority of the Philippines does not have appropriate canopy cover for the occurrence of Nepenthes. The areas with the highest suitability are generally found on the eastern sides of the Philippines, with some intermittent patches found on each island of the Luzon and Visayas areas. Moderate suitability areas can be found in small areas on both the east and west of Luzon and Mindanao. Palawan shows only a mix of both low and moderate suitability.

Figure 3: Regional divisions of the Philippines, taken from Bravo et al., (2014).

Discussion

1 Previous collection distribution

As expected, overall findings indicate that previous collection points are commonly found in areas where there is moderate to high suitability for Nepenthes species. Within Luzon, previous collections are largely found on both the Sierra Madre northern mountain range, Cordellia Central range (Cordellia administrative region and Region II), and on Mount Halcon located in central Mindoro

© The Author. All rights reserved. 9 (Region IVB). When observing Visayas, a large cluster of points can be seen in the centre of Romblon, with singular points scattered across adjacent islands, such as Leyte and Samar (Region VIII). Other previous collections points within Visayas include Central Antique, Malalison island, and Negros (Regions VI to VII). Mindanao has sparsely distributed points across the northern, central, and southern regions (Regions X, XI, XII) with a notable cluster towards the Caraga region (Region XIII). Palawan (located west of Region VI) also has a variety of points distributed across the island, with a majority being found in the central and southern regions.

2 Areas of suitability

Although these previous collections points are found in or around moderate-high to high suitability areas, there are a still large number of areas that are largely unexplored and can provide guidance for future collections. Figure 2 indicates three large areas of high suitability, thus named for the purpose of this study Eastern Luzon distribution, Visayas Leyte distribution and North Eastern Mindanao distribution. They differed in which environmental variables made them suitable, their remaining habitat and their accessibility.

The Philippines can be divided into 5 climate types, Type I, Type II, Type III and Type IV; details of each climate can be found in appendix 4 (Tolentino et al., 2016). Type II and Type IV climates both have high amounts of rainfall distributed throughout the year with no notable dry season, whereas Type I and Type III both have notable dry seasons from November to April. Results for annual rainfall and precipitation of the driest quarter indicate that higher rainfall is where the highest frequency of Nepenthes are present, suggesting Type II and Type IV are the most suitable climates (Tolentino et al., 2016).

2.1 The Eastern Luzon Distribution

Although six collections have been made around the northern and central section of eastern Luzon, the entirety of the high suitability area seems significantly under collected. The high suitability strip observed in eastern Luzon suggests that the variables used in this paper closely correlate with the temperature and the rainfall of the area. The Eastern Luzon coast experiences a Type II climate, with rainfall distributed throughout the year. This is supported by Findings of Figure 1 which shows that high annual rainfall are where Nepenthes occur at the highest frequencies (Tolentino et al., 2016).

© The Author. All rights reserved. 10 Another key driver for the notable distributions seen in Eastern Luzon is the smaller temperature seasonality range, which can be the result of a high altitude (Linacre, 1982). This effect is most notable when viewing the Sierra Madre mountain range with Luzon. As suggested by Figure 1, a smaller temperature seasonality between 4-10°C are where the highest frequencies of Nepenthes species are found. This suggests that high altitude areas in general may be more suitable for Nepenthes due to their more stable temperature seasonality (Linacre, 1982). Furthermore, when viewing the Sierra Madre distribution in comparison to surrounding areas, there is a notable outline where forest cover becomes reduced in low altitude areas (Rickart, Heaney, Balete and Tabaranza, 2011). The low forest cover observable in central Luzon can be largely accredited to intensive agricultural practices and increasing urbanisation in response to a growing population (Cheek and Jebb, 2013). These disturbances overtime have resulted in a highly fragmentated ecosystem losing areas that once may have been suitable according to the parameters of this study (Rickart, Heaney, Balete and Tabaranza, 2011). Higher altitude areas currently have reduced rates of deforestation (Spracken et al, 2015), although these areas may be affected in the future. Fortunately, some of the Sierra Madre range is protected to prevent any further forest cover reduction (Minter et al., 2014). One such protected area is the Northern Sierra Madre Natural park which covers 3594.86 Km² (Minter et al., 2014). This park contains a variety of habitats such as Luzon montane rainforests and mossy forests, which are known to harbour Nepenthes populations (Gronemeyer et al., 2014; Minter et al., 2014). As this study aims to highlight areas of the Philippines which are favourable for Nepenthes but lack observations, a more favourable area may be the central or southern mountain ranges. Although not protected, the central and southern Sierra Madre range have identical climatic conditions and habitats to the northern range (Tolentino et al., 2016). This suggests there may be potential for new populations and new species to be found in this area.

2.2 Samar/Leyte distribution

The results of this study also identify Samar and Leyte as high suitability areas. This finding is supported by two collections found centrally on each of the islands. As with Luzon, Samar experiences a Type II climate, the characteristic high rainfall of which is favourable for Nepenthes populations (Tolentino et al., 2016). A vast area of Samar and its forests is protected by the Samar Island National Park (SINP). This park encompasses 3333 km² centrally on the island and contains the largest tract of old growth forest in the Philippines (Holden, 2012). This area has been designated a national park, protecting it from deforestation, and thus previously sampled populations may still be

© The Author. All rights reserved. 11 present (Holden, 2012). Furthermore, future exploration will also be viable in this area as there will be local guides and appropriate infrastructure in place to aid with accessibility (Holden, 2012).

The rainfall climate of Leyte is categorised as Type IV. Although different to the Type II climate in Samar, both areas lack a dry season and have rainfall distributed throughout the year, with the distinction of Type II having a pronounced rain season from November to January (Tolentino et al., 2016). The most suitable areas are found centrally extending southwards to southern Leyte. These areas are uninhabited, have a high altitude, and contain necessary habitats for Nepenthes populations to be present. The surrounding low forest cover seen in the northern and eastern costs of Leyte is the consequence of intensive agriculture and urban expansion (Siebert, 1984). Leyte is one of the most densely populated islands in the Philippines, with a rapidly growing population. To sustain this population growth much of the lowlands has been converted from forests to intensive agricultural land growing primarily coconut, sugar, and rice (Mukul, Herbohn and Firn, 2016). The ongoing increase in need for agricultural land is likely to decrease in area cover of land considered here to be highly suitable for Nepenthes.

Overall, the importance of Samar and Leyte for present Nepenthes populations has been underestimated in its potential when considering the amount of previous collections and GBIF occurrences (Gbif, 2020). Although islands have high suitability and appropriate habitats, Samar may be more favourable than Leyte for fieldwork due to the protections afforded to the forest by the SINP. However, as Leyte still has areas of high suitability but is at risk of deforestation, expeditions should be in Leyte to locate new species or threatened populations which may be at risk from deforestation (Mukul, Herbohn and Firn, 2016).

2.3 North Eastern Mindanao distribution

The north eastern distribution in Mindanao displays areas of moderate to high suitability. As with Leyte, the rainfall in this area is typical of a Type IV climate, and is appropriate for Nepenthes populations (Tolentino et al., 2016). Moreover, as seen in the Sierra Madre range Luzon, the sections of Mindanao with the highest suitability seem to surround high altitude areas (>1000m). Within Northern Mindanao these high-altitude areas are found on the border of the Caraga Region and Northern Mindanao surrounding Mount Mangabon (extending south to the Davao Region). These high-altitude areas contain primarily Dipterocarp forests and mossy forests, where Nepenthes can usually be found (Gronemeyer et al., 2014). At the base of this area, slightly east, is the Agusan River

© The Author. All rights reserved. 12 basin. Unlike most areas highlighted as high suitability, this area is marshland, low in gradient and altitude. (Gronemeyer et al., 2014). This area would be important to investigate as it is one of the few areas of lowland outlined here.

3 The Case of Palawan

Figure 1 suggests that Palawan has no areas of high suitability but instead only has areas of low to moderate suitability. Palawan is unique in the Philippines, which may be the causation for the lower suitability observed in this study. Palawan has a longer dry season from April to November in the north, and an infrequent dry to rainy seasonality in the south (Type I and Type III) which would not be considered suitable for Nepenthes species by the variables used in this study (Tolentino et al., 2016). Despite the apparently unsuitable environment, collections of Nepenthes have been made in Palawan, and there is location point data for populations recorded on GBIF (Gbif, 2020). This indicates that there are further factors present in Palawan which influence suitability for Nepenthes.

A possible explanation for the extensive Nepenthes species found in Palawan is the island’s unique geology. A third of the island’s surface is formed of ultrabasic rock present in most high altitude areas, notably the central mountain chain (Robinson et al., 2009; Cheek and Jebb, 2013). The soil formed from ultrabasic rock is low in phosphorus, potassium and calcium which is essential for the growth of most plant species (Proctor, 2003). As a result, only plants adapted to this soil type, such as carnivorous plants, can occupy these environments. The geology of Palawan is therefore an important variable to consider for future studies and seems to be a key factor in understanding how the Nepenthes species are distributed across Palawan. As geology is a key factor in Palawan it would also be worth investigating in other areas which have perceived low suitability (Andal, Arai and Yumul, 2005). It is notable that areas within Mindanao, Samar/Leyte, and Luzon which are considered to be low/moderate suitability are also areas in which ultrabasic rock deposits have been recorded, and therefore it is possible that there are extant Nepenthes populations there (Andal, Arai and Yumul, 2005; Guotana et al., 2017).

Another factor which may be influencing distribution in not only Palawan but other low suitability areas throughout the Philippines is the air humidity (Bauer, Bohn and Federle, 2007). Within this study, the factor of humidity on Nepenthes was considered and replicated using a combination of temperature and rainfall and forest cover. However, a direct measurement of humidity would be necessary to fully assess an area’s suitability, an area with high humidity may have been discounted

© The Author. All rights reserved. 13 because of low rainfall, and thus would be overlooked when working to find new populations (Bauer, Bohn and Federle, 2007). Areas which are labelled as moderate or low suitability (particularly as Palawan and Eastern Mindanao), may see increased suitability when air humidity is considered.

4 Threats

Despite the areas of suitability indicated in this study, much of the Philippines is not viable for exploration due to deforestation (Figure 2). Much of the lowland forest cover has already been lost, but deforestation is starting to occur in higher altitude areas. This poses a problem for future exploration as forests which are highlighted in this study soon be removed, potentially endangering new species to science.

The main causes of forest cover reduction currently in the Philippines is slash and burn (Kaingin) and mining (Lawrence, 1997; Bravante and Holden, 2009). Slash and burn agriculture is an intensive method to convert forest cover to swidden, for use in agriculture (Lawrence, 1997). Slash and burn practices are common in all areas highlighted here to have high suitability (Lawrence, 1997). It should be noted that where communities are able to allow periods of regrowth, slash and burn techniques are less destructive to forest ecosystems (Edem, 2012). However, all slash and burn practices eventually impact the biodiversity of an area, reducing likelihood of Nepenthes populations surviving long enough to be found (Edem, 2012).

The high suitability areas found across the Philippines are all at risk from increased mining practices (Bravante and Holden, 2009). Two areas currently at risk from the impact of mining are the Visayas islands and Palawan. Records from the Republic of the Philippines Department of Environment and Natural Resources (2015) indicate that there is proposed planning for mines to be created in Eastern Samar and Leyte to extract iron and nickel. Moreover, there has been an increase of mining activity across the Palawan island as much of the ultrabasic areas are rich in rare metals such as cobalt and chromium (Cheek and Jebb, 1999). Although environmental impact assessments are included within mining proposal documents such as tree planting post operation, the assessment fails to acknowledge how biodiversity is affected (Bravante and Holden, 2009). The endemism and nature of

© The Author. All rights reserved. 14 Nepenthes species means that they often exist in large populations with restricted distributions, meaning that the removal of a population can often equate to the destruction of an entire species, including those that have yet to be identified (Bravante and Holden, 2009). As the Philippines is categorised as a developing country by the United Nations, harmful industrial practices such as mining are required to boost the country’s economy (United Nations, 2014). Therefore, mitigation in the way of more extensive and highly detailed environmental impact assessments should be undertaken before mines become operational in new areas (Bravante and Holden, 2009).

5 Accessibility Risks

The safety of these highlighted areas of high suitability should also be considered prior to exploration (Macapagal, Montiel and Canuday, 2018). Insurgence and armed conflict in some areas of the Philippines can make collecting sites extremely unsafe for field work (Macapagal, Montiel and Canuday, 2018). Unfortunately, deaths from plant expeditions in the Philippines has occurred in the past, making safety prior to collection a necessity (Cheek and Jebb, 2014). Co’s Digital Flora of the Philippines, used to provide some herbarium specimens for this study, is a memorial website for the botanist Leonardo L. Co who unfortunately died in armed conflict within the Philippines (Leyte) (Philippineplants, 2020). Therefore, it is crucial to ensure areas recommended for exploration will be safe in the future or ensure that the area can be checked for conflict prior to fieldwork.

6 Finalised areas for future fieldwork

For future exploration of Nepenthes genus, it is important to explore areas which have differing species. The endemic Nepenthes of the Philippines can be divided into two groups Sect. Alatae and Sect Micramphorae outlined in King and Cheek, in press. Sect. Alatae includes species from Luzon and Sibuyan (with the exception of N. graciliflora which occurs from Luzon to Mindanao). Sect. Micramphorae occurs from to Southern Visayas to Mindanao (King & Cheek, in press)

The first recommended area for future exploration is the central and southern Sierra Madre. Unlike its northern section, this area has far fewer collection points. Moreover, as there is evidence of some exploration indicated by two sightings of Nepenthes on GBIF, the area is likely to be accessible from settlements (Gbif, 2020). As a safety precaution the Aurora Memorial National Park is accessible to

Northern central Samar (highlighted in figure 4) would also be an area suitable for future exploration. As this area is encompassed by the SINP, and is largely unaffected by human activities, meaning the likelihood of present Nepenthes populations is high. Moreover, in terms of safety the national park can be contacted prior to collection to ensure there is no security risk in the area. Species belonging to Sect Micramphorae and N. graciliflora should be found in this area (King & Cheek, in press).

As Mindanao island has one of the highest radiation of Nepenthes diversity, there is a high likelihood of new species being discovered. As shown in Figure 2, the most promising area for Nepenthes exploration would be following the high-altitude areas between Mount Mangabon and the west of the Agusan river. It is particularly important to focus collection efforts around the lowland areas surrounding the Agusan river basin, as suitable low altitude areas are shown to be rare by the parameters of this study (Figure 2). Species belonging to Sect Micramphorae and N. graciliflora can be found in these areas.

Although Palawan is known to be another hotspot for Nepenthes species, and a main area where the Palawan clade can be found (King & Cheek, in press), accurate recommendations for future fieldwork areas cannot be provided, due to variables used in this study lacking accuracy in Palawan (Figure 2).

Limitations of project

In this study a relatively small sample size of 71 specimens was used. From these 71 species, appropriate Bioclim variable were selected which were distinct from the general climate of the Philippines (Worldclim, 2020). Both ultrabasic and non-ultrabasic dwelling species were grouped together, in order to have a large enough sample size. Investigating a larger selection of specimens with more species occurrence data will be more representative of climatic conditions for Nepenthes, and will give a better understanding of climatic factors’ influence on distribution.

This study also favoured the use of R over the programme Maxtent to analyse species histogram due to the small dataset (Phillips, Dudík and Schapire, 2020). The use of Maxtent can produce highly in-

© The Author. All rights reserved. 16 depth analysis of climatic variables and their impact on species distribution (Phillips, Dudík and Schapire, 2020). However, as this study only required general observations of the distribution of species compared to climatic variables, in-depth analysis of species against variables may create false conclusion due to the low sample size (Radosavljevic and Anderson, 2013).

Conclusion

Here we have designated areas of high suitability, in which Nepenthes populations and new species may be found. Using rainfall and temperature data from Bioclim, these areas can be observed in numerous regions across the Philippines. Despite how extensive some of these areas are, potential for future research is reduced by ongoing habitat destruction.

The immense loss of forest in the Philippines is a threat to almost all the areas highlighted in this study, with slash and burn agriculture and mining being the main causes. It is therefore important to maximise exploration efforts in outlined areas of interest to identify any Nepenthes populations and potentially new species. Conservation of these areas should be considered a priority. Some of the areas outlined in this paper may be affected by political conflict, and so national parks are recommended for future field research, given their protected status and accessibility

Although the variables used in this study are seen to be less accurate on Palawan, research should focus on including further variables that are present in the niche areas where Nepenthes can be found. Observations from this preliminary study of Nepenthes distribution within the Philippines indicate that geology and air humidity would be important factors to consider in future studies. Future SDM of Nepenthes at species level is also recommended following the findings of this study, both for the identification of new populations and species, and to highlight areas of high suitability in which in situ conservation may be appropriate.

References Almadrones-Reyes, K. and Dagamac, N., 2018. Predicting local habitat suitability in changing climate scenarios: Applying species distribution modelling for Diderma hemisphaericum. Current Research in Environmental & Applied Mycology, 8(5), pp.492-500.

Balete, D., Rickart, E., Heaney, L., Alviola, P., Duya, M., Duya, M., Sosa, T. and Jansa, S., 2012. Archboldomys(Muridae: Murinae) Reconsidered: A New Genus and Three New Species of Shrew Mice from Luzon Island, Philippines. American Museum Novitates, 3754(3754), pp.1-60.

Bauer, U. and Federle, W., 2009. The insect-trapping rim of Nepenthes pitchers. Plant Signaling & Behavior, 4(11), pp.1019-1023.

© The Author. All rights reserved. 17 Beaumont, L., Hughes, L. and Poulsen, M., 2005. Predicting species distributions: use of climatic parameters in BIOCLIM and its impact on predictions of species’ current and future distributions. Ecological Modelling, 186(2), pp.251-270.

Bohn, H. and Federle, W., 2004. Insect aquaplaning: Nepenthes pitcher plants capture prey with the peristome, a fully wettable water-lubricated anisotropic surface. Proceedings of the National Academy of Sciences, 101(39), pp.14138-14143.

Bravante, M. and Holden, W., 2009. Going through the motions: the environmental impact assessment of nonferrous metals mining projects in the Philippines. The Pacific Review, 22(4), pp.523-547.

Bravo, L., Roque, V., Brett, J., Dizon, R. and L'Azou, M., 2014. Epidemiology of Dengue Disease in the Philippines (2000–2011): A Systematic Literature Review. PLoS Neglected Tropical Diseases, 8(11), p.e3027.

Cheek, M. and Jebb, MHP., 1999. Nepenthes (Nepenthaceae) in Palawan, Philippines. Kew Bulletin, 54(4), p.887.

Cheek M, Jebb MHP. 2001. Nepenthaceae. Flora Malesiana 15. Nationaal Herbarium Nederland, 614

Leiden. 157.

Cheek M, Jebb MHP. 2013a. Recircumscription of the Nepenthes alata group (Caryophyllales: Nepenthaceae), in the Philippines with four new species. European Journal of Taxonomy 69: 1–23 http://dx.doi:10.5852/ejt.2013.69

Cheek M, Jebb MHP. 2013b. The Nepenthes micramphora (Nepenthaceae) group, with two new species from Mindanao, Philippines. Phytotaxa 151(1): 25–34 http://dx.doi.org/10.11646/phytotaxa.151.1.2

Cheek M, Jebb MHP. 2013c. Identification and Typification of Nepenthes blancoi Blume, with N. abalata Jebb & Cheek sp. nov. from the Western Visayas, Philippines. Nordic Journal of Botany 31(2): 151–156 http://dx.doi.org/10.1111/j.1756-1051.2012.00012.x

Cheek M, Jebb MHP. 2013d. Nepenthes alzapan (Nepenthaceae), a new species from Luzon, Philippines. Phytotaxa 100 (1): 57–60 http://dx.doi.org/10.11646/phytotaxa.100.1.6

Cheek M, Jebb MHP. 2013e. Nepenthes ramos (Nepenthaceae), a new species from Mindanao, Philippines. Willdenowia 43: 107–111 http://dx.doi.org/10.3372/wi.43.43112

Cheek M, Jebb MHP. 2013f. Nepenthes samar (Nepenthaceae), a new species from Samar, Philippines. Blumea 58(1): 82-84 http://dx.doi.org/10.3767/000651913X673513

Cheek M, Jebb MHP. 2013g. Typification and redelimitation of Nepenthes alata Blanco with notes on the N. alata group, and N. negros sp. nov. from the Philippines. Nordic Journal of Botany 31(5): 616– 622 http://dx.doi.org/10.1111/j.1756-1051.2012.00099.x

Cheek, M. and Jebb, M., 2014. Expansion of the Nepenthes alata group (Nepenthaceae), Philippines, and descriptions of three new species. Blumea-Biodiversity, Evolution and Biogeography of Plants, 59(2), pp.144-154.

Edem, I., 2012. Effects of Biomass Burning On Carbon Sequestration and Air Quality under Slash- And-Burn Agriculture. IOSR Journal of Agriculture and Veterinary Science, 1(2), pp.39-44.

© The Author. All rights reserved. 18 Galindon, J.M., Pasion, B., Tongco, M.D., Fidelino, J., Duya, M.R. and Ong, P., 2018. Plant diversity patterns in remnant forests and exotic tree species-based reforestation in active limestones quarries in the Luzon and Mindanao biogeographic sub-regions in the Philippines. Ecological research, 33(1), pp.63-72

Garcia, K., Lasco, R., Ines, A., Lyon, B. and Pulhin, F., 2013. Predicting geographic distribution and habitat suitability due to climate change of selected threatened forest tree species in the Philippines. Applied Geography, 44, pp.12-22.

Gbif. 2020. Nepenthes L. Available at: https://www.gbif.org/species/3190711 (Accessed 27 August 2020).

Global Forest Change. 2020. Global Forest Change | Google Crisis Map. [online] Available at: https://earthenginepartners.appspot.com/science-2013-global-forest [Accessed 27 August 2020].

Global Forest Watch. 2020. Tree cover loss in the Philippines. Available at: www.globalforestwatch.org (Accessed 27 August 2020).

Gronemeyer, T., Coritico, F., Wistuba, A., Marwinski, D., Gieray, T., Micheler, M., Mey, F. and Amoroso, V., 2014. Four New Species of Nepenthes L. (Nepenthaceae) from the Central Mountains of Mindanao, Philippines. Plants, 3(2), pp.284-303.

Guotana, J., Payot, B., Dimalanta, C., Ramos, N., Faustino-Eslava, D., Queaño, K. and Yumul, G., 2017. Petrological and geochemical characteristics of the Samar Ophiolite ultramafic section: implications on the origins of the ophiolites in Samar and Leyte islands, Philippines. International Geology Review, 60(4), pp.401-417.

Heany, L., Tabaranza, B., Rickart, D. and Balete, E., 2006. The Mammals of Mt. Kitanglad Nature Park, Mindanao, Philippines. Fieldiana Zoology, 112, pp.1-63.

Hughes, A.C., 2017. Understanding the drivers of S outheast A sian biodiversity loss. Ecosphere, 8(1), p.e01624.

Hill, S., Arnell, A., Maney, C., Butchart, S., Hilton-Taylor, C., Ciciarelli, C., Davis, C., Dinerstein, E., Purvis, A. and Burgess, N., 2019. Measuring Forest Biodiversity Status and Changes Globally. Frontiers in Forests and Global Change, 2.

Holden, W., 2012. Ecclesial Opposition to Large-Scale Mining on Samar: Neoliberalism Meets the Church of the Poor in a Wounded Land. Religions, 3(3), pp.833-861.

IUCN. 2012. IUCN Red List Categories and Criteria: Version 3.1. Second edition. Gland, Switzerland and Cambridge, UK: IUCN. Available at: http://www.iucnredlist.org/ (Accessed 26 August 2020).

JStor Global Plants. 2020. (continuously updated) Available at http://plants.jstor.org/ (Accessed 26 August 2020).

King, C. and Cheek, M., in press (Appendix 1). Nepenthes maximoides (Nepenthaceae) a new, Critically Endangered (Possibly Extinct) species in Sect. Alatae from Luzon, Philippines showing striking pitcher convergence with N. maxima (Sect. Regiae) of Indonesia. PeerJ, pp.1-15.

Lawrence, A., 1997. Kaiñgin in the Philippines: is it the end of the Forest? Network Paper-Rural Development Forestry Network, pp. 2-8.

© The Author. All rights reserved. 19 Lillo, E., Malaki, A., Alcazar, S., Redoblado, B., Diaz, J., Pinote, J., Rosales, A. and Buot Jr, I., 2020. Native trees in Nug-as forest Key Biodiversity Area, Cebu, Philippines. Biodiversitas Journal of Biological Diversity, 21(9).

Limpert, E. and Stahel, W., 2011. Problems with Using the Normal Distribution – and Ways to Improve Quality and Efficiency of Data Analysis. PLoS ONE, 6(7), p.e21403.

Linacre, E., 1982. The effect of altitude on the daily range of temperature. Journal of Climatology, 2(4), pp.375-382.

Macapagal, M., Montiel, C. and Canuday, J., 2018. The Unifying and Divisive Effects of Social Identities: Religious and Ethnopolitical Identities Among Mindanao Muslims in the Philippines. Journal of Pacific Rim Psychology, 12.

Minter, T., van der Ploeg, J., Pedrablanca, M., Sunderland, T. and Persoon, G., 2014. Limits to Indigenous Participation: The Agta and the Northern Sierra Madre Natural Park, the Philippines. Human Ecology, 42(5), pp.769-778.

Mukul, S., Herbohn, J. and Firn, J., 2016. Tropical secondary forests regenerating after shifting cultivation in the Philippines uplands are important carbon sinks. Scientific Reports, 6(1).

Murphy, B., Forest, F., Barraclough, T., Rosindell, J., Bellot, S., Cowan, R., Golos, M., Jebb, M. and Cheek, M., 2020. A phylogenomic analysis of Nepenthes (Nepenthaceae). Molecular Phylogenetics and Evolution, 144, p.106668.

Pecchi, M., Marchi, M., Burton, V., Giannetti, F., Moriondo, M., Bernetti, I., Bindi, M. and Chirici, G., 2019. Species distribution modelling to support forest management. A literature review. Ecological Modelling, 411, p.108817.

Phillips, S., Dudík, M and Schapire, R., 2020. Maxent software for modeling species niches and distributions (Version 3.4.1). Available at: http://biodiversityinformatics.amnh.org/open_source/maxent/. (Accessed 26 August 2020). Philippineplants, 2020. Nepenthaceae. [online] Available at: https://www.philippineplants.org/Families/Nepenthaceae.html (Accessed 26 August 2020).

Plants of the World Online. 2020. Nepenthes L.. [online] Available at: http://www.plantsoftheworldonline.org/taxon/urn:lsid:ipni.org:names:327014-2#children (Accessed 26 August 2020).

Poirazidis, K., Bontzorlos, V., Xofis, P., Zakkak, S., Xirouchakis, S., Grigoriadou, E., Kechagioglou, S., Gasteratos, I., Alivizatos, H. and Panagiotopoulou, M., 2019. Bioclimatic and environmental suitability models for capercaillie (Tetrao urogallus) conservation: Identification of optimal and marginal areas in Rodopi Mountain-Range National Park (Northern Greece). Global Ecology and Conservation, 17, p.e00526.

R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org/.

Radosavljevic, A. and Anderson, R., 2013. Making better Maxentmodels of species distributions: complexity, overfitting and evaluation. Journal of Biogeography, 41(4), pp.629-643.

© The Author. All rights reserved. 20 Rathore, T., Tandon, P. and Shekhawat, N., 1991. In Vitro Regeneration of Pitcher Plant (Nepenthes khasiana Hook. f.) — A rare Insectivorous Plant of India. Journal of Plant Physiology, 139(2), pp.246- 248.

Republic of the Philippines Department of Environment and Natural Resources, 2015. Leyte Mineral Profile. Mines and Geosciences Bureau, pp.1-16.

Robinson, A., Fleischmann, A., Mcpherson, S., Heinrich, V., Gironella, E. and Peña, C., 2009. A spectacular new species ofNepenthesL. (Nepenthaceae) pitcher plant from central Palawan, Philippines. Botanical Journal of the Linnean Society, 159(2), pp.195-202.

Siebert, S., 1984. Conserving Tropical Rainforests: The Case of Leyte Mountains National Park, Philippines. Mountain Research and Development, 4(3), p.272.

Spracklen, B., Kalamandeen, M., Galbraith, D., Gloor, E. and Spracklen, D., 2015. A Global Analysis of Deforestation in Moist Tropical Forest Protected Areas. PLOS ONE, 10(12), p.e0143886.

Tolentino, P., Poortinga, A., Kanamaru, H., Keesstra, S., Maroulis, J., David, C. and Ritsema, C., 2016. Projected Impact of Climate Change on Hydrological Regimes in the Philippines. PLOS ONE, 11(10), p.e0163941.

United Nations, 2014. Country Classification. Available at:

Williams, J., Seo, C., Thorne, J., Nelson, J., Erwin, S., O’Brien, J. and Schwartz, M., 2009. Using species distribution models to predict new occurrences for rare plants. Diversity and Distributions, 15(4), pp.565-576.

Worldclim. 2020. Bioclimatic Variables. Available at: https://www.worldclim.org/data/bioclim.html (Accessed 26 August 2020).

Appendix 1: N. maximoides paper

Nepenthes maximoides (Nepenthaceae) a new, Critically Endangered (Possibly Extinct) species in Sect. Alatae from Luzon, Philippines showing striking pitcher convergence with N. maxima (Sect. Regiae) of Indonesia.

Charles King¹, Martin Cheek¹

¹Science, Royal Botanic Gardens, Kew, Richmond, Surrey, U.K.

Corresponding author:

Email address: [email protected]

ABSTRACT

Nepenthes maximoides sp. nov. (Sect. Alatae) is described and assessed as Critically Endangered (Possibly Extinct) from Luzon, Philippines and appears unrecorded in 110 years. The spectacular, large, narrowly funnel-shaped upper pitchers, lids with recurved basal, and filiform apical appendages, unlike any other species in the Philippines, closely resemble those of N. maxima (Sect. Regiae) of Sulawesi-New Guinea, likely due to convergent evolution. Following recent phylogenomic analysis, sect. Alatae is divided into two, Sect. Alatae sensu stricto of Luzon to Sibuyan (including N. maximoides), and Sect. Micramphorae, expanded and recircumscribed to encompass those species of the southern Visayas, and Mindanao. A key is provided to the six species now recognised in the newly narrowly recircumscribed Sect. Alatae. The number of Nepenthes species recorded from Luzon has increased from two in 2001, to eight in 2020, all but one of which are endemic to that island, and four of which appear to be point endemics.

INTRODUCTION This paper is one in a series leading to a monograph of the genus Nepenthes building on a skeletal revision of the genus (Jebb & Cheek, 1997) and the account for Flora Malesiana (Cheek & Jebb, 2001). While in 2001 only 85 species were accepted for the genus, today the figure lies at 181 (Murphy et al., 2020). In 2001 the geographic units with highest Nepenthes species diversity were Sumatra and Borneo, each with over 30 species, while just 12 species were recorded for the Philippines (Cheek & Jebb, 2001). However since then the Philippines has dominated for discovery of new species. In 2013 alone, twelve new species were published from the Philippines (Cheek & Jebb, 2013a; Cheek & Jebb, 2013b; Cheek & Jebb, 2013c; Cheek & Jebb, 2013d; Cheek & Jebb, 2013e; Cheek & Jebb, 2013f; Cheek & Jebb, 2013g; Micheler et al. 2013), and new species continue to be added every year (Cheek & Jebb, 2014; Cheek et al., 2015; Gronemeyer et al., 2016; Lagunday et al., 2017; Amoroso et al., 2018; Robinson et al., 2019a). The current total of Nepenthes species for the Philippines is 59 (Pelser et al., 2011 onwards, accessed May 2020) now far exceeding the totals for other phytogeographic units such as Sumatra and Borneo.

However, new species of Nepenthes continue to be discovered elsewhere in SE Asia, from Indonesia, with new species described from Halmahera (Cheek, 2015), Sulawesi (Cheek & Jebb, 2016a,b) and New Guinea (Cheek et al., 2018), and also from Malaysia, with new species from Borneo (Robinson et al., 2019b; Golos et al., 2020)

© The Author. All rights reserved. 22 During a study of specimens of Nepenthes from the Philippines, one specimen, Curran s.n. (Fig. 1), previously determined as N. alata Blanco (in fact N. graciliflora Elmer (Cheek & Jebb 2013g)), was initially set aside since it seemed to have been mislabelled as from Luzon when in fact the specimen appeared to be Nepenthes maxima Nees, a widespread and variable species occurring from Sulawesi to New Guinea but absent from Philippines (Cheek & Jebb, 2001). Nepenthes maxima is distinctive for its narrowly funnel-shaped upper pitchers, flattened peristome with undulate outer margin, large, ovate-elliptic lid, a usually hooked basal appendage and a filiform apical lid appendage (Cheek & Jebb, 2001). No known species in the Philippines remotely resembles it. Nepenthes maxima (Borneo-New Guinea) is placed in Sect. Regiae Danser which is entirely absent from Philippines (Cheek & Jebb, 2015). However, a second inspection of Curran s.n. showed that features of the stem and petiole, in contrast to the pitchers, are not those of sect. Regiae Danser (Danser, 1928), but instead are consistent with those of Sect. Alatae Jebb & Cheek, which as currently defined is endemic to the Philippines, occurring from Luzon to Mindanao (Cheek & Jebb, 2015). Curran s.n. has petioles with conspicuous, patent wings (T-shaped in transverse section) while those of Sect. Regiae are inconspicuous or erect (U-shaped in transverse section); the axillary buds are inconspicuous and the indumentum of stem and leaf-blades is absent or inconspicuous (in Sect. Regiae the axillary buds are conspicuous, several mm long, and spike-like; and the indumentum is robust, branched, brown and present at least at the stem apex). In Sect. Alatae moreover, the phyllotaxy is spiral and not, as in Regiae, distichous (Danser, 1928; Cheek & Jebb, 2015). It seems clear that Curran s.n. might have been correctly labelled as a Philippine species, after all.

In this paper we key out Curran s.n., distinguishing it from all other species of Sect. Alatae in northern Philippines (Luzon and the northern Visayas) describing it as Nepenthes maximoides Cheek, diagnosing it from N. graciliflora which it had previously been identified as. We also compare it with N. copelandii Macfarl. of Mindanao which unusually also shares the narrowly funneliform upper pitcher shape. We present biographical notes on the collector and deduce the location of the geographic source of his collection, likely Mount Banahaw, and consider the likelihood of the extinction of this species. We consider the presence of terminal lid appendages in the genus and their homology; we also discuss morphological convergence with N. maxima. Finally, we discuss the division and new, narrow recircumscription of Sect. Alatae based on new phylogenomic data.

MATERIALS & METHODS

The specimen on which this paper is centred is on loan from PH and was compared with material presently at K including that on loan from A, BISH, BRIT, CAS, GH, L, NY, US and that studied at BO by the second author.

The electronic version of this article in Portable Document Format (PDF) will represent a published work according to the International Code of Nomenclature for algae, fungi, and plants (ICN), and hence the new names contained in the electronic version are effectively published under that Code from the electronic edition alone. In addition, new names contained in this work which have been issued with identifiers by IPNI will eventually be made available to the Global Names Index. The IPNI LSIDs can be resolved and the associated information viewed through any standard web browser by appending the LSID contained in this publication to the prefix "http://ipni.org/". The online version of

Herbarium citations follow Index Herbariorum (Thiers et al., 2020) and binomial authorities the International Plant Names Index (IPNI, 2020). The conservation assessment was made using the categories and criteria of IUCN (2012). Herbarium material was examined with a Leica Wild M8 dissecting binocular microscope fitted with an eyepiece graticule measuring in units of 0.025 mm at maximum magnification. The drawing was made with the same equipment using Leica 308700 camera lucida attachment. The specimen was partly unmounted to expose the characters needed to first identify and second characterise the species. It had originally been mounted with the lower surface of the lid face down, obscuring critical features. The map was made using SimpleMappr (https://www.simplemappr.net).

Information on the likely location of the collection was gathered by analysing other herbarium specimens collected by H.M. Curran recorded on JStor Global Plants (2020)

TAXONOMIC TREATMENT Key to the species of Nepenthes sect. Alatae in Luzon and northern Visayas

1 Monopodial erect shrublets 30 cm tall (including inflorescence), climbing stems and upper pitchers always absent; lower pitchers with column extending onto lower surface of the lid; lid lower surface lacking appendages. Sibuyan…..…………………….N. argentii Jebb & Cheek

1 Climbers exceeding 1 m tall (where known), climbing stems and upper pitchers present at maturity; pitchers with column usually absent, if present, not extending onto lower surface of the lid; lid lower surface with basal appendage (except N. armin)………………………….………….2

2 Upper pitchers narrowly infundibulate, broadest at the peristome; lower surface of lid with both basal and apical appendage. Luzon……………………………………….…..…..N. maximoides Cheek

2 Upper pitchers ovoid-cylindric, broadest in the basal half; lower surface of the lid lacking an apical appendage (and in N. armin a basal appendage also)………………..………………3

3 Upper pitchers lacking convex appendages on their lower surface, outer edge of peristome shallowly lobed; stem angular; male pedicels 3-4.5 mm long. Sibuyan….N. armin Jebb & Cheek

3 Upper pitchers with basal appendages on their lower surface; outer edge of peristome entire; stem terete; male pedicels > 10 mm long…………………….……………………………………..4

4 Stems glabrous, rarely glabrescent; upper pitcher lacking fringed wings; outer surface <20% covered covered in red stellate hairs or lacking them entirely...... 5

4 Stems persistently pubescent; upper pitchers with fringed wings in upper part; outer pitcher surface >50% covered in grey stellate hairs. Northern Luzon………………..…N. alata Blanco

5 Upper pitcher subcylindrical, outer surface 10-15% covered in minute red stellate hairs; lowland coastal ultramafic scrub of N & E Luzon...... N. ultra Jebb & Cheek

© The Author. All rights reserved. 24 5 Upper pitcher with ellipsoid base constricted abruptly to the narrow, cylindrical upper 2/3. outer pitcher surface lacking stellate hairs. Submontane forest of southern Luzon to Mindanao...... …….....….N. graciliflora Elmer

Nepenthes maximoides Cheek, sp. nov. - Fig. 1, Fig. 2

Differing from Nepenthes graciliflora Elmer in the upper pitchers narrowly infundibulate, widest in the distal half at the peristome (not ovoid-cylindric, widest in the proximal half), the peristome broad, flattened, and lobed on the outer edge (not narrowly cylindrical and entire on the outer edge), the lid with an asymmetrically hooked basal appendage and a filiform apical appendage (not symmetrical no-hooked, and absent, respectively). - Type: Curran s.n., Herb. Univ. Pennsylvania sheet number 70707, Academy of Natural Sciences Philadelphia sheet number 01113309 (holotype PH; isotype PNH destroyed, not seen), Philippines, Luzon, ‘Tayabas Province’ (deduced to be Mt Banahaw, Quezon Prov.) st. December 1911.

Etymology. Meaning that the species looks like Nepenthes maxima Nees, (since it looks so similar to this species that it was confused with it).

Terrestrial climber (probably), height unknown. Rosette and short stems unknown. Climbing stem rounded 8-10 mm diam; phyllotaxy spiralled; internodes 3-6 cm long; axillary buds not conspicuous; indumentum glabrous. Leaves petiolate, blades elliptic-oblong (10-)12.5-14 cm by 2.75-3.8 cm wide; apex acute; base gradually decurrent to the petiole; longitudinal nerves arising in the basal 2-2.5 cm of the blade, (2-)3 on each side of the midrib in the outer quarter of the blade, indistinct, pennate nerves more or less patent, indistinct; abaxial surface with sessile depressed-globose glands 0.02- 0.03 mm diam., c. 10 per mm2 (Fig. 2M); midrib when young densely, c.30-40% covered in erect simple hairs 0.1-0.2 mm long (Fig. 2L), margin of young-leaf blade densely hairy, hairs 1-3-armed, patent, acute, 0.06-0.25 mm long (Fig. 2N). Petiole clasping the stem by 2/5-½ its circumference, not decurrent, winged, T-shaped in section (2-)3-4 cm long, wings patent, each 2-3 mm wide, glabrous. Lower and intermediate pitchers unknown. Upper pitchers (coiled tendril, lid facing away from tendril), narrowly infundibulate or infundibulate-cylindrical in outline, 14-22.5 cm in length; 2.5-3 cm wide at base widening to 4-6.5 cm wide below the peristome, fringed wings reduced to a pair of low ridges; outer surface of pitcher glossy, drying yellowish brown, subglabrous, indumentum extremely sparse, minute and inconspicuous, hairs white bifid, 0.05 mm long, arms spreading, apices rounded. Mouth ovate, slightly concave, at 45 degrees from the horizontal, forming a column below the lid, inner surface of pitcher surface glandular, immediately below column waxy; peristome rounded, inverted U-shaped in transverse section (Fig.1K), 6-6.5 mm wide, c. 8 mm deep, ridges 0.5- 0.8 mm apart, developed as acute low ridges 0.1-(0.2) mm high (Fig. 2I), outer edge with 0-2(-3) shallow lobes (Fig. 2A), recurved and rolled (Fig. 2K), inner edge flat, held parallel to pitcher wall, the distal part with teeth inconspicuous, triangular, <0.2 mm long (Fig. 2J); column triangular, 0.8 cm by 0.5 cm (Fig. 2B). Lid elliptic or ovate-elliptic 4.8-5.8 cm by 3.5-4.5 cm, apex broadly rounded, base shallowly cordate, upper surface glabrous apart from sparse sessile red glands c. 0.05 mm diam.; lower surface densely covered (apart from the marginal 4-5 mm band) in monomorphic orbicular to

© The Author. All rights reserved. 25 slightly elliptic crater-like nectar glands, nectar glands slightly larger around base of lid, (0.15-)0.25(- 0.3) mm diameter (Fig. 2E), gradually becoming smaller towards the lid margin, 0.15(-0.2) mm long (Fig. 2D); both a basal and apical appendage arising from the midline ridge, basal appendage slightly hook-shaped, directed towards base of lid, (Fig. 2F) arising at c. 45 degrees from midline ridge, laterally flattened, oblong-arcuate, 7 mm high and 4-5 mm wide, apex rounded, 1-3 nectar glands per mm2; apical appendage filiform-cylindrical 9.5 by 0.8 mm, arising from midline 4-5 mm from lid apex, nectar glands present on midline; marginal band of lid 3-4(-5) mm wide, lacking nectar glands and with sessile red glands c. 0.05 mm diam., and in the outermost mm, short, erect, 3-5-branched hairs 125-250 μm long and broad. Spur (Fig.1B) inserted c. 4 mm below lid insertion, straight, erect, simple, 6-6.2 mm long, apex shortly bifid, surface moderately densely covered in erect 2-3-armed brown hairs, 0.05-0.08 mm long. Male, female inflorescence, infructescence and seeds all unknown.

Conservation – if our deduction is correct that the only known specimen of Nepenthes maximoides derives from Mt Banahaw (see discussion below), then there is yet some hope that the species might yet have survived extinction since the mountain and its forest are regarded as sacred by the local population. It also has a high level of formal, government protection, designated as a ‘Protected Landscape’ since 2003. Nonetheless, although its formal protection has increased since it was designated as a forest reserve in 1921, upgraded to National Park in 1941, at each stage the area has been reduced – probably reflecting the steady clearance of its forest upslope for rice cultivation, which can be seen to this day on Google Earth (viewed Feb. 2020). Barcelona et al. (2007) note that the area of Mount Banahaw has previously been affected by large-scale human disturbance, predominantly from pilgrims (it is a holy mountain), tourists and mountaineering groups. According to Birdlife International (2020), 300,000 pilgrims and hikers per annum visit the mountain, using four trails, resulting in habitat degradation. This activity which occurred for many years prior to Mount Banahaw becoming a protected landscape area in 2004, together with illegal logging, and quarrying at two sites (Birdlife International, 2020) may have significantly reduced the population of N. maximoides within this area. Should our deduction be incorrect and the location for N. maximoides be elsewhere in the province, the outlook for its survival is more negative. Myers et al. (2000) estimated that remaining primary vegetation amounts to only 3%, and Sohmer & Davis (2007) estimate extinction levels due to habitat destruction as between 9 and 28% in one representative, mainly forest genus, Psychotria L.

We here assess N. maximoides as Critically Endangered (Possibly Extinct) CR B2ab(iii), since only a single location, represented by a single specimen, is known, with threats, namely habitat degradation and destruction as reported. The area of occupancy is designated as 4 km² to comply with IUCN (2012) directions. The threats referred to above are ongoing: even at Mt Banahaw a forest fire was recently reported to have destroyed 50 Ha of the mountain’s forest (Ranada, 2014). While one can hope that this species is simply under collected and will be refound at Banahaw and other sites where the required habitat survives (as was the case in Mindanao with Nepenthes robcantleyi Cheek (Cheek, 2011)), this seem unlikely. Nepenthes are among the most charismatic plant groups in the Philippines and numerous citizen scientists and botanists have targeted them for study, resulting in numerous exciting new discoveries in the last two decades. Given this, it seems improbable that N. maximoides, the largest and most spectacular Nepenthes in Luzon, the most heavily populated island of the Philippines should not have been refound in 110 years if it actually

© The Author. All rights reserved. 26 survives. Should the species be refound, measures should be taken to determine the size of the population and the state of regeneration, and seed collected as a priority to enable mass in vitro propagation for the Nepenthes collector market in order to reduce the certain pressure of poaching that would otherwise result from rediscovery.

DISCUSSION

The top set of this collection was probably deposited at PNH but, with the entire herbarium, destroyed there with much of the city in February 1945 during the battle of Manila between the forces of Japan and U.S.A (Scott, 2019). Studies of specimens either requested on loan from or studied during visits at USA herbaria known to have Philippine specimens, namely AA, BISH, BRIT, CAS, GH, NY, PH, US, have failed to find either additional duplicates, or additional collections of this species. Curran s.n. had been determined (undated) as Nepenthes alata Blanco by Macfarlane, the last monographer of the genus. He had a broad concept of this species and included within it N. graciliflora, and other species, which are now widely recognised (Macfarlane, 1908). Macfarlane’s determination was adopted by Sutton in 1995. However, both those botanists seem to have only looked at this specimen superficially because the original mountings had remained in place. Because the pitchers are mounted face down, it is only by demounting the specimen that most of the diagnostic characters, such as the lid appendages, can be seen.

With the description in this paper of Nepenthes maximoides, the number of Nepenthes species accepted from Luzon has increased to eight, from two in 2001 (Cheek & Jebb, 2001). All but one of these species are endemic to that island, and four, including Nepenthes maximoides, appear to be point endemics on current evidence.

Within the Philippines, Nepenthes maximoides is most similar in pitcher shape to N.copelandii of Mt Apo, Mindanao. This is because both species have narrowly infundibulate upper pitchers, an unusual character in the Philippines. However, N. copelandii has dimorphic nectar glands on the lower surface of the lid, 2-flowered partial-peduncles and hairy stems (not monomorphic and glabrous respectively), also the basal lid appendage of the upper pitchers is only obtuse and is not hooked, and not asymmetric, nor is there an apical appendage. Finally, the peristome of the upper pitcher in N. copelandii is not flattened, and lobed on its outer edge as in N. maximoides, and the inner edge is minutely toothed, not with teeth inconspicuous as in N. maximoides.

Hugh McCollum (or McCullam or McCullom) Curran (1875-1960) and his Philippine collections

The collector of the type and only known specimen of Nepenthes maximoides is stated on the printed specimen label to be H.M. Curran. Searching under ‘Curran’ on the IPNI (continuously updated) database gives only two people of this name, both American, only one of which appears to have been active in the Philippines. This is Hugh McCollum Curran (1875-1960) listed on IPNI as a collector, not an author. The other, active in U.S.A., was Mary Katharine Curran (1844-1920). Hugh Curran was a prolific collector and a search of IPNI (continuously updated) produced 84 species names with the epithet “curranii” most of which are from the Philippines and all of which appear to commemorate H.M. Curran, probably because he collected the type specimen. According to his

© The Author. All rights reserved. 27 Wikipedia page (Anon., continuously updated) he was trained in N. Carolina and then at Yale Universities before spending 1906-1912 in the Philippines as a forester with the Forestry Bureau of Manila. Thereafter he visited South America for several years but returned to Philippines 1929-41 as Professor of Forestry, survived internment there in the second world war, afterwards returning to Venezuela.

Analysis of specimens under the name Curran on JStor Global Plants (2020) which is dominated by type specimens, returns 1,087 items, about half of which are specimens collected by H.M. Curran, and most of these, 534, are from Philippines. H.M. Curran collected specimens under the Forest Bureau of Manila series. His earliest collections, from 1906, are from Palawan and include Curran 3891, the holotype of Nepenthes deaniana Macf. (destroyed with the Manila herbarium, PNH in 1945), and the lectotype of N. philippinensis Macf., Curran 3896 (Cheek & Jebb, 2000). While his Palawan collections often had notes giving written descriptions of the specimens with altitudes and locations (e.g. Curran 3473) those collected in subsequent years tended to lack these details and to consist of only printed labels, often only with the province, month and year: as for the specimen of N. maximoides. However, sometimes the name of a mountain or settlement might be given. It is unusual that no Forest Bureau number is assigned to a specimen as in the case of this type specimen. We speculate that this might be because the specimen is sterile. Even today, in the 21st century, collecting sterile specimens for herbaria is often considered dubious practice and such specimens, collected because they excite interest in the collector, are often sadly left unnumbered (M. Cheek pers. obs. 1991-2020). Such sterile specimens are often collected only as unicates, when the collector might otherwise have a system of collecting large sets. This would help explain why only a single sheet is known of this specimen. However, in any case, Curran usually appeared to collect only small sets – inspection of the JStor Global Plants data (continuously updated), which details herbaria at which duplicates of type specimens collected by him are known, suggests it is rare that more than two duplicates are known of any of his specimens outside of the Philippines. Van Steenis-Kruseman & van Steenis (1950: 123-124) state that of Curran’s collections “Many plants in Herb. Manila, at least partially numbered in the F.B. series” yet this may have been written before the destruction in 1945 or in ignorance of it. Today none are known to survive there.

Tayabas Province, given as location of the type specimen collected by Curran (no further details are given), has been renamed and divided into two: Aurora Province to the north, along the eastern coast of Luzon comprising the narrow coastal plain and the seaward edge of the N-S Sierra Madre range, and in the south Quezon Province – the S.E. corner of Luzon and an adjoining part of the Bicol Peninsula. We can only deduce where within this range the type specimen of N. maximoides might have been collected. We can be moderately certain that is was in forest on a mountain above 1000 m. alt., since all but one of the other seven species of Nepenthes on Luzon occur in such habitats (Cheek & Jebb, 2013a; Cheek & Jebb, 2013c; Cheek & Jebb, 2013d; Cheek & Jebb, 2013g; Cheek et al., 2015). The exception is Nepenthes ultra Jebb & Cheek, restricted to ultramafic habitat just above sea-level (Cheek & Jebb, 2013h).

Eight mountains in Quezon and Aurora (former Tayabas) meet this specification (see Table 1, Fig. 3). Most of these (six of the eight) are clustered in two of four areas (Fig. 3). However, we contend that

© The Author. All rights reserved. 28 the most likely of these to have furnished the type specimen is the Mt Banahaw complex which apart from Mt Banahaw itself include the mountains of San Cristobal and Banahao de Lucuban.

Mt Banahaw is also spelled as Banhao, Banàhao, or as Banajao, and in the Spanish colonial period was known as as Monte de Majayjay or Monte San Cristobal. Of the 13 specimens detailed on JStor as collected by H.M.Curran from Tayabas Province, six are given as from Mt Banahaw. The remaining seven specimens either have no further locality data or give Municipality Macuban or Paete-Piapi, lowland settlements to the NE of Mt Banahaw. Mount Banahaw lies on the boundary between Tayabas (now Quezon) and Laguna Provinces. Of the 42 H.M. Curran specimens recorded from Laguna Province on JStor Global Plants (2020), 12 are also from Mt Banahaw. Therefore, on the basis of the JStor data, Mt Banahaw is the only collection location in former Tayabas Province that was a Curran location where Nepenthes might have been expected to be found (above 1000 m alt.). There is no evidence that any other mountain in Tayabas Province where Nepenthes might be found was visited by Curran, although this is possible. Moreover, Mt Banahaw was clearly a target for Curran since he collected there also while in Laguna Province. On the balance of probability, it seems likely, though not certain that the type and only specimen of N. maximoides derives from Mt Banahaw.

At the time the specimen was collected in 1911, the then provincial capital, Tayabas, was still a major administrative centre, giving its name to the former province. It is situated on the SE slopes of Mount Banahaw. The possible reasons for H.M. Curran selecting Banahaw are that it would then have been 1) the most readily accessible area of forest in the province from Curran’s base near Manila, 2) it was and remains the largest block of surviving forest in the province and 3) he had visited it on a productive earlier visit: Curran 3039 collected with M.L. Merritt, is syntype of Ahernia glandulosa Merr. (Flacourtiaceae, now Achariaceae) – both a new species and genus, collected from “Mt Banajao, Tayabas Province” on 1 Nov. 1907 (JStor Global Plants, 2020).

Mount Banahaw

At 2158 m high Mount Banahaw is the highest of a group of volcanoes south and east of Manila. Banahaw is flanked by the less high and more recent San Cristobal volcano on the west and Banahaw de Lucban on the NE. Andesitic-to-dacitic lava domes occur on the flanks of Banahaw and San Cristobal. The summit crater is about 2 km wide and 300 m deep. The last eruption is thought to have been in 1909 but this is uncertain (Global Volcano Program, continuously updated). Hot springs are present at several sites. The slopes of the mountain are completely forested apart from the lower altitudes which have largely been cleared from sea-level upwards to about 700 m alt. (see Conservation above). This complex contains c. 100 km² of forest above c. 500 m alt. There is no checklist of the plant species, but trees above 10 cm diameter at breast height have been characterised from 25 20 m x 20 m plots placed along a transect from 700 m alt. to the summit, which recorded 455 stems and 92 species (Gascon et al., 2013). Rainfall varies from 2350-2400 mm p.a. on the NW slope to 4470 m p.a. on the NE slope spread evenly over the year with 262 rainy days p.a. (Gascon et al., 2013). Details of the history of the protected status of the mountain are given under conservation (above). Currently it is part of the 10, 900 ha Mounts Banahaw–San Cristobal Protected Landscape. In describing Rafflesia banahaw Barcelona, Pelser & Cajano which is now

© The Author. All rights reserved. 29 considered a synonym of R. philippensis Blanco (Barcelona et al., 2009)), Barcelona et al. (Barcelona et al., 2007) document nine botanists who made collections at Mount Banahaw, some who visited on more than one occasion. To these can be added the collector Azaola who obtained at this location on 22 April 1840 original material of Rafflesia lagascae Blanco (Blanco, 1845; Pelser et al., 2013) and who appears to have been the first collector of plant specimens at Mount Banahaw. Additional to these ten, van Steenis-Kruseman & van Steenis (1950) also list: W. Kerr (1805), C. Wilkes (1842), N.J. Andersson (1853), A. Marche (1880) and E. Langlassé (1895) as having visited Mt. Banahaw. So, it might be that Kerr rather than Azaola was the first western botanist to collect on the mountain. In this paper we add Curran (see above). Mount Banahaw is therefore one of the most botanically visited of locations in the Philippines. However, it is possible that important parts of the mountain, with intact vegetation, have yet to be thoroughly surveyed for plant species, and perhaps Nepenthes maximoides might be found in one of these places.

To the best of our knowledge, no other Nepenthes species is known from Mt. Banahaw.

Recircumscription of Sect. Alatae

The Nepenthes alata group was first designated (Cheek & Jebb, 2013a) to include all species occurring from Luzon to Mindanao (excluding Palawan) excepting those of Sect. Insignes Danser, and also excluding the only Philippine non-endemic species, N. mirabilis (Lour.) Druce.

Subsequently the N. micramphora group was designated (Cheek & Jebb, 2013b) for three species from Mindanao that lacked key attributes of the N. alata group as then defined – particularly the basal lid appendage and a distinct petiole. These two groups were later formalised as Sect. Alatae Cheek & Jebb and Sect. Micramphorae Cheek & Jebb respectively (Cheek & Jebb, 2015).

However, a recent near-comprehensive species-level phylogenomic study of Nepenthes revealed that Sect. Alatae was not monophyletic (Murphy et al., 2020). The northern Sect. Alatae species, i.e. of Luzon and Sibuyan, are sister to the species of Palawan, previously not considered to be closely related. Unexpectedly, the southern Alatae species i.e. of the southern Visayas (Negros, Leyte and Mindanao (excluding N. graciliflora Elmer which extends from Luzon), including the Micramphorae, are sister to the Palawan clade & the northern Alatae. The division of the northern from the southern Alatae revealed by phylogenomic analysis is supported by morphology as seen in the primary division in the most recent key to the group (Cheek & Jebb, 2014). The species of the north have monomorphic, uniformly small, moderately crater-like nectar glands evenly spread on the lower surface of the lid. In contrast, the species of the south usually have dimorphic glands in different size-classes and always have at least some glands very much larger than those seen in the northern species. While the northern species have 1-flowered partial-peduncles, those of the south have 2-flowered partial peduncles (where known).

The separation between the two redelimited sections can be summarised in key form as follows:

Lower surface of lid, including basal appendage (if present), densely and evenly covered in uniformly minute circular or shortly elliptic nectar glands (0.15–0.25(–0.3) mm diam.); inflorescence partial- peduncles 1-flowered. Luzon & Sibuyan (except N. graciliflora Luzon to Mindanao)……………………………………………………………………………..Sect. Alatae

Lower surface of lid with nectar glands either absent from the appendage and/or, sparse, large or dimorphic (larger glands 0.35–0.4 mm diam. or larger); inflorescence partial-peduncles 2-flowered. Southern Visayas & Mindanao…………………………………………..… Sect. Micramphorae

Nepenthes argentii Jebb & Cheek of Sibuyan was formerly unplaced in a species group (Cheek & Jebb, 2001). Due to the shortly cylindrical pitchers, a broad peristome extending as a column to the lower lid surface and with blade-like peristome ridges it has similarities with species of the Palawan group. But it is embedded in the northern Alatae clade on phylogenomic data (Murphy et al., 2020). This unexpected placement is morphologically supported by its lid nectar gland and inflorescence morphology: it has uniform, minute nectar glands on the lower surface of the lid and 1-flowered partial-peduncles. Since the type species of Sect. Alatae is N. alata Blanco of Luzon, the northern species must retain this sectional name. The only available sectional name for the southern species of Sect. Alatae is Sect. Micramphorae, so this name must perforce now be adopted for the ‘southern Alatae’ which comprise the most species-diverse of the Nepenthes clades of the Philippines. A key to Sect. Alatae sensu stricto (as here recircumscribed), informally referred to as the ‘northern Alatae’ above, is presented in the results (above). It now consists of six species, all but one of which occur on Luzon, with N. graciliflora extending from Luzon to Mindanao, and N. armin being restricted to Sibuyan.

The apical lid appendage in Nepenthes

Nepenthes maximoides is highly remarkable and among all the known species of the genus in the Philippines unique, in its filiform apical appendage (Fig. 2G). This structure also occurs in several species of sect. Regiae (Borneo to New Guinea), where it appears homologous – the appendage appears to be a continuation of the distal terminus of the midline of the lid. This midline is thickened, raised above the adjoining tissue as a low ridge, and appears to be highly vascularised. In Sect. Regiae the apical appendage often carries the largest nectar glands present on the lower surface of the lid (Cheek & Jebb, 2001), and is also often raised above the surface as a ridge for part of its length: see e.g. Nepenthes minima Jebb & Cheek (Cheek & Jebb, 2016b). Lid appendages also occur in some species of sect. Montanae Danser, but these appear likely to be non-homologous, arising differently, as lobes from the midline far before its distal terminus e.g. N. lingulata Chi C. Lee, Hernawati & Akhriadi (Lee et al,. 2006).

© The Author. All rights reserved. 31 The earliest clade in the most comprehensive phylogenetic tree of Nepenthes (Murphy et al., 2020) in which an apical lid appendage is developed is in the Nepenthes danseri group, as seen in N. weda Cheek (Cheek, 2015). In this species all stages of the pitchers were available for study and there appears to be stage dependent heteromorphy of the distal lid appendage. In rosette pitchers the appendage is a transversely crescent-shaped ridge, while in upper pitchers it protrudes further from the surface and is more elaborate but not filiform (Cheek, 2015: 225). While potentially proto- filiform appendages such as these are unknown in Sect. Alatae as now delimited, they do occur in Sect. Micramphorae in N. robcantleyi Cheek (Cheek, 2011) and in N. tboli Jebb & Cheek (Cheek & Jebb, 2014).

Convergence of pitcher morphology

The shared pitcher morphology of Nepenthes maximoides with that of Nepenthes maxima appears due to convergence, not due to phylogenetic proximity. Convergence of pitcher morphology is recorded in other cases in the genus (Thorogood et al., 2018). Recently, 12 functional pitcher types have been recognised, each postulated to target capture of nutrients from animals (sometimes plants) in a different manner (Cheek et al., 2020a). Nepenthes maxima, and Nepenthes maximoides fall under pitcher type 2 (“narrow-funnel” Cheek et al., 2020a).

The dramatic rise in the numbers of Philippine species of Nepenthes in the 21st century (see introduction) is mirrored in other plant groups such as Rafflesia R.Br. (Rafflesiaceae). Before 2002 only two species of Rafflesia were thought to be known from the Philippines (subsequently two additional, long-overlooked species came to light), and, as in Nepenthes, the genus was thought to be most diverse in Borneo and Sumatra. Intensive fieldwork in remaining patches of forest in the Philippines, however, has raised species numbers steadily from two species in 2002 to 13 species in 2019, and Philippines now is the most species-diverse country for Rafflesia globally (Barcelona et al., 2007; 2009; Pelser et al., 2013; 2019).

The number of flowering plant species known to science is disputed (Nic Lughadha et al., 2017), but a reasonable estimate is 369 000 (Nic Lughadha et al., 2016), while the number of species described as new to science has been at about 2000 per annum for at least 10 years (Cheek et al., 2020b). The conservation status of 21–26% of plant species has been established using evidence-based assessments, and 30–44% of these rate the species assessed as threatened, while only c. 5% of plant species have been assessed using the IUCN (2012) standard (Bachman et al., 2018). Newly discovered species such as Nepenthes maximoides, are likely to be threatened, since widespread species tend to have been already discovered and it is the more localised, rarer species that remain to be found although there are exceptions such as Gouania longipedunculata Cahen, Stenn & Utteridge (Cahen et al., 2020) which is widespread. This makes it urgent to discover and protect such localised species before they become extinct due to habitat clearance as was the case with Nepenthes extincta Jebb & Cheek (Cheek, 2013a). However, it may be too late for Nepenthes maximoides, which may be extinct already, although efforts to rediscover it should be made in case not.

Acknowledgements.

The authors thank Dr Laurence Dorr (US) for facilitating the study visit of Martin Cheek to PH (at the Academy of Natural Sciences of Drexel University, Philadelphia, incorporating the Pennsylvania State University herbarium ) and to that Institute for providing on loan to K of the material for the basis for the description in this paper of Nepenthes maximoides. Janis Shillito typed up much of the manuscript. We give praise and thanks to two reviewers Pieter Pelser and Victor Amoroso for their painstaking, valuable and constructive reviews of an earlier version of this paper.

We thank the staff at the Tropical Nursery, RBG Kew for cultivating so expertly over many years the species Nepenthes collection used in this paper for comparison purposes, especially Tom Pickering, Rebecca Hilgenhof, Nick Johnson, James Beattie, Lara Jewitt and Kath King.

REFERENCES

Amoroso VB, Lagunday NE, Coritico FP, Colong RD. 2018. Nepenthes alfredoi (Caryophyllales, Nepenthaceae), A New Species of Pitcher Plant from Mindanao, Philippines. Philippine Journal Systematic Biology 11(2):14-9 https://doi.org/10.26757/pjsb.2017b11018

Anon. 2020 onwards. Hugh McCollum Curran. Available at https://en.wikipedia.org/wiki/Hugh_McCollum_Curran (accessed 14 June 2020)

Bachman SP, Nic Lughadha EM, Rivers MC. 2018. Quantifying progress towards a conservation assessment for all plants. Conservation Biology 32(3):516-524. https://doi.org/10.1111/cobi.13071

Barcelona JF, Pelser PB, Cajano MO. 2007. Rafflesia banahaw (Rafflesiaceae), a new species from Luzon, Philippines. Blumea 52(2):345-350. https://doi.org/10.3767/000651907x609089

Barcelona JF, Pelser PB, Balete DS, Co LL. 2009. Taxonomy, ecology, and conservation status of Philippine Rafflesia (Rafflesiaceae). Blumea 54(1-2):77- 93. https://doi.org/10.3767/000651909x474122

BirdLife International. 2020. Important Bird Areas factsheet: Mounts Banahaw-San Cristobal National Park. Available at http://www.birdlife.org (accessed 14 June 2020)

Blanco FM. 1845. Flora de Filipinas. segun el sistema sexual de Linneo: por el P. Fr. Manuel Blanco. Segunda Impression. Manila: M. Sanchez.

Cahen D, Stenn KS, Utteridge TM. 2020. A revision of the genus Gouania (Rhamnaceae) in the Philippines and Sundaland. Kew Bulletin, 75:1-24. https://doi.org/10.1007/s12225-020-9867-5

Cheek M. 2011. Nepenthes robcantleyi sp. nov. (Nepenthaceae) from Mindanao, Philippines. Nordic Journal Botany 29: 1-5 https://doi.org/10.1111/j.1756-1051.2011.01449.x

Cheek M. 2015. Nepenthes (Nepenthaceae) of Halmahera, Indonesia. Blumea 59(3): 215–225 http://dx.doi.org/10.3767/000651915X689091

Cheek M, Jebb M. 2000. Nepenthes (Nepenthaceae) in Palawan, Philippines. Kew Bulletin 54(4): 887- 895 https://doi.org/10.2307/4111166

Cheek M, Jebb MHP. 2013b. The Nepenthes micramphora (Nepenthaceae) group, with two new species from Mindanao, Philippines. Phytotaxa 151(1): 25–34 http://dx.doi.org/10.11646/phytotaxa.151.1.2

Cheek M, Jebb MHP. 2013d. Nepenthes alzapan (Nepenthaceae), a new species from Luzon, Philippines. Phytotaxa 100 (1): 57–60 http://dx.doi.org/10.11646/phytotaxa.100.1.6

Cheek M, Jebb MHP. 2013e. Nepenthes ramos (Nepenthaceae), a new species from Mindanao, Philippines. Willdenowia 43: 107–111 http://dx.doi.org/10.3372/wi.43.43112

Cheek M, Jebb MHP. 2013f. Nepenthes samar (Nepenthaceae), a new species from Samar, Philippines. Blumea 58(1): 82-84 http://dx.doi.org/10.3767/000651913X673513

Cheek M, Jebb MHP. 2013h. Nepenthes ultra (Nepenthaceae), a new species from Luzon, Philippines. Blumea 58(1): 241–244 http://dx.doi.org/103767/000651913X675124

Cheek M, Jebb MHP. 2014. Expansion of the Nepenthes alata group (Nepenthaceae), Philippines, and descriptions of three new species. Blumea 59(2): 144-154. https://doi.org/10.3767/000651914x685861

Cheek M, Jebb MHP. 2015. Nepenthes, Three new Infrageneric names and lectotypifications. Planta Carnivora 37(2): 34-41

Cheek M, Jebb MHP. 2016a. A new section in Nepenthes (Nepenthaceae) and a new species from Sulawesi. Blumea 61:59-62 http://dx.doi.org/10.3767/000651916X691510.

Cheek M, Jebb MHP. 2016b. Nepenthes minima (Nepenthaceae), a new pyrophytic grassland species from Sulawesi, Indonesia. Blumea 61(3):181–185 https://doi.org/10.3767/000651916X693509

Cheek M, Jebb MHP, Murphy B, Mambor F. 2018. Nepenthes section Insignes in Indonesia, with two new species. Blumea 62:174-178. https://doi.org/10.3767/blumea.2018.62.03.03

Cheek M, Jebb M., Murphy B. 2020a. A classification of functional pitcher types in Nepenthes (Nepenthaceae). Planta Carnivora 40(2):22-43 https://doi.org/10.1101/852137

Cheek M, Nic Lughadha E, Kirk P, Lindon H, Carretero J, Looney B, Douglas B, Haelewaters D, Gaya E, Llewellyn T, Ainsworth M, Gafforov Y, Hyde K, Crous P, Hughes M, Walker BE, Forzza RC, Wong KM, Niskanen T. 2020b. New discoveries: plants and fungi. Plants, People Planet (in press).

© The Author. All rights reserved. 34 Cheek M, Tandang DN, Pelser PB. 2015. Nepenthes barcelonae (Nepenthaceae), a new species from Luzon, Philippines. Phytotaxa 222(2): http://dx.doi.org/10.11646/phytotaxa.222.2.7

Danser BH. 1928. The Nepenthaceae of the Netherlands Indies. Bulletin Jardin Botanique Buitenzorg III, 9:249-438. Gascon CN, Garcia RC, Beltran FN, Faller WC, Agudilla MAR. 2013. Biodiversity Assessment of Mt. Banahaw de Dolores. Asian Journal of Biodiversity 4(1):23-45 https://doi.org/10.7828/ajob.v4i1.295

Global Volcano Program. 2020. Banahaw. Smithsonian Institution. Museum of Natural History. Available at https://volcano.si.edu/volcano.cfm?vn=273050#:~:text=Geological%20Summary,of%20Banahaw%2 0and%20San%20Cristobal. (accessed 14 June 2020)

Golos MR, Robinson AS, Barer M, Dančák M, De Witte J, Limberg A, Noorhana MS, Tjiasmanto W. 2020. Nepenthes fractiflexa (Nepenthaceae), a new Bornean pitcher plant exhibiting concaulescent metatopy and a high degree of axillary bud activation. Phytotaxa 432: 125-143. http://dx.doi.org/10.11646/phytotaxa.432.2.3

Gronemeyer J, Suarez W, Nuytemans H, Calamaro M, Wistuba, A, Mey FS, Amoroso VB. 2016. Two new Nepenthes from the Philippines and an emended description of Nepenthes ramos. Plants 5(2):23 https://doi.org/10.3390/plants5020023

IPNI. 2020. International Plant Names Index. The Royal Botanic Gardens, Kew, Harvard University Herbaria & Libraries and Australian National Botanic Gardens. Available at http://www.ipni.org, (accessed 05 June 2020).

IUCN. 2012. IUCN Red List Categories and Criteria: Version 3.1. Second edition. Gland, Switzerland and Cambridge, UK: IUCN. Available at: http://www.iucnredlist.org/ (accessed: 01/2020).

Jebb M, Cheek M. 1997. A Skeletal Revision of Nepenthes (Nepenthaceae). Blumea 42:1-106. JStor Global Plants. 2020. (continuously updated) Available at http://plants.jstor.org/ (accessed 14 June 2020)

Lagunday NE, Acma FM, Cabana VG, Sabas NM, Amoroso VB. 2017. Two New Nepenthes Species from the Unexplored Mountains of Central Mindanao, Philippines. Philippine Journal of Science 146(2):159-65.

Lee C, Hernawati, Akhriadi P. 2006. Two new species of Nepenthes (Nepenthaceae) from northern Sumatra. Blumea 51(3):561-568. https://doi.org/10.3767/000651906x622120

Macfarlane JM. 1908. Nepenthaceae. In: A. Engler, Das Pflanzenreich Heft 36, 4, 3: 1–92. Micheler M, Gronemeyer T, Wistuba A, Marwinski D, Suarez W, Amoroso V. 2013. Nepenthes viridis, eine neue Nepenthes-Art von der Insel Dinagat, Philippinen. Das Taublatt 76: 4–21. Murphy B, Forest F, Barraclough T, Rosindell J, Bellot S, Cowan S Golos M, Jebb M, Cheek M. 2020. A phylogenomic analysis of Nepenthes (Nepenthaceae). Molecular Phylogenetics and Evolution 144. https://doi.org/10.1016/j.ympev.2019.106668

Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GAB, Kent J. 2000. Biodiversity Hotspots for Conservation Priorities. Nature 403: 853–858. https://doi.org/10.1038/35002501

© The Author. All rights reserved. 35 Nic Lughadha E, Bachman SP, Govaerts R. 2017. Plant Fates and States: Response to Pimm and Raven. Trends in Ecology & Evolution 32: 887–889. https://doi.org/10.1016/j.tree.2017.09.005

Nic Lughadha E, Govaerts R, Belyaeva I., et al. 2016. Counting counts: Revised estimates of numbers of accepted species of flowering plants, seed plants, vascular plants and land plants with a review of other recent estimates. Phytotaxa 272:82–88. https://doi.org/10.11646/phytotaxa.272.1.5

Pelser PB, JF Barcelona, Nickrent DL. (eds.). 2011. onwards. Co's Digital Flora of the Philippines. Available at www.philippineplants.org (accessed on 14 June 2020)

Pelser PB, Nickrent DL, Callado JR, Barcelona JF. 2013. Mt. Banahaw reveals: The resurrection and neotypification of the name Rafflesia lagascae (Rafflesiaceae) and clues to the dispersal of Rafflesia seeds. Phytotaxa 131(1):35-40. https://doi.org/10.11646/phytotaxa.131.1.6

Pelser PB, Nickrent DL, van Ee BW, Barcelona JF. 2019. A phylogenetic and biogeographic study of Rafflesia (Rafflesiaceae) in the Philippines: Limited dispersal and high island endemism. Molecular phylogenetics and evolution 139:106555. https://doi.org/10.1016/j.ympev.2019.106555

Ranada P. 2014. Mount Banahaw Fire Burns 50 ha of Mountain Forest. Rappler. Available at https://www.rappler.com/science-nature/environment/53460-mt-banahaw-fire-burns-hectares- forest (accessed 13 June 2020)

Robinson AS, Zamudio SG, Caballero RB. 2019a. Nepenthes erucoides (Nepenthaceae), an ultramaficolous micro-endemic from Dinagat Islands Province, northern Mindanao, Philippines. Phytotaxa 423(1):21-32 https://doi.org/10.11646/phytotaxa.423.1.3

Robinson AS, Golos MR, Barer M, Sano Y, Forgie JJ, Garrido D, Gorman CN, Luick AO, McIntosh NWR, McPherson SR, Palena GJ, Pančo I, Quinn BD & Shea J. 2019b. Revisions in Nepenthes following explorations of the Kemul Massif and the surrounding region in north-central Kalimantan, Borneo. Phytotaxa 392 (2): 97–126. https://doi.org/10.11646/phytotaxa.392.2.1

Scott JM. 2019. Rampage: MacArthur, Yamashita, and the Battle of Manila. New York: W.W. Norton & Co.

Sohmer SH, Davis AP. 2007. The Genus Psychotria (Rubiaceae) in the Philippine Archipelago. Sida, Botanical Miscellany, No. 27. Fort Worth: Botanical Research Institute of Texas. 247.

Thiers B. 2020. continuously updated. Index Herbariorum: A global directory of public herbaria and associated staff. New York Botanical Garden's Virtual Herbarium. Available at http://sweetgum.nybg.org/ih/ (accessed June 2020).

Thorogood CJ, Bauer U, Hiscock SJ. 2018. Convergent and divergent evolution in carnivorous pitcher plant traps. New Phytologist, 217(3):1035-1041 https://doi.org/10.1111/nph.14879

van Steenis-Kruseman MJ, van Steenis CGGJ. 1950. Malaysian plant collectors and collections being a Cyclopaedia of botanical exploration in Malaysia and a guide to the concerned literature up to the year 1950. Flora Malesiana - Series 1, Spermatophyta, 1(1), 2–639.

Many thanks for submitting your manuscript to PeerJ and the careful consideration of the comments, in particular including the comparison with N. copelandii which strengthened the paper. I

am therefore delighted to recommend that the paper be accepted.

[# PeerJ Staff Note - this decision was reviewed and approved by Patricia Gandini, a PeerJ Section Editor covering this Section #]

Appendix 2 Specimen list

Nepenthes abalata Nepenthes abgracilis Nepenthes aenigma Nepenthes alata Nepenthes alfredoi Nepenthes alzapan Nepenthes argentii Nepenthes armin Nepenthes attenboroughii Nepenthes barcelonae Nepenthes bellii Nepenthes burkei Nepenthes cabanae Nepenthes ceciliae Nepenthes cid Nepenthes copelandii Nepenthes cornuta Nepenthes deaniana Nepenthes erucoides Nepenthes extincta Nepenthes gantungensis Nepenthes graciliflora Nepenthes hamiguitanensis Nepenthes justinae Nepenthes kitanglad Nepenthes kurata Nepenthes leonardoi Nepenthes leyte Nepenthes malimumuensis Nepenthes manobo Nepenthes mantalingajanensis Nepenthes merrilliana Nepenthes micramphora Nepenthes mindanoaoensis Nepenthes mira Nepenthes mirabilis Nepenthes nebularum Nepenthes negros Nepenthes palawanensis

Nepenthes peltata Nepenthes petiolata Nepenthes philippinensis Nepenthes pulchra Nepenthes ramos Nepenthes robcantleyi Nepenthes samar Nepenthes saranganiensis Nepenthes sibuyanensis Nepenthes sumagaya Nepenthes surigaoensis Nepenthes talaandig Nepenthes tboli Nepenthes truncata Nepenthes ultra Nepenthes ventricosa Nepenthes viridis Nepenthes zygon

Appendix 3: Climate types

Following the Coronas classification outlined in Tolentino et al. (2016), the Philippines can be divided into 4 climates. Type I climates have a distinct dry period from November to April, with a wet season from May to October. Type II has rainfall evenly distributed throughout the year, with notable rainfall from November to January. Type III climates have less notable seasons with a distinct dry season from November to April. Type IV has rainfall which is evenly distributed throughout the year.

Appendix 4: R Markdown library('raster') library(rgdal) library(ggplot2) library(tidyr)

#grab data

#forestcover note in file fc1 <- raster('C:/Users/Charlie/Documents/data/Hansen_GFC-2019- v1.7_treecover2000_10N_110E.tif') fc2 <- raster('C:/Users/Charlie/Documents/data/Hansen_GFC-2019- v1.7_treecover2000_10N_110E.tif')

© The Author. All rights reserved. 38 fc3 <- raster('C:/Users/Charlie/Documents/data/Hansen_GFC-2019- v1.7_treecover2000_20N_110E.tif') fc4 <- raster('C:/Users/Charlie/Documents/data/Hansen_GFC-2019- v1.7_treecover2000_20N_120E.tif')

#worldclim

climate1 <- getData('worldclim', var='bio',res=0.5,lon=120,lat=1)

climate2 <- getData('worldclim', var='bio',res=0.5,lon=90,lat=1)

#rename names to be same

names(climate1) <- c("bio1","bio2","bio3","bio4","bio5","bio6","bio7","bio8","bio9","bio10","bio11","bio12","bio13","bi o14","bio15","bio16","bio17","bio18","bio19") names(climate2) <- c("bio1","bio2","bio3","bio4","bio5","bio6","bio7","bio8","bio9","bio10","bio11","bio12","bio13","bi o14","bio15","bio16","bio17","bio18","bio19")

##extent

e <- extent(118,130,6,19)

#clip/crop

climate1 <- crop(climate1,e) climate2 <- crop(climate2,e)

#merge into one climate <- merge(climate1,climate2)

#release a little memory rm(climate1,climate2)

#rename again, not sure why it does not conserve the names!!!

names(climate) <- c("bio1","bio2","bio3","bio4","bio5","bio6","bio7","bio8","bio9","bio10","bio11","bio12","bio13","bi o14","bio15","bio16","bio17","bio18","bio19")

#sort out forest cover

#crop

#merge fc <- merge (merge (fc1,fc2),merge (fc3,fc4))

#clear memory of old rm(fc1,fc2,fc3,fc4)

#resample to worldclim resolution etc fc1km <- resample(fc,climate$bio1,method='bilinear')

#add to main stack allimages <- addLayer(climate,fc1km)

#export to save probelms later saveRDS(allimages, file = "allimages.rds")

##################################################

###Analysis#######################################

################################################## allimages <- readRDS(file = "C:/Users/Charlie/Documents/data/allimages.rds")

#note forestcover called layer???

#csv

spcsv <- read.csv('C:/Users/Charlie/Documents/data/species.csv') sp <- SpatialPoints(data.frame(x= spcsv$longitude,y=spcsv$latitude))

#Looking at bio1 (note tempx10, so for real temp /10) plot(allimages$bio1) points(sp)

#get background histogram hist(allimages$bio1,breaks=20)

#extract our samples

#group histogram (same scale as above allowing comparison) hist(tempsp,breaks=10,xlim=c(minValue(allimages$bio1),maxValue(allimages$bio1)))

#I would suggestion you go with limits the min and max (161 - 269) may as well round to 160 - 270 (ie 16-27 oC) min(tempsp,na.rm=TRUE) max(tempsp,na.rm=TRUE)

#Looking at bio12 rainfall plot(allimages$bio2) points(sp)

#get background histogram hist(allimages$bio12,breaks=20)

#sample precsp=extract(allimages$bio12, sp)

#group histogram (same scale as above allowing comparison) hist(precsp,breaks=20,xlim=c(minValue(allimages$bio12),maxValue(allimages$bio12)))

#bio1 values from 0-160 = 0 and 160-270=1 270+=0 bio1_reclass <- c(0,181.5,0,181.5,268,1,268,300,0) bio1_matrix <- matrix(bio1_reclass,ncol=3,byrow=TRUE) bio1_bin <- reclassify(allimages$bio1,bio1_matrix) plot(bio1_bin)

#Manual thresholding bio1_reclass <- c(0,160,0,160,270,1,270,300,0) bio1_matrix <- matrix(bio1_reclass,ncol=3,byrow=TRUE) bio1_bin <- reclassify(allimages$bio1,bio1_matrix) plot(bio1_bin)

#Charles attempt at reclass

################################################

#bio17

bio17_reclass <- c(0,95.2,0,95.2,604.3,1,604.3,750,0)

bio17_matrix <- matrix(bio17_reclass,ncol=3,byrow=TRUE)

bio17_bin <- reclassify(allimages$bio17,bio17_matrix) plot(bio17_bin)

#Manual treshold bio17_reclass <- c(0,200,0,200,600,1,600,750,0) bio17_matrix <- matrix(bio17_reclass,ncol=3,byrow=TRUE) bio17_bin <- reclassify(allimages$bio17,bio17_matrix) plot(bio17_bin)

#bio4 bio4_reclass <- c(0,353,0,300,1580.9,1,1580.9,1973,0) bio4_matrix <- matrix(bio4_reclass,ncol=3,byrow=TRUE) bio4_bin <- reclassify(allimages$bio4,bio4_matrix) plot(bio4_bin)

#Maunual threshold bio4_reclass <- c(0,300,0,300,1400,1,1400,1973,0) bio4_matrix <- matrix(bio4_reclass,ncol=3,byrow=TRUE) bio4_bin <- reclassify(allimages$bio4,bio4_matrix) plot(bio4_bin)

#bio12

allimages$bio12

© The Author. All rights reserved. 42 bio12_reclass <- c(0,1973,0,1973,3656,1,3656,6000,0) bio12_matrix <- matrix(bio12_reclass,ncol=3,byrow=TRUE) bio12_bin <- reclassify(allimages$bio12,bio12_matrix) plot(bio12_bin) hist(bio12_bin)

#Manual threshold bio12_reclass <- c(0,1800,0,1800,3800,1,3800,6000,0) bio12_matrix <- matrix(bio12_reclass,ncol=3,byrow=TRUE) bio12_bin <- reclassify(allimages$bio12,bio12_matrix)

#merging rasters

#forestcover as we have only one limit it's easy maths

plot(allimages$layer)

fc_bin <- allimages$layer > 60 plot(fc_bin)

#get out the simple map resultMap <- (bio1_bin + bio4_bin + bio12_bin + bio17_bin) * fc_bin #not forest cover is behaving as simple mask plot(resultMap) points(sp)

biostack<- stack(bio1_bin + bio4_bin + bio12_bin + bio17_bin) * fc_bin biostackd <- as.data.frame(biostack,xy=TRUE)%>%drop_na()

plot(biostack)

trim_t <- function(x){

x[(x > quantile(x, 0.25)-1.5*IQR(x)) & (x < quantile(x, 0.75)+1.5*IQR(x))]

}

hist(trim_t(allimages))

hist(allimages$bio17)

precsp <- extract(allimages$bio4, sp) quantile(precsp,c(.05,.95),na.rm=TRUE)

bt <- quantile(precsp,c(.05,.95),na.rm=TRUE) bio12bin<- (allimages$bio1 > bt[[1]]) + (allimages$bio1 < bt[[2]]) hist(allimages$bio12,breaks=10) hist(allimages$bio12-bio12bin,breaks=10)

#New map with label x=c(121.8,121.8,121.2,121.2,121.8,121.8,122.5,121.8)

y=c(18.3, 16.5, 16.3, 14, 14, 16, 17, 18.3)

East_luzon=tibble(x,y)

x=c(124.2,125,125.9,124.2)

y=c(12.5, 10, 12.5, 12.5)

Samar_Leyte=tibble(x,y)

x=c(125.3,124.8,126.2,127,125.3)

North_Eastern_Mindanao=tibble(x,y)

ggplot()+

geom_raster(data=biostackd,aes(x=x,y=y,fill=layer))+

geom_point(data=species,aes(x=longitude,y=latitude), size = 2,colour="blue") +

annotate('text',x=123,y=16,label='East\nLuzon') +

annotate('text',x=125.8,y=13,label='Samar & Leyte') +

annotate('text',x=126.8,y=9.5,label='North Eastern\nMindanao') +

scale_fill_viridis_c(option = "magma") +

coord_sf(expand=c(0,0))+

labs(x='Longitude',y='Latitude',

title="",

subtitle='',

caption='')+

cowplot::theme_cowplot()+

theme(panel.grid.major = element_line(color = gray(.5),

linetype = 'dashed',

size = 0.5),

panel.grid.minor = element_blank(),

panel.ontop = TRUE) +

xlim(117,128)