Variation in the Manna of Eucalyptus Viminalis

Home , Eriococcus coriaceus, Eucalyptus viminalis, Honeydew (secretion)

Variation in the manna of Eucalyptus viminalis

Department and University: School of Natural Sciences, University of Tasmania Supervisory team: Geoff While, Julianne O’Reilly-Wapstra and Peter Harrison

Amy Wing May 2020

A thesis submitted in partial fulfillment of the requirements of the degree Bachelor of Science with Honours

Declaration I hereby declare that this thesis contains no material which has previously been accepted for the award of any other degree or diploma and contains no copy or paraphrase of material previously published or written by any other person, except where due reference is made in the text of this thesis.

Statement on authority to access

This thesis may be made available for loan and limited copying in accordance with the Copyright Act 1968.

Signed: Amy Wing

Date: 20/05/2020

Acknowledgements I would love to thank my three amazing supervisors, Geoff, Julianne and Peter. This project was a blast, I loved all the knowledge and enthusiasm you all had towards me and my project. Thank you Geoff for getting me into such an amazing project and team, despite the fact I am not being a Lizard person. The immense support and advice on my work, no matter the hour, was such a huge help. Thanks to Peter for all the expertise in all things statistics. The hours upon hours you dedicated to helping me wrap my head around the data was absolutely essential for what I achieved. And thank you to Julianne for being such a great support and encouragement in my writing. Towards the end of this project, when we had to all work at a distance, all those small words of encouragement helped me keep going. I was honestly blessed to have all three of you as apart of my honours experience and I hope you all know how much you are appreciated. A huge thank you to Sally Bryant and the Tasmanian Land Conservancy for awarding me the Bird Conservation Fund Scholarship. Thank you for taking on this bright and bubbly character. Sally, your commitment to the protecting the forty-spotted pardalote and so many other creatures is inspiring and it was lovely to spend time on Bruny Island with you as I tried my best to absorb your extensive knowledge. Thank you to Greening Australia and all the property owners that allowed me to come and sample on their properties. There was so much enthusiasm and encouragement for my research, and always a chatty face to almost all properties I rocked up too. Thank you to Jimmy Collinson from Greening Australia for helping me find my common garden field trials. As well as a big thank you to Roderic O’Connor and Adrian for allowing me onto their beautiful property in Connorville, and to Geoff and Anthea Hendley for access to their farm in Relbia. I would also love to thank all the property owners on Bruny Island and the overall enthusiasm to my research. Thanks Tonia Cochran of Inala Nature Tours, Lauchlan Story, Daniel Sprod, Ben Kienhuis and the managers at the Murryfield Station. Thank you, David Nichols, and your team in the Central Science Laboratory for helping me with the chemical analysis. It was fun being a lab coat wearing scientist, even if just for a few days. Thank you to the School of Natural Sciences for all the cake and tea sessions early on. There is a great family feel within this group and you all helped me and my group feel so welcome and comfortable here! Also thanks to David Green, Glen Bain, Paul Tilyard and Thomas Baker for all the help in my field work and the many silly questions around paper work. Also, a big shout out to Hugh Fitzgerald, for your extra help in my chemical analysis and organising of field work, you went above and beyond, so thank you! Thank you, Tanya Bailey, for your help with my sampling in Midlands, and for giving me a place to stay. Our therapy discussions amongst the trees and all your time in the field will always be remembered.

Thank you to all my friends and volunteers who kept me company in the field, some days with only a few hours’ notice. Sophia, Maryanne, Jake, Richard, Tom, Zoe and Melanie, you were all such a huge help and made the hours in the sun even better! A big thanks to my mid-year cohort, Zoë, Zach, Elizabeth, Laura and Fiona. Thanks for coming along for this crazy adventure. None of us expected the turn of events that happened due to COVID-19, but we managed to remain as a great group and support each other throughout, and I hope we remain friends for a long time. I’d like to thank my friends and family for the extra support throughout this project. Thanks to my parents for all the encouragement throughout my whole science degree and the huge amount of support from my partner, Tom, especially in those final few weeks. Finishing my thesis in isolation was not easy, but I can thank every one of you for picking up the phone or logging onto Zoom when I just needed a chat or study buddy. And finally, thanks to the University of Tasmania and particularly the Natural Sciences for accepting me as a honours student. I know I cannot have possibly remembered all the people that helped me out this year, as there were just so many people who helped me and made this experience great. I thank every single one of you as this will be an experience I remember forever.

Contents ABSTRACT ...... 5 Introduction ...... 6 Eucalyptus woodlands as foundation species ...... 6 The importance of sugary resources ...... 7 The important relationship between the Forty-Spotted Pardalote and Eucalyptus viminalis ...... 8 What do we know about manna? ...... 9 Aims ...... 10 Methods ...... 12 Part one: Common-garden field trials ...... 12 Part two: Bruny Island ...... 16 Manna Analysis by Liquid-Chromatography Mass-Spectrometry ...... 18 Tree traits and environmental variables ...... 19 Data Analysis ...... 19 Common garden field trials ...... 20 Natural populations of E. viminalis at Bruny Island ...... 21 Results ...... 23 Discussion ...... 34 Variation in manna across the common garden field trials ...... 35 Variation in manna across the native distribution of E. viminalis ...... 36 Manna quality and the forty-spotted pardalote distribution ...... 38 Future directions and Implications of research ...... 39 REFERENCES ...... 41 SUPPORTING MATERIAL ...... 46 LITERATURE REVIEW ...... 47 ...... 55

ABSTRACT Plant resources are essential for animal distribution and survival in all communities. Therefore, it is of uppermost importance that ecologists understand how resources respond in changing conditions to aid in future conservation and management of natural spaces. This is particularly so regarding foundation species, where subtle intraspecific changes in their plant resources can have broad scale effects throughout the community.

The manna of Eucalyptus viminalis is an essential food source of the endangered forty-spotted pardalote (Pardalotus quadragintus). Manna is not well understood or studied and the potential response under climate change is unknown. This study aimed to explore the mechanisms behind manna trait variation in a two separate approaches. 1) to explore if manna exhibits variation influenced by genetic and environmental factors, by utilising two common-garden field trials in the Tasmanian Midlands. 2) to explore manna variation across native distributions of E. viminalis on Bruny Island and investigate relationships between climatic and soil variables.

This research further reinforces the importance of sucrose and raffinose in manna, as well as the composition of manna. Manna quality as well as sucrose and raffinose sugars varies spatially across Bruny Island. Manna quality is influenced by the environmental variable annual heat moisture index (AHM), across Bruny Island. This research did not find any indication of genetic control over manna quality. These results indicate that manna may be vulnerable to climate change, which could have devastating effects on the endangered pardalote species. The results can be used to aid future management and conservation efforts, as well as supplementary feeding trials.

Introduction The survival and success of all animal species relies on the resources they can acquire. This includes resources for food, habitat and protection. These resources can vary either spatially or temporally, in quality or in quantity, and variation resources has the potential to impact animal use around the resource. Trait variation within plant resources, such as higher or lower quality resources, may influence the behaviour of the herbivore using the plant resource (Price 1987; Wallis et al. 2002). Intraspecific variation within resources has been proven to influence herbivore behaviours and choice around resource use, which can span many ecosystem levels within the community (Bailey et al. 2006; Barbour et al. 2009; Frye et al. 2013). This behaviour and choice around resource uses by a single species can cause wider spread influences on the community. Two examples of this is it can lead to plants resources investing in defences against browsing (a variation in trait) (Gowda 1996) and it can lead to increased competition for resources (a variation in abundance) (Arsenault and Owen‐Smith 2002). The ways in which species respond to variation in resources can have transformative effects on the community and ecosystem around them, by influencing other species within the community, and cascade through animal plant interactions within the community.

Trait variation in plant resources is controlled primarily by one of three factors. Traits can vary based on genotype (the underlying genetic composition of an organism), due to environmental variation (such as climatic factors of precipitation and aspect) and an interaction between genotype and environment (for example, the response to the environment depends on different genetic backgrounds)(Nicotra et al. 2010). Some resources of the same species have high importance in the community structuring and composition then others. These are known as foundation resources, and they play important roles in structuring ecosystems, biodiversity and community productivity. Variations in resources, particularly those that are genetically underpinned, can influence broad aspects of the community (Bailey et al. 2009).

Eucalyptus woodlands as foundation species Eucalyptus woodlands are a critical habitat in Australia, that are habitats for a large diversity of species. They can form the basis of very productive communities and are considered a foundation species in many of these systems (O'Reilly-Wapstra et al. 2004). Eucalypts can also

6 play important roles in regulating forest structure function and stability by stabilizing ecosystem functions (Barbour et al. 2009; Whitham et al. 2006) and intraspecific variation in their trees resources can influence the wider community. Variation in eucalypt defensive chemistry can be influenced by both genetic and genetic by environmental factors (O'Reilly- Wapstra et al. 2005), making it a topic of interest in relation to community genetics.

Many studies in the Eucalyptus system have examined intraspecific variation in plant defensive traits and shown that these traits are under genetic control (Lawler et al. 2000; O'Reilly-Wapstra et al. 2002; O'Reilly-Wapstra et al. 2004) and genetic by environmental interaction (Herrera and Bazaga 2011). Although the majority of intraspecific trait variation in defensive compounds appears to have some element of genetic control (Herrera and Bazaga 2011; Lawler et al. 2000; O'Reilly-Wapstra et al. 2002; O'Reilly-Wapstra et al. 2004), some eucalypt traits are also influenced by environmental effects (Macfarlane et al. 2004; McLean et al. 2014; Nicotra et al. 2010).

While there has been a large amount of work in the past two decades on understanding how eucalypt defensive traits vary within a species and the implications on feeding herbivores, little work has examined intraspecific variation in positive tree traits that feed animals but do not have a defensive role for the tree, such as sugars resources such as nectar, honeydew, sap and manna. Does intraspecific trait variation in positive resources also major community wide influences on the broader community?

The importance of sugary resources Plant resources high in sugar content provide high levels of energy and nutrients to many animals and bird species (Bolnick et al. 2011). Research into the important plant resources of nectar show that trait variation can be influenced by genetic, environmental or genetic by environmental interaction, but environmental factors mediate the majority of its traits variation (Mitchell 2004; Parachnowitsch et al. 2019). Manna is a liquid sugary secretion that oozes from wounds mostly caused by insect bites on the tree stems and hardens to an almost solid state. It has only been observed within the Myrtaceae (Myrtle) family, on several species of Eucalyptus (including, but not limited to, E. viminalis, E. Paucifora, E. panctata, E. maculate,

E. manifera) and two Angophora species (A. floribunda and A. costata) (Basden 1965; Paton 1980; Steinbauer 1996). Manna is an important resource to woodland birds, particularly species that rely on sugar rich compounds all year around. However, research is limited on manna, and as a result even basic understanding such as how it arises or what it is influenced by, is not yet understood. It was previously hypothesized that manna secretion was a response to insect saliva on tree stems (Basden 1965), but recent research has shown that manna secretion can be promoted by scalpel wounds as well (Case and Edworthy 2016). Therefore, manna secretion is likely to be linked to sap, like other Eucalypt resources such as extrafloral nectaries (nectar-secreting glands that develop outside of flowers, Bentley 1977) or kino (a reddish-brown exudate occurs in response to wounds on tree; Tippett 1986), but this is still unknown. Only a couple of studies have investigated manna’s chemical composition, and as a result we know manna does vary interspecifically between trees and intra-specifically between individuals within species (Basden 1965; Swart 2017). There is intraspecific variation within one Eucalypt species in traits of sugar concentration and manna quality (Swart 2017), however there is no research into what mechanisms relate to manna variation within species.

Manna is an important resource of eucalypt systems, being consumed by a vast range of animals including insects, marsupials and birds (Paton 1980; Paton 1982; Steinbauer 1996). It also feeds marsupial species. Sugar gliders (Wallis and Goldingay 2014) have been seen opportunistically eating this carbohydrate rich resource. Many insects from the Hymenoptera family feed on manna when available, particularly ants, which feed on manna directly from tree, or when fallen on ground (Paton 1980; Steinbauer 1996). Up until recently, the research surrounding how manna was produced was uncertain. A recent study uncovered that manna could be created by artificial wounds to tree stems (Case and Edworthy 2016). This rejected previous hypothesis that manna was an enzymic reaction of the insect saliva on the exposed wound of the stem (Basden 1965).

The important relationship between the Forty-Spotted Pardalote and Eucalyptus viminalis The forty-spotted pardalote (Pardalotus quadragintus) has a close life history with Eucalyptus viminalis trees. It lives and feeds from them and can also perform a rare action of inducing manna secretion. The small habitat specialist has a hooked beak and scrapes at tree stems to

8 produce manna, an action only first recorded in 2016. The small canopy gleaning bird returns to the wound sites and collects manna secretions once hardened (Case and Edworthy 2016). This is evidence of ‘farming’ and may have implications for the broader community, as manna is consumed by a wide array of animals who opportunistically consume it. Therefore, the act of producing manna, and therefore the presence of the pardalote in any given community, may change the shape of the community, and influence the presence and abundance of other species within.

Manna is an important resource to the pardalote and makes up a large proportion of its diet (Bryant 2010; Case and Edworthy 2016). Manna makes up 80% of the diet of young fledglings that are growing and still reliant on their parents. This makes manna as a resource particularly important as the forty-spotted pardalote is an endangered species that has experience major range restrictions and loss of population. As an endangered species, the health and survival of offspring is absolutely crucial for their future.

What do we know about manna? Manna is a sugary carbohydrate of great complexity. Around 60% of the compound is pure carbohydrates, the remaining proportion is mostly water. Research on manna chemical sugar compositions shows us that it does vary between tree species (Basden 1965) and within trees of the same species (Swart 2017). Specifically, the manna of E. viminalis contains more than 20 different types of sugars and just 5 sugars make up 99% of the total sugar composition. They are raffinose, sucrose, stachyose, fructose and glucose, and 5 sugars vary in proportion between different trees. This is most evident when looking at raffinose and sucrose contents. Previous research by Swart (2017) looked at the behaviours of pardalotes in association to their tree use. She uncovered that the pardalotes have primary and secondary trees in which they farm manna from, and that these trees have different sugar proportions. Primary trees, in which pardalotes spent upwards of 80% of their time foraging in, had significantly higher proportions of sucrose relative to raffinose, when compared to secondary trees (which pardalotes spent less than 15% of time in) (Figure 1). Therefore, Swarts results shows a preference of high sucrose and low raffinose in the manna. This is referred to as the sucrose to raffinose ratio (SUC:RAF), and is an indicator of manna quality.

Manna as a resource, and the variation within its traits, is still very misunderstood. Understanding how manna varies and the mechanism that underpin it is crucial, as the forty- spotted pardalote is a species a vulnerable species to the impacts of climate change (Foden et al. 2013). It relies heavily on E. viminalis, and has already experienced significant population declines (Bryant 2010). In order to make proactive decisions in conservation and management around this vulnerable species, knowledge in an essential food source is critical moving forward trying to protect this species.

Aims The main aim of this research was to explore the intraspecific variation in manna quantity of manna in E. viminalis and understand the genetic and environmental effects mediating this variation. To do this, I separated the study into two approaches.

In part one I investigated variation in manna in two common garden field trials to determine the degree to which variation in manna is genetically and environmentally influenced.

In part two I investigated how manna quality varies across a native distribution of E. viminalis and assessed if variation in manna related to climatic and soil variables.

Figure 1: Variation in manna quality between primary and secondary used trees of the forty- spotted pardalote (Pardalotus quadragintus) on Bruny Island. Figure of (Swart) 2017.

Methods Part one: Common-garden field trials To determine whether the quality and quantity of manna varied due to provenance (genetic), site (environment), or their interaction (genetic by environment), I accessed two pre- established common-garden field trials in the northern midlands of Tasmania (Connorville: - 41.828636°, 147.138727°; Relbia: -41.520906°, 147.240920°). The two trial sites were 35km apart and occupied similar climates (mean annual temperature: Connorville 11.5°C, Relbia 12.2°C; mean annual precipitation: Connorville 626mm; Relbia 654mm) at different elevations (Connorville: 179m; Relbia: 58m). The trials were established in September 2016 (Connorville) and June 2017 (Relbia) using open-pollinated seed (hereafter ‘families’) collected from 8-10 trees from each of the seven provenances (i.e. the geographical location of the germplasm) (Table 1). These provenances were sampled along a temperature and precipitation gradient that transgressed the midlands of Tasmania (Whitmore 2015). One- year old seedlings were planted into rip mounds (Connorville) and Wilko mounds (Relbia) as single-tree plots in a randomised complete block design with 2 (Connorville) and 6 (Relbia) replicates. As early establishment at Relbia was low and to keep sampling for manna balanced across the two sites, I collected manna from plants from the two replicates in Relbia with the highest survival (Replicates 2 and 5). This sampling strategy resulted in each family being represented at least twice at Relbia.

Table 1: Description of the seven populations planted in the common garden field trials. Shown are the seven provenances and their maximum temperature of the warmest period (TMXWP), mean annual precipitation (MAP) and annual heat moisture index (AHM) calculated as (MAT + 10) / (MAP/1000) where MAT is the mean annual temperature of the provenance. The grey shaded cells correspond to the northern and southern provenances sampled outside the midlands of Tasmania.

Region Provenance Code TMXWP (°C) MAP (mm) AHM

Non-arid Brushy Lagoon BL 22.5 998.1 21.0

(North) Nunamara N 22.1 1064.2 19.3

Symmons Plains SP 24.2 616.7 35.0 Arid Connorville C 24.4 611.5 34.8 (Central) Ross R 23.8 509.1 41.5

Non-arid Tooms Lake TL 20.5 557.9 35.5

(South) Stonor S 21.4 604.8 33.1

Manna extraction: Manna was extracted for each individual in the above-mentioned replicates at Connorville and Relbia by making small lateral incisions in the branch with a scalpel (Figure 2A) following the protocol outlined by Case and Edworthy (2016). Incisions were only made on healthy stems between 1mm and 8mm in diameter, where up to four stems per individual that were at least 1.5m above ground (where possible) were sampled for manna. On each stem, two separate cuts more than 2 cm apart were made usually above a stem axis or below a leaf petiole, giving a total of up to eight incisions per tree (i.e. four stems per individual and two incisions per stem). A plastic zip-lock bag was then used to cover the stem and incision sites (Figure 2B,C), with holes cut in the bottom of the bag to allow for ventilation and the release of built up condensation. The bags were used to protect manna secretions from foragers, such as ants and birds. However, in some instances, the bags did not protect the manna from ants and other invertebrates and unavoidable foraging of manna was observed at Connorville. The

13 stems were left for 24 hours before manna was collected into 50mm vials. Total secretions from each tree were pooled into one vial and it was noted how many secretions were produced by each tree and the stem diameters of where manna secretions formed. Vials of manna were labelled with the individual plants unique identifier and placed in a -20°C freezer at the end of the day to limit any deterioration of the manna before chemical processing could occur.

To determine whether sugar composition of manna varied through time, manna samples were collected across at least two sampling periods. In all cases, manna was extracted from the same individuals across the sampling periods, with the exception at Relbia, where an additional third replicate was sampled during the second sampling period. This was due to low success rates extracting manna during the first sampling period, as to obtain a suitable representative sample size across provenances. A third sampling period was undertaken at only Connorville following a heavy two-day rain event (total precipitation 28ml), providing an opportunity to test whether success in extracting manna may be coincidental with increased soil moisture. For Connorville specifically, 7 trees produced manna in all three time points, 8 trees produced manna across two time points and 19 trees produced manna once. For Relbia, a total of 5 trees produced manna over the two sampling periods and a further 24 trees produce manna a single time. Sixty-three trees produce manna over all sampling periods from both field trials. The total number of samples collected from each population within each common garden site is detailed in Table 2. Of the 63 trees that produced manna, thirteen of the samples were not collected due to the manna sample being either too small to collect (n = 10), contaminated by dirt and animal faeces (n = 1), or dropped in the wind (n = 2).

A B C

Figure 2: Example of the manna extraction procedure. Using a scalpel blade, an incision was made to the stem of Eucalyptus viminalis (A). The branch, including the incision site, was bagged to protect the manna from degradation and harvesting by insects (B). The incision site was left for 24 hours, where the crystalised manna was collected (C).

Table 2: The manna sample success from each provenance across the Connorville and Relbia common garden field trials in the Tasmanian Midlands. Including the number of Eucalyptus viminalis trees sampled and success rates of manna extraction for each provenance. The grey shaded cells correspond to the northern and southern provenances sampled outside the midlands of Tasmania.

Connorville Relbia

Provenance Trees sampled Manna success Sample size Manna success

Bushy Lagoon (BL) 9 5 14 4

Nunamara (N) 12 5 15 1

Stonor (S) 6 2 15 3

Connorville (C) 15 7 21 1

Ross (R) 10 4 20 9

Symmons Plains (SP) 10 7 18 6

Tooms Lake (TL) 12 4 16 5

Total 74 34 119 29

Success rate 45.9% 24.4%

Part two: Bruny Island To determine whether manna quality varies across the native stands of E. viminalis, I assessed manna quality across ten sites on Bruny Island in February 2020. Bruny Island was selected because it represents a stronghold for the forty-spotted pardalote, supporting around 450 individuals, which constitutes a third of all known occurrences of this species. Sites were distributed evenly across North and South Bruny (five sites within each region, Figure 3) and selected to reflect varying soil types and aridity gradients. At least five of the selected sites had (or had in the past few years) forty-spotted pardalotes nesting within trees at that site (Bryant 2010). Trees within each of the ten sites were sampled at two canopy heights or 100m apart. Unlike the common garden field trials where manna was sampled directly from tree branches on site, two small branches were collected from within individual tree canopies on Bruny island using an extendable pruning pole, up to a maximum height of 10 metres. The branch was labelled with a unique identifier and placed in a bucket of water as soon as possible before being transported back to the University of Tasmania. Branches were then stored in buckets of water and incisions made with a scalpel on multiple stems from each branch (see ‘Manna extraction’ in Part 1 above). Incisions were checked every 24 hours for visible manna secretions. Manna from separate stems of each tree were placed in separate vials to see if manna varied between branches on the same tree. If no manna had been produced, repeat cuts were made and checked every 24 hours. This continued until manna was extracted from at least one branch per tree, or the branch wilted. Dates were noted from which manna secretions were collected. A total of 66 trees were sampled within an eight-day period, 48 trees successfully produced manna.

Figure 3: Distribution of sampled populations of Eucalyptus viminalis on Bruny Island. Black points indicate the sampled trees within locations which were grouped by region, where the red circles correspond to the populations sampled on North Bruny (N1-N5) and blue circles correspond to the sampled South Bruny (S1-S5) populations.

Manna Analysis by Liquid-Chromatography Mass-Spectrometry All manna samples produced in part one and part two were processed at the University of Tasmania’s Central Science Laboratory. Sugars were extracted and analysed by Liquid- Chromatography-Mass Spectrometry (LC-MS). The manna sample in each vial was weighed and total weight recorded in milligrams (mg) to five decimal places and recorded for overall weight. A subsample of between 1.00 and 2.00mg (maximum of 2.00mg and minimum of 0.01mg) from each vial was taken and placed into a 2mL LC-MS vial. Manna was dissolved with equal proportions of distilled water (eg. 1mg of manna with 1mL of distilled water). The solution was vortexed for 5 seconds until manna was dissolved. Sugars were analysed by LC- MS following the below methods. I focussed on the five main sugar groups; raffinose, sucrose, stachyose, fructose and glucose, which had previously been shown to make up 99% of the total sugar composition (Swart 2017). Glucose and galactose were unable to be differentiated due to their similar molecular structure (Gaucher and Leary 1998) and were therefore grouped together.

LC-MS was conducted at the University of Tasmania’s Central Science Laboratory, using a Waters Acquity H-Class UPLC instrument coupled to a Waters Xevo TQ triple quadrupole mass spectrometer. Waters a Acquity UPLC BEH Amide column (50 x 2.1 mm, 1.7um) was used and the mobile phase used two solvents of 0.4% (v/v) Ammoniun hydroxide in water (solvent A) and Acetonitrile (solvent B). The UPLC program ran for 6 minutes at 20% A:80% B to 45% A:55% B, which was followed by immediate re-equilibration to starting conditions for three minutes. The column was held at 50°C, and sample compartment at 6°C. Flow rate was 0.35 mL min-1 with and injection volume of 1 uL.

The mass spectrometer was operated under negative ion electrospray mode with a needle voltage of 2.7kV. Analytes were detected using single ion monitoring with [M-H]- deproinated molecular species. To detect C6 monosaccharides, (m/z) 179.05 with a cone voltage of 17V was used. For disaccharides, (m/z) 341.10 at 24V, trisaccharides, (m/z) 503.15 at 30V and finally, tetrasaccharides used (m/z) 665.20 at 36V. Dwell time was 0.057 seconds and the ion source temperature 130°C. Desolvation gas was N2 at 950 L h-1 (with a temperature of 450°C ) and the cone gas flow was 100 L h-1. Data were processed using MassLynx software (Nichols, D. pers comm).

Individual sugars (raffinose, stachyose, glucose, fructose, sucrose) were identified and quantified (into parts per million, PPM) using the external calibration approach. Five individual calibration standards of each sugar were produced at known concentrations in distilled water and analysed in sequence with every 20 samples. Unidentified tetrasaccharides (in addition to stachyose) were quantified as “Stachyose equivalents” using the Stachyose external calibration curve. A small number of trees (N=8) found traces of tetrasaccharides, and therefore, due to the low sample size it was not included in statistical analysis.

Tree traits and environmental variables To determine whether sugar composition of manna traits co-varied with tree size, tree heights of individuals within the field trials were measured using height poles (in cm), and individuals across Bruny Island were measured for tree trunk size at diameter at breast height (1.3m, DBH) .To examine how variation in manna quality co-varied with home-site environment, I obtained data for each provenance (field trials) and population (Bruny Island) for key climate and soil traits thought to mediate variation in plant traits (Harrison 2017; Prober et al. 2016), downloaded from the Atlas of Living Australia (https://spatial.ala.org.au/, accessed 24 March 2020) using the ALA4R package in R (R Core Team 2013). This included three bioclimatic variables (maximum temperature at warmest period [TMXWP], mean annual precipitation [MAP], mean annual temperature [MAT]), a measure of home-site aridity 푇퐴푁푁+10 (annual heat moisture index [AHM]) estimated as (AHM= ) following Wang et al. 푅퐴푁푁/1000 (2006), and four soil attributes (total phosphorous [PTO], total nitrogen [NTO], plant extractable depth [DPE], soil organic carbon [SOC]; (Viscarra Rossel et al. 2015)) calculated as the mean value within a 2 m soil profile.

Data Analysis Manna sugar composition obtained from the common garden field trials and natural field population were analysed separately using R version 3.6.2. Raw sugar outputs were in in present amounts (PPM), and the total of all sugar outputs (total PPM) gives total sugar concentration within the manna sample. For data analysis, sugars were standardised across samples as the proportion of each sugar relative to the total sugars within given sample. Swart (2017) identified the ratio of sucrose to raffinose (hereafter SUC:RAF) as an important

19 predictor variable associated with tree preference by the forty-spotted pardalotes. Thus I also calculate SUC:RAF as the proportion of SUC divided by the summed proportion of SUC and RAF. Prior to analyses, data was explored Zuur et al. (2010). Univariate and bivariate outliers were detected using a modified z-score test and Mahalanobis distance, respectively. Outliers were checked for data entry errors and removed where necessary to limit adverse effects of leverage and influence. This procedure removed 4 and 9 points from the common garden field trial and Bruny Island data sets, respectively.

Common garden field trials To explore associations between mannaa produced by the sampled E. viminalis individuals over time points (t), I used a Spearman’s Rank correlation test. This was pursued by comparing the sugars within a site for the following contrasts: t0 vs. t1 (both field trials), t0 vs. t2 (Connorville only) and t1 vs. t2 (Connorville only). As sugar quality was significantly positively associated among the time periods (see ‘Results’), duplicated samples across the time periods were removed.

To test whether the proportion of each sugar and the SUC:RAF ratio varied among sampled provenances, site and the interaction of provenance by site, I fitted the following model: Y = µ + provenance + site + prov*site + height + ε (Model 1) where µ is the mean response, provenance, site and their interaction are fixed effects, and height is the normalised height of the tree fitted as a covariate. This presence/absence of manna was fitted using a generalized linear model (GLM) assuming a Bernoulli error with a logit link function. Model assumptions of normality and homoscedasticity were visually assessed following Zuur et al. (2016) using diagnostic plots produced from simulated model residuals using the DHARMa package (Hartig 2019). Statistical significance of fixed effects were assessed using a likelihood ratio test implemented using the ‘drop1’ function.

As the proportional data followed a beta distribution (i.e. bounded between the open interval (0,1); (Douma and Weedon 2019; Ferrari and Cribari-Neto 2004), models were fitted using generalized additive models (GAM) with a logit link function using the ‘gam’ function the mgcv package (Wood 2017). As no smoother terms were incorporated into the model, the GAM

20 was effectively a penalised GLM (Wood 2017). In cases where the proportion of a sugar was at the boundary of the beta distribution, the sugar values were re-scaled following Smithson and Verkuilen (2006). Model assumptions of normality and homoscedasticity were visually assessed using diagnostic plots produced by the ‘gam.check’ function following Zuur et al. (2016). Significance of fixed effects were assessed using a likelihood ratio test implemented using the ‘anova.gam’ function As I a priori hypothesised that manna would vary by home- site environment (where the seed was originally sourced from), I tested associations between provenance mean sugar and home-site climate and soil using a Spearman’s Rank correlation test.

Natural populations of E. viminalis at Bruny Island To test whether the proportion of a sugar varied within a tree, I assessed the association between manna samples collected across different branches (9 pairs) using a Spearman’s Rank correlation test. Following this, duplicated manna samples within a tree were removed.

To test whether the probability of obtaining manna among sampled populations and their test environment, I fitted the following model:

Y = µ + population + DBH + ε (Model 2) where µ is the mean response, population is a fixed effect, and DBH is the normalised diameter at breast height (1.3m) fitted as a covariate. As above-mentioned, success was fitted using a generalized linear model (GLM) assuming a Bernoulli error with a logit link function. Model assumptions and statistical significance of the fixed effect of population and covariate of DBH were assessed following the same procedure as above. To test whether the proportion of each sugar and the SUC:RAF ratio varied among populations and covaried with tree growth (DBH), Model 2 was fitted using a generalized additive models (GAM) with a logit link function. Model assumptions and significance of the population and DBH terms were assessed as above. To test whether the proportion of each sugar covaried with population home-site climate and soil, I extracted the marginal least-square means from Model 2 using the emmeans package (Lenth 2019) and fitted separate beta regression GAMs, fitting the

21 marginal least-square mean as the response and each climate and soil variable as the predictor.

Results Common garden field trials One-hundred and ninety-two Eucalyptus viminalis trees were sampled across the Connorville and Relbia common garden field trials planted in the Tasmanian Midlands. There was low success in extracting manna across the 2-3 attempts, with 45.9% and 24.4% success at Connorville and Relbia respectively (Table 2). The manna from Connorville on average consisted of 64.62% raffinose, 33.23% sucrose, 1.26% stachyose and less than 1% glucose and fructose combined (Table 3). This was similar to Relbia, with raffinose making up 60.05%, followed by sucrose at 37.31% and stachyose at 1.79%. Glucose and fructose again made up less than 1% of the manna. The results of exploration into any variation over the different time periods showed significant correlation for the main sugar types (FRU, SUC, RAF and SUC:RAF; Table 4). This positive correlation between attempts suggests that the variation in sugar does not change the provenance mean rank across time periods.

There was variation between sugar content across the field trails, and between provenances at each site. Relbia has a higher overall means of SUC:RAF than Connorville, of 0.3828 and 0.3394 respectively (Table 3). Provenance variation across field trials was also observed, most notably provenance C, which has a SUC:RAF difference of 0.2%, attributed to around 20% variations in SUC and RAF at Connorville compared to Relbia. Despite these observed 2 patterns, there were no differences in the concentrations between site (푋1 = 1.11, 푃 = 2 0.292), provenance (푋5 = 7.69, 푃 = 0.174), or a significant provenance by site interaction 2 (푋6 = 4.63, 푃 = 0.592). Exploration of the environmental data of maximum temperature at warmest period (TMXWP) and annual heat moisture index (AHM) indicated that there was no significant association between climate and SUC:RAF across provenances for either common garden trial (Table 5).

Table 3: Summary of sugar proportions within the manna of Eucalyptus viminalis in the common-garden field trials of Connorville and Relbia of the Tasmanian Midlands. Includes breakdown of samples within each provenance established in the common gardens.

Total Individuals Prov manna GLU (%) FRU (%) STA (%) SUC (%) RAF (%) SUC:RAF sampled success Connorville BL 4 9 0.24 0.56 1.31 41.23 56.66 0.4210 N 4 12 0.57 0.69 1.02 34.16 63.57 0.3493 S 1 6 0.25 0.47 1.03 35.91 62.34 0.3655 C 5 15 0.27 0.34 1.85 21.47 76.08 0.2209 R 4 10 0.75 0.79 1.08 30.44 66.94 0.3121 SP 5 10 0.51 0.62 1.23 30.17 67.47 0.3086 TL 3 12 0.00 0.18 1.27 39.25 59.29 0.3982 Overall Means: 0.37 0.52 1.26 33.23 64.62 0.3394

Relbia BL 3 15 0.41 0.52 1.92 33.99 63.16 0.3497 N 0 15 S 3 12 0.55 0.74 1.74 39.44 57.54 0.4060 C 2 20 0.00 0.36 1.31 42.15 56.19 0.4286 R 6 13 0.33 0.52 1.94 34.02 63.19 0.3499 SP 4 14 0.65 0.74 1.69 28.93 67.99 0.2979 TL 4 12 0.03 0.33 2.12 45.32 52.20 0.4648 Overall Means 0.33 0.56 1.79 37.31 60.05 0.3828

Table 4: Spearman’s rank correlation test across attempts at the Connorville common garden field trials: Sugars included are fructose (FRU), glucose (GLU), stachyose (STA), sucrose (SUC) and raffinose (RAF).

Sugar type RHO P-value FRU 0.767 0.016 GLU 0.000 1.000 STA 0.450 0.224 SUC 0.667 0.050 RAF 0.700 0.036 SUC:RAF 0.667 0.050

Table 5: Associations between climate variables and the sucrose to raffinose ratio at the common-garden field trial sites Connorville and Relbia of the Tasmanian midlands: Including variables of maximum temperature of the warmest period (TMXWP) and annual heat moisture index (AHM).

Connorville Relbia

RHO P-value RHO P-value

TMXWP -0.643 0.1194 -0.2 0.704

AHM -0.5 0.253 0.257 0.623

B A

C D

Figure 4: Proportion of sugar contents within total manna sugar composition of Eucalyptus viminalis provenances grown at the Connorville and Relbia common-garden field trials in the Tasmanian Midlands: A and B provides visual representation of the sugar proportions in manna at Connorville and Relbia. Sugars included are fructose (FRU), glucose (GLU), stachyose (STA), sucrose (SUC) and raffinose (RAF). C and D are boxplots of the sucrose to raffinose ratio variation between provenances across the two field trials.

Natural populations of E. viminalis at Bruny Island

Seventy-four Eucalyptus viminalis trees were sampled across Bruny island, of which 78.7% successfully produced manna to cuts on tree stems. Thirteen trees did not produce any manna. Success in extracting manna did not vary between populations (see Table 6, Figure 7) 2 (푋7 = 5.758, 푃 = 0.568 . The size of tree, DBH, was not associated with the probability of 2 success of manna extraction (푋1 = 1.605, 푃 = 0.205).

Of the five sugar types analysed, raffinose (RAF) makes up the highest proportion of manna analysed at Bruny Island, at 57.98%. This was followed by sucrose (SUC) at 39.15% (Table 6; Figure 8), the two sugars making up around 97% of the total sugar content in manna. The sugars stachyose (STA; 2.45%), fructose (FRU; 0.27%) and glucose (GLU; 0.15%) make up a small proportion of the overall manna composition. When exploring whether there is variation in manna within a tree, the results showed high association between branches for SUC, SUC:RAF and a close to significant correlation for RAF.(Table 7).

The sugar content in manna varied between populations on Bruny Island. Manna concentration, or total sugars present within the manna, was on average, 845 PPM. However, I found variation in manna concentration between populations, from high concentrations in S2 and low concentrations in N4 and N5 (Figure 7). In relation to specific sugar proportions, 2 2 of the five sugars, two (SUC: 푋7 = 23.29, 푃 = 0.002 and RAF: 푋7 = 26.83, 푃 < 0.001) showed significant variation between populations (Table 8). The SUC:RAF also showed 2 significant variation between populations (푋7 = 24.73, 푃 < 0.001)(Table 8; Figure 10). The SUC:RAF also exhibited significant variation at a broader spatial scale, at a regional scale between north and south (Table 8; Figure 8).

The results indicate manna quality is associated with the climate variable annual heat 2 moisture index (AHM). Figure 9shows the significant association (X1 = 5.075, P = 0.078) between SUC:RAF and AHM. As AHM increases, so does manna quality. Maximum temperature at warmest period (TMXWP) was explored for associations, but the relationship 2 was non-significant (X1 = 2.49, P = 0.114). The results in this study show no significant associations between manna quality and soil variables (Table 10). The soil variables of total 2 2 phosphorous in soil (PTO, X1 = 0.195, P = 0.659), total nitrogen in soil (NTO, X1 = 0.91,

2 P = 0.34), plant extractable depth (DPE, X1 = 0.216, P = 0.642) and soil organic carbon 2 (SOC, X1 = 0.079, P = 0.778) all showing non-significant associations.

Table 6: Mean sugar composition and tree diameter at breast height for each sampled population across Bruny Island. Shown are the number of individual trees sampled per population (sample size), number of individuals that produced manna (manna success), the mean proportion of each sugar relative to the total sugar content within a manna sample, expressed as a percentage (Fructose [FRU], Glucose [GLU], Stachyose [STA], Sucrose [SUC], Raffinose [RAF], and sucrose to raffinose ratio [SUC:RAF)). DBH refers to the diameter at breast height (1.3 m) and are the mean of the successful manna producing individuals. Populations N1 and S1 were removed from data set due to only having one manna sample produced.

Sample Manna DBH Population GLU (%) FRU (%) STA (%) SUC (%) RAF (%) SUC:RAF size success (cm) N2 10 9 0.14 0.31 2.05 36.08 61.42 0.3701 52.5 N3 8 7 0.22 0.30 2.78 48.85 47.85 0.5048 28.6 N4 4 4 0.06 0.18 2.50 44.38 52.89 0.4563 34.0 N5 4 3 0.42 0.44 2.19 37.87 59.08 0.3908 31.5 S2 9 8 0.04 0.19 2.39 34.67 62.71 0.3552 60.0 S3 4 2 0.14 0.42 2.66 42.07 54.71 0.4352 31.3 S4 4 3 0.15 0.17 3.02 31.61 65.06 0.3269 26.8 S5 5 2 0.10 0.30 2.49 31.90 65.21 0.3288 19.6 Overall Mean 0.15 0.27 2.45 39.15 57.98 0.4030 39.2

Table 7: Association between manna samples within individual trees Results from the Spearman’s rank correlation between branch 1 and branch 2 showing the correlation coefficient (RHO) and the probability of observing this coefficient given the data. Sugars are abbreviated as follows: Fructose (FRU), glucose (GLU), stachyose (STA), sucrose (SUC), raffinose (RAF) and sucrose to raffinose ratio (SUC:RAF). FRU GLU STA SUC RAF SUC:RAF RHO 0.518 0.576 0.536 0.964 0.714 0.821 P-value 0.234 0.176 0.215 >0.001 0.071 0.025

Table 8: The significance of sugar variation amongst populations on Bruny Island. Results from ANOVA showing significant values for variation in sugar quantity of sucrose (SUC), raffinose (RAF) and the sucrose to raffinose ratio (SUC:RAF), between populations.

Sugar type chi square P-value FRU 4.083 0.770 GLU 7.980 0.334 STA 1.952 0.963 SUC 23.29 0.002 RAF 26.825 <0.001 SUC: RAF 24.734 <0.001

Figure 5: Percentage of manna extraction success across sampled populations of Eucalyptus viminalis on Bruny Island.

Figure 6: Proportion of sugar contents within total manna sugar composition across populations of Eucalyptus viminalis on Bruny Island: Visual representation of how the five sugars vary among native stands of Eucalyptus viminalis on Bruny Island. Sugars represented include fructose (FRU), glucose (GLU), stachyose (STA), sucrose (SUC) and raffinose (RAF).

Figure 7: Variation in the total amount of sugar across populations of Eucalyptus viminalis on Bruny Island: Total sugar refers to the total amount (in PPM) of sugar within samples. The green triangles indicate above mean results of total sugars, and the red circles indicate below mean results. The larger the shape (triangle or circle), the further from the mean result the population is.

Figure 8: Variation in sucrose to raffinose ratio of Eucalyptus viminalis manna across populations on Bruny Island, Tasmania. Northern region in red and southern region in blue. 2 Significant associations were seen on both regional ( X1 = 5.78, P = 0.0162) and population 7 levels ( X1 = 24.734, P < 0.001).

Figure 9: Association between manna quality (SUC:RAF) and annual heat moisture index (AHM) of populations of Eucalyptus viminalis on Bruny Island: With 95% confidence intervals. AHM is Incorporates annual mean temperature with annual precipitation and gives better representation of soil moisture than precipitation alone.

Table 10: Associations between manna quality and soil variables: ANOVA of sucrose to raffinose ratio and the soil variables of total phosphorous in soil (PTO), total nitrogen in soil (NTO), plant extractable depth (DPE) and soil organic carbon (SOC)

Chi square P-value PTO 0.195 0.659 NTO 0.91 0.34 DPE 0.216 0.642 SOC 0.079 0.778

Discussion This research explored how manna quality of Eucalyptus viminalis varies over the Tasmanian landscape, and the mechanisms that may underpin this variation. Two key findings are evident. First, two key sugars (sucrose and raffinose) and their ratio (indicating manna quality) varies spatially across Bruny Island. Second, manna quality relates to the environmental variable annual heat moisture index (AHM), across Bruny Island.

This research provides insights on the composition of manna. Past research has identified five dominant sugar groups (fructose, FRU; glucose, GLU; stachyose, STA; sucrose, SUC; raffinose, RAF) comprise the majority of manna composition and the results of this study has further reinforced this. The sugar most prominent in manna across this study (as well as in previous research) is raffinose (RAF), varying from between 58-80% across three previously researched species of eucalyptus and an angophorous species. Of the previous research, E. viminalis has the highest proportion of SUC in manna compared to all other investigated species, ranging between 6% and 26.4% (Table 11). This study found SUC to vary anywhere between 21.5% (in the field trials, table 5) and 48.9% (on Bruny Island, table 7). This high sucrose proportion may be an important factor in why the pardalote species formed such an intimate relationship with E. viminalis. Other Birds and animals do feed on manna, but the forty-spotted pardalote is currently the only species known to rely on it so heavily.

Table11: Percentage manna composition comparison from previous research: Results from this study are separated from Bruny Island and averaged across the common-garden field trials. Other data is from Swart (2017) unpublished data and Basden (1965).

Source Common gardens Bruny Island, Swart Basden Basden Basden (1965) (This study) (This study) (2017) (1965) (1965) Species E. viminalis E. viminalis E. viminalis E. maculata E. punctata Angophorous costata. Stachyose 1.53 2.45 3.9 - 2 10 Raffinose 62.3 58 68.6 80 80 65 Sucrose 35.3 39 26.4 6 10 20 Glucose & <1 <1 <1 4 8 5 Fructose Melibiose - - - 10 - -

Variation in manna across the common garden field trials The aim of the common-garden trials was to understand if quality in manna varies across different provenances of E. viminalis, different environments and the interaction between provenance and environment. Common garden trails provide an excellent framework to tease apart the effects of genetics and environment on plant traits. Differences in manna quality between provenances planted in a randomised common garden trial, may indicate a genetic influence of manna variation. Furthermore, differences in manna traits across sites could indicate environmental influence of manna variation. This study found no significant differences between provenances, sites and no significant interaction between provenance and site in manna quality of E. viminalis.

Manna quality showed high consistencies across different sampling periods. This was to explore temporal variation in manna at the common garden field trials. This consisted of three separate sampling periods throughout summer at the Connorville field trial. Past research suggests seasonality in manna quantity (Spring and early summer saw greater quantities of manna than winter, coinciding with rapid tree leaf growth (Basden 1965), however, this study only looked at variation over a 6-week period, and found no results. This is potentially due to the low sample size, the difficulty extracting manna, or the time of sampling (early to mid- summer).

At this stage it is difficult to interpret these results due to low sample sizes. The low success rate of extracting manna from the common garden trials may have affected the ability to robustly test the genetic, environment and interactive effects. The method of manna extraction by using a scalpel blade was adopted from (Case and Edworthy 2016). During the study, some areas of manna extraction varied. Across the native Bruny Island sampling, manna extraction occurred off-site, where branches were brought back to the university, and many stems were cut on each branch to increase the likelihood of getting a manna secretion. On top of that, repeated cuts occurred on stems over a number of days, further increasing the chances of manna success. This was different from the method used in the common garden field trials, where up to four branches and eight stems were cut on each tree and only on one instance. This meant there were both less cuts and less attempts at gaining manna.

This may help explain the large decrease in manna success from the native stands (71%) and the field trials (45.9% and 24.4%).

There are potentially other reasons why manna success was so low at the field trials. One explanation is the age of trees, as the native tree stands were significantly older than the common garden trees. It is currently unknown whether manna is connected to any life history stages. The trees within the field trials were on the cusp of flowering and budding, and it was potentially the first reproductive season many of these trees had. If manna was linked to tree maturity, many trees within the field trials may simply have been too young to produce manna samples.

Manna quality showed high consistencies across different sampling periods. This was to explore temporal variation in manna at the common garden field trials. This consisted of three separate sampling periods throughout summer at the Connorville field trial. Past research suggests seasonality in manna quantity (Spring and early summer saw greater quantities of manna than winter, coinciding with rapid tree leaf growth (Basden 1965), however, this study only looked at variation over a 6-week period, and found high consistencies, indicating no variation. This is potentially due to the low sample size, the difficulty extracting manna, or the time of sampling (early to mid-summer).

Variation in manna across the native distribution of E. viminalis The composition of manna on Bruny Island was dominated by SUC and RAF sugars, and manna quality showed significant variation across populations on Bruny Island. Manna quality, SUC and RAF portions showed significant variation across populations and manna quality also varied across region. These results could indicate a relationship between manna traits and climatic variables. This study found a positive association between manna quality and the annual heat moisture index (AHM) of populations on Bruny Island AHM incorporates average temperature and precipitation variables instead of the two alone, to represent soil moisture content (Wang et al. 2006). AHM may also be a more sufficient variable to use in relation to effects from climate change (Castellanos-Acuña et al. 2018; Wang et al. 2006), as the predictions of changes in precipitation are small by comparison to changes in temperature (Flato et al. 2000).

Eucalyptus viminalis is a tree species which is already struggling due to climate change. The species has been experience large dieback due to the decrease in rainfall and precipitation (Jurskis 2016), and is vulnerable to drought (Li et al. 2018). This relationship E.viminalis and AHM in relation to manna indicates manna could be vulnerable to climate change. This would have negative results for the vulnerable pardalote species, as specialist species, species of low population and species of low dispersal are all characteristics of the forty spotted pardalote and are also indicators of climate change vulnerable species (Foden et al. 2013).

The environmental variables of temperature (TMXWP) was insignificant on Bruny Island indicating that this is a specific response of AHM and therefore soil moisture content. This may be due to the likely mechanism behind manna formation. Although it is not yet known exactly how manna forms, it is potentially related to sap, and sap flow is negatively affected by drought, due to its relationship to the water transport systems in the tree (Nadezhdina 1999). There were no significant associations between environmental or soil variables at for the provenances established in the common garden field trials.

Climatic and soil variables were explored in this study due to the association between other positive plant traits and environmentally influenced variation. Environmental factors lead to the largest sources of variation within traits such as nectar(Bertsch 1983; Boose 1997; Carroll et al. 2001; Jakobsen and Kritjansson 1994; Leiss and Klinkhamer 2005; Parachnowitsch et al. 2019; Petanidou and Smets 1996; Zimmerman 1983). This may be because environmental plasticity can lead to increased survival and success in a community, due to increased adaptability in climates (Chambel et al. 2005; Nicotra et al. 2010). The trait of nectar production rate is associated with water availability (Boose 1997; Carroll et al. 2001; Leiss and Klinkhamer 2005; Zimmerman 1983), humidity (Bertsch 1983) and temperature (Jakobsen and Kritjansson 1994; Petanidou and Smets 1996). For honeydew, more humid conditions may dilute honeydew droplets (Gaze and Cloud 1983) and an increase in temperature can lead to a more concentrated sugar content of the honeydew (Gaze and Cloud 1983), as well as greater quantities due to increased aphid productivity (Özder and SAGLAM 2013). Soil nutrients are known to effect both nectar and flower traits (Baude et al. 2011; Burkle and

Irwin 2009; Gardener and Gillman 2001). Increased nutrients can have both positive and negative effects on plants and vegetation (Baude et al. 2011; Bowman et al. 2004). Due to this, soil nutrient variables were explored in this study to see if they would also affect manna traits, however, no associations were seen.

Manna quality and the forty-spotted pardalote distribution A frequent question of ecologists surround why species are distributed the way they are? One potential answer surrounds the availability and quality of food resources in a habitat. The populations of Dennes hills and Inala Nature Reserve (populations N2 and S2) have the strongest colonies of the pardalotes on Bruny Island. If the pardalotes colonised sites based on manna quality, we would expect that these two sites, to correspond with the highest quality of manna, however, this was not the case. Both N2 and S2 have lower SUC:RAF and therefore manna quality (0.37 and 0.35 respectively) compared to other populations (SUC:RAF varied between 0.33 to 0.50) (Table 6). It is likely that another factor is more important for the pardalotes distribution around Bruny Island, such as habitat requirements. Both N2 and S2 populations had the largest averages of tree trunk diameter of all sampled populations (Table 6). This is consistent with previous research of the pardalote species, which has identified mature E. viminalis trees and nest cavities as essential requirements for the species (Bryant 2010; Edworthy 2016; Woinarski and Bulman 1985).

The trait of high quality manna was adopted by work of Swart (2017), who explored the difference between primary, secondary and unused trees and found a significance in this SUC:RAF and therefore manna quality, seeing increase quality manna in primary tree when compared to secondary. This indicated a choice for trees with higher quality manna. The primary tree referred to time spent foraging, and pardalotes spent around 80% of time farming manna in these trees. The primary trees weren’t always nest trees for the birds in Swarts study, and coincidentally usually had the highest tree trunk diameter of all other trees within the colonies territory. Swarts work also had lower scaled manna quality then in this study, where primary tree manna varied between 0.29-0.36 SUC:RAF, and in this research the highest quality manna was 0.50 SUC:RAF (population N3). The populations N2 and S2 had manna that were considered high quality by Swarts research, but fairly low in comparison to populations across this study. The results of populatiosn N2 and S2 have high quality manna

38 across the scale presented in Swarts research, a result may make sense considering both populations at Dennes hill and Inala Nature Reserve were sampled by Swart as well.

Future directions and Implications of research An area for future research could focus on species distribution models in relation to manna quality. At what level would food quality influence the pardalotes distribution? Previous research has identified habitat requirements as a crucial element for the pardalotes presence in any given community, but food resources are also important for a species survival. Manna has been identified as essential but previous work (Case and Edworthy 2016), but to what extent in quality causes choices for and against manna? Is the quality of manna required for the species lower than the average in this study? Or is it simply that increased amount of sucrose in the manna not important to the species?

A question worth investigating in the future may examine if manna from branches in buckets vary from manna on trees? Does this method have any consequence on the manna? Research in manna composition is so new that our results do not signify any abnormalities to indicate whether this may be an issue. Swart (2017) method was similar, as branches were also taken off trees for extraction. The obvious reason for this is that mature E. viminalis trees are very tall and would require a method that includes tree climbing if a method similar to the common-garden field trials were used. Another area worth investigating is quantity of manna. The quantity of manna was initially attempted to be studied, but due to the initial difficulties extracting manna in the common-gardens, and the subsequent variations in extraction method, it was later removed.

The main objective of this study was to explore mechanisms that may relate to manna variation in E. viminalis, and aid to further understanding ofthe relatively understudied resource. of the finding of significant spatial variation in manna quality is an important result and signifies there is significant variation in manna intra-specifically. This will pave the way for further exploration as to what may be controlling manna variation, and how it may arise. Manna quality was associated with annual heat moisture index and therefore environmental variables may be a mechanism controlling variation in manna. However, this did not account

39 for all the variation seen in manna quality. Small sample sizes may have hindered some results and therefore replication and further exploration in these variables are essential.

These results indicate future research must focus on the potential impacts of climate change. The forty-spotted pardalote is a species already vulnerable to climate change (Webb et al. 2019) and the effects of climate change on the endangered species could be devastating. Therefore, this study may help aid future management and conservation practices in relation to the pardalote. The results in manna composition and concentration may also aid in the establishment of feeding trials.

REFERENCES Arsenault R, Owen‐Smith N (2002) Facilitation versus competition in grazing herbivore assemblages. Oikos 97, 313-318.

Bailey JK, Hendry AP, Kinnison MT, Post DM, Palkovacs EP, Pelletier F, Harmon LJ, Schweitzer JA (2009) From genes to ecosystems: an emerging synthesis of eco-evolutionary dynamics. The New phytologist 184, 746-749.

Bailey JK, Wooley SC, Lindroth RL, Whitham TG (2006) Importance of species interactions to community heritability: a genetic basis to trophic‐level interactions. Ecology letters 9, 78-85.

Barbour RC, O'Reilly-Wapstra JM, Little DWD, Jordan GJ, Steane DA, Humphreys JR, Bailey JK, Whitham TG, Potts BM (2009) A geographic mosaic of genetic variation within a foundation tree species and its community-level consequences. Ecology 90, 1762-1772.

Basden R (1965) The occurrence and composition of manna in Eucalyptus and Angophora. 90, 152- 6.

Baude M, Leloup J, Suchail S, Allard B, Benest D, Meriguet J, Nunan N, Dajoz I, Raynaud X (2011) Litter inputs and plant interactions affect nectar sugar content. Journal of Ecology 99, 828-837.

Bentley BL (1977) Extrafloral Nectaries and Protection by Pugnacious Bodyguards. Annual Review of Ecology and Systematics 8, 407-427.

Bertsch A (1983) Nectar production of Epilobium angustifolium L. at different air humidities; nectar sugar in individual flowers and the optimal foraging theory. Oecologia 59, 40-48.

Bolnick DI, Amarasekare P, et al. (2011) Why intraspecific trait variation matters in community ecology. Trends in Ecology & Evolution 26, 183-192.

Boose DL (1997) Sources of variation in floral nectar production rate in Epilobium canum (Onagraceae): Implications for natural selection. Oecologia 110, 493-500.

Bowman WD, Steltzer H, Rosenstiel TN, Cleveland CC, Meier CL (2004) Litter effects of two co‐ occurring alpine species on plant growth, microbial activity and immobilization of nitrogen. Oikos 104, 336-344.

Bryant S (2010) Conservation assessment of the endangered forty-spotted pardalote 2009–2010. Report to Threatened Species Section, DPIPWE and NRM South, Hobart.

Burkle LA, Irwin RE (2009) The effects of nutrient addition on floral characters and pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant Ecology 203, 83-98.

Carroll AB, Pallardy SG, Galen C (2001) Drought stress, plant water status, and floral trait expression in fireweed, Epilobium angustifolium (Onagraceae). American Journal of Botany 88, 438-446.

Case SB, Edworthy AB (2016) First report of ‘mining’ as a feeding behaviour among Australian manna-feeding birds. Ibis 158, 407-415.

Castellanos-Acuña D, Vance-Borland KW, St. Clair JB, Hamann A, López-Upton J, Gómez-Pineda E, Ortega-Rodríguez JM, Sáenz-Romero C (2018) Climate-based seed zones for Mexico: guiding reforestation under observed and projected climate change. New Forests 49, 297-309.

Chambel MR, Climent J, Alía R, Valladares F (2005) Phenotypic plasticity: a useful framework for understanding adaptation in forest species. Forest Systems 14, 334-344.

Douma JC, Weedon JT (2019) Analysing continuous proportions in ecology and evolution: A practical introduction to beta and Dirichlet regression. Methods in Ecology and Evolution 10, 1412-1430.

Edworthy AB (2016) Competition and aggression for nest cavities between Striated Pardalotes and endangered Forty-spotted Pardalotes. The Condor 118, 1-11.

Ferrari S, Cribari-Neto F (2004) Beta regression for modelling rates and proportions. Journal of applied statistics 31, 799-815.

Flato G, Boer G, Lee W, McFarlane N, Ramsden D, Reader M, Weaver A (2000) The Canadian Centre for Climate Modelling and Analysis global coupled model and its climate. Climate Dynamics 16, 451- 467.

Foden WB, Butchart SH, et al. (2013) Identifying the world's most climate change vulnerable species: a systematic trait-based assessment of all birds, amphibians and corals. PloS one 8, (6).

Frye GG, Connelly JW, Musil DD, Forbey JS (2013) Phytochemistry predicts habitat selection by an avian herbivore at multiple spatial scales. Ecology 94, 308-314.

Gardener MC, Gillman MP (2001) The effects of soil fertilizer on amino acids in the floral nectar of corncockle, Agrostemma githago (Caryophyllaceae). Oikos 92, 101-106.

Gaucher SP, Leary JA (1998) Stereochemical differentiation of mannose, glucose, galactose, and talose using zinc (II) diethylenetriamine and ESI-ion trap mass spectrometry. Analytical chemistry 70, 3009-3014.

Gaze P, Cloud M (1983) Honeydew and its importance to birds in beech forests of south island, New Zealand. New Zealand journal of ecology 6, 33-37.

Gowda JH (1996) Spines of Acacia tortilis: what do they defend and how? Oikos, 279-284.

Harrison PA (2017) Integrating climate change into conservation and restoration strategies: the case of the Tasmanian eucalypts. University of Tasmania, Hobart

Hartig F (2019) DHARMa: residual diagnostics for hierarchical (multi-level/mixed) regression models. R package version 0.2.4 https://CRAN.R-project.org/package=DHARMa.

Herrera CM, Bazaga P (2011) Untangling individual variation in natural populations: ecological, genetic and epigenetic correlates of long‐term inequality in herbivory. Molecular ecology 20, 1675- 1688.

Jakobsen H, Kritjansson K (1994) Influence of temperature and floret age on nectar secretion in Trifolium repens L. Annals of Botany 74, 327-334.

Jurskis V (2016) ‘Dieback’(chronic decline) of Eucalyptus viminalis on the Monaro is not new, unique or difficult to explain. Australian Forestry 79, 261-264.

Lawler IR, Foley WJ, Eschler BM (2000) Foliar concentration of a single toxin creates habitat patchiness for a marsupial folivore. Ecology 81, 1327-1338.

Leiss K, Klinkhamer P (2005) Genotype by environment interactions in the nectar production of Echium vulgare. Functional Ecology, 454-459.

Lenth R (2019) Estimated Marginal Means, aka Least-Squares Means. R package version 1.4. https://CRAN.R-project.org/package=emmeans.

Li X, Blackman CJ, Choat B, Duursma RA, Rymer PD, Medlyn BE, Tissue DT (2018) Tree hydraulic traits are coordinated and strongly linked to climate‐of‐origin across a rainfall gradient. Plant, Cell & Environment 41, 646-660.

Macfarlane C, Adams MA, White DA (2004) Productivity, carbon isotope discrimination and leaf traits of trees of Eucalyptus globulus Labill. in relation to water availability. Plant, Cell and Environment 27, 1515-1524.

McLean EH, Prober SM, Stock WD, Steane DA, Potts BM, Vaillancourt RE, Byrne M (2014) Plasticity of functional traits varies clinally along a rainfall gradient inEucalyptus tricarpa. Plant, Cell & Environment 37, 1440-1451.

Mitchell RJ (2004) Heritability of nectar traits: Why do we know so little? Ecology 85, 1527-1533.

Nadezhdina N (1999) Sap flow index as an indicator of plant water status. Tree physiology 19, 885- 891.

Nicotra AB, Atkin OK, et al. (2010) Plant phenotypic plasticity in a changing climate. Trends in plant science 15, 684-692.

O'Reilly-Wapstra JM, McArthur C, Potts BM (2002) Genetic variation in resistance of Eucalyptus globulus to marsupial browsers. Oecologia 130, 289-296.

O'Reilly-Wapstra JM, McArthur C, Potts BM (2004) Linking plant genotype, plant defensive chemistry and mammal browsing in a Eucalyptus species. Functional Ecology 18, 677-684.

O'Reilly-Wapstra JM, Potts BM, McArthur C, Davies NW (2005) Effects of Nutrient Variability on the Genetic-Based Resistance of Eucalyptus globulus to a Mammalian Herbivore and on Plant Defensive Chemistry. Oecologia 142, 597.

Özder N, SAGLAM Ö (2013) The effects of temperature for development time, fecundity and reproduction on some ornamental aphid species. Journal of Central European Agriculture 14, 0-0.

Parachnowitsch AL, Manson JS, Sletvold N (2019) Evolutionary ecology of nectar. Annals of Botany 123, 247-261.

Paton DC (1980) The importance of manna, honeydew and lerp in the diets of honeyeaters. Emu 80, 213-226.

Paton DC (1982) The diet of the New Holland honeyeater, Phylidonyris novaehollandiae. Australian Journal of Ecology 7, 279-298.

Petanidou T, Smets E (1996) Does temperature stress induce nectar secretion in Mediterranean plants? New Phytologist 133, 513-518.

Price T (1987) Diet variation in a population of Darwin's finches. Ecology 68, 1015-1028.

Prober SM, Potts BM, et al. (2016) Climate adaptation and ecological restoration in eucalypts. Proceedings of the Royal Society of Victoria 128, 40-53.

Smithson M, Verkuilen J (2006) A better lemon squeezer? Maximum-likelihood regression with beta- distributed dependent variables. Psychological methods 11, 54.

Steinbauer MJ (1996) A note on manna feeding by ants (Hymenoptera: Formicidae). Journal of Natural History 30, 1185-1192.

Swart I (2017) Forty-spotted pardalote (Pardalotus quadragintus) tree use and the influence of tree and manna variation. University of Tasmania, Hobart

Team RC (2013) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/.

Tippett JT (1986) Formation and Fate of Kino Veins in Eucalyptus L'hérit. IAWA journal 7, 137-143.

Viscarra Rossel RA, Chen C, Grundy MJ, Searle R, Clifford D, Campbell PH (2015) The Australian three-dimensional soil grid: Australia.

Wallis I, Watson M, Foley W (2002) Secondary metabolites in Eucalyptus melliodora: field distribution and laboratory feeding choices by a generalist herbivore, the common brushtail possum. Australian Journal of Zoology 50, 507-519.

Wallis IR, Goldingay RL (2014) Does a sap feeding marsupial choose trees with specific chemical characteristics? Austral Ecology 39, 973-983.

Wang T, Hamann A, Yanchuk A, O'neill G, Aitken S (2006) Use of response functions in selecting lodgepole pine populations for future climates. Global Change Biology 12, 2404-2416.

Webb M, Alves F, Tulloch A, Shaw J, Bryant S, Stojanovic D, Crates R, Heinsohn R (2019) All the eggs in one basket: Are island refuges securing an endangered passerine? Austral Ecology 44, 523-533.

Whitham TG, Bailey JK, et al. (2006) A framework for community and ecosystem genetics: from genes to ecosystems. Nature Reviews Genetics 7, 510-523.

Woinarski J, Bulman C (1985) Ecology and breeding biology of the forty-spotted pardalote and other pardalotes on North Bruny Island. Emu 85, 106-120.

Wood SN (2017) 'Generalized additive models: an introduction with R (2nd edition).' (CRC press).

Zimmerman M (1983) Plant reproduction and optimal foraging: experimental nectar manipulations in Delphinium nelsonii. Oikos, 57-63.

Zuur AF, Ieno EN (2016) A protocol for conducting and presenting results of regression‐type analyses. Methods in Ecology and Evolution 7, 636-645.

Zuur AF, Ieno EN, Elphick CS (2010) A protocol for data exploration to avoid common statistical problems. Methods in ecology and evolution 1, 3-14.

SUPPORTING MATERIAL

Appendix 1: Sugar proportion variations across populations of Eucalyptus viminalis manna on Bruny Island: Sugars represented include A) fructose (pcFRU); B) glucose (pcGLU); C) stachyose (pcSTA); D) sucrose (pcSUC); E) raffinose (pcRAF); and F) sucrose to raffinose ratio (SucTOraf).The green triangles indicate above mean results sugar proportions, and the red circles indicate below mean results. The larger the shape (triangle or circle), the further from the mean result the population is. Note that proportions of fructose and glucose (A and B) are too low to be listed in key.

LITERATURE REVIEW

Genetic, environmental and GxE control of sugar- based exudates used as food resources by animals

Amy Wing

May 2020

Department and University: School of Natural Sciences, University of Tasmania Supervisory team: Geoff While, Julianne O’Reilly-Wapstra and Peter Harrison

A literature review submitted in partial fulfillment of the requirements of the degree Bachelor of Science with Honours

Introduction

Animals rely upon plants for their survival, either directly or indirectly, through the resources or habitat they provide. The complexity of the vegetative community (i.e. the structure and composition of the flora) can often provide contrasting resources and habitat quality, thus shaping patterns of biodiversity (Bennett 2010; Litteral and Shochat 2017; Virgós 2002). Animals exploit their environment by forming interment relationships with their host plant (Altshuler 2003; Temeles 1996) . Antagonistic relationships often come at a cost to the host plant. Such is the case with many plant-herbivore interactions, where foraging can damage the plant, impact growth rates and the plants overall fitness (Vourc'H et al. 2002). Conversely, mutualistic relationships result in either neutral (i.e. no detrimental effect) or beneficial responses between the host species and animal (Boucher et al. 1982). A direct mutualistic relationship is seen where both species benefit from the interaction such as the interactions between pollinators and flowers, pollination occurs by insect (or other animals) when collecting nectar and pollen from multiple flowers (Thomson et al. 1989). This process has direct impacts of dispersing pollen for the flower and providing a food resource for pollinator. Another excellent example is the relationship between plants investment in extrafloral nectaries to encourage ant protection against herbivory on reproductive traits (Bluthgen et al. 2000; Ness et al. 2009; Zhang et al. 2015). Indeed, there are many examples demonstrating the close evolutionary history between animal species and the environment they exist in (Macior 1971).

It is unclear whether antagonistic and mutualistic relationships will decouple with their host- species as environments change (i.e. with climate change). As climates warm, it is predicted species will disperse to follow their optimal environmental niche (Suppiah et al. 2007; Thuiller et al. 2011; Zhang et al. 2015). Already, there is clear evidence for increased elevation and latitudinal migration of both plants and animals, as well as effects on biotic interactions within communities and ecosystems (Chen et al. 2011; Feeley et al. 2013; Guillaumet 2017; Parmesan and Yohe 2003; Serra‐Diaz et al. 2016; Thomas 2010; Thuiller et al. 2005). However, in the case of long-lived and sessile forest tree species that are often foundation species of an ecosystem, there is a disparity between their ability to disperse verses the pace required to track the optimal environmental niche (Aitken et al. 2008). It is expected tree species may

48 either persist, adapt, or die as their home-site environment becomes maladaptive (Aitken et al. 2008). Indeed, there is already range-wide dieback across the world (Allen et al. 2010) and within Australia (Hoffmann et al. 2019). Facilitated migration within (assisted migration) and outside (assisted colonisation) a species range has been proposed as an approach to mitigate such dieback events as climates warm (Prober et al. 2016). However, we have little knowledge on how to move species around and the broader implications such translocations will have on dependent organisms.

One step towards understanding the potential implication of facilitated migration on dependent organisms is through gaining an understanding of the quality and quantity different resources offered by tree species vary through space and time. Variation in phenotypic expression of traits is fundamentally influenced in one of three ways. First, traits can vary based on genotype, the underlying genetic composition of an organism. Second, the trait variation may be due to environmental variation, such as environmental pressure on the organism throughout its development. Third, an interaction between genotype and environment (hereafter GxE) may occur when, for example, the response to the environment depends on different genetic Figure one: a conceptual diagram representation of the effect of genetic, environmental, and genetic by backgrounds (see Figure 1, Nicotra et al. environmental interaction on the production of 2010). Within species variation in the anthocyanins in a leaf (reproduced from Nicotra et al. 2010). The response is illustrated under two sets phenotype of plants can be seen both of environmental conditions (Env1 and Env2). within- and between- populations, as well Picture 1 visualises phenotypic plasticity, where the influence of environment is the mechanism for as within individual plants, however this phenotypic expression. Picture 2 shows genetic review will primarily focus on within- variance as the mechanism that controls phenotypic expression. Picture 3 shows an interaction between population variation on plant traits. genetic and environmental (GxE) factors explaining phenotypic expression. 49

Plant traits which provide an unproportionally higher resource content for many animals are those with high sugar contents. Resources such as nectar, honeydew, manna and lerp have high glucose and sucrose contents (Basden 1965; Basden 1966; Basden 1967; Basden 1970; Paton 1980), which feed large numbers of avian species and insects, as well as some mammals and reptiles (Carthew et al. 1999; Ford and Paton 1982; Olsson et al. 2000; Paton 1980; Steinbauer 1996; Woinarski et al. 1989). Foliar herbivory may decrease the plants overall fitness (Rausher 2001; Vourc'H et al. 2002). However, that is not likely the case with these sugary exudates (nectar, honeydew, manna and lerp). In case of nectar, consumption by animals is encouraged to fulfil reproductive cycles (Parachnowitsch et al. 2019). Consequently, the foraging of these positive plant traits rarely shows negative effects on the plant’s overall fitness, and in many cases, positive mutualistic relationships have evolved.

To understand the ecological and evolutionary relationships between animals and plants it is important to understand what underpins the variation in these resources. The variation in the quality and quantity of resources not only has direct effects on the herbivores feeding on the plant, but has flow-on consequences on community composition and ecosystem functioning (Whitham et al. 2006). This is particularly true in the context of foundation species, which are relied upon by an unproportionally high amount of animals compared to other resources (Ellison et al. 2005), and variation in these traits will have cascading effects on individuals or whole populations (Bailey et al. 2004; Barbour et al. 2009; Ellison et al. 2005). Recent studies have shown that traits can have predictable effects on community structure and ecosystem processes, particularly if that trait has a heritable component (Whitham et al. 2006). Species do not evolve in a vacuum (Whitham et al. 2006), and therefore it is important to understand the whole picture when dealing with community functioning and foundation species. Understanding how traits vary and the mechanism that controls this has great importance and can be used in the fields of conservation and restoration.

For the purposes of this review, discussion will focus around the sugar-based exudates of nectar, honeydew, manna and lerp. Pollen has been excluded, despite its high importance to

50 many nectar consumers, as it is not a sugary-based trait. Phloem sap has also been excluded from this review, despite being consumed by aphids and other insects which produce honeydew and lerp, as the complexity of interaction between host tree species, aphid food source and also biotic and abiotic factors, is beyond the scope of this study.

While many species have examined genetic and environmental influences on some plant traits (i.e. plant secondary metabolites (Bailey et al. 2005; Close et al. 2003; Mutikainen et al. 2000; O'Reilly-Wapstra et al. 2004; O’Reilly-Wapstra et al. 2005)) significantly less is known about variation in sugar-based plant traits. Such is the case with manna and lerp, two traits which are key resources to many birds and insects, and endangered animals (Paton 1980), have received little attention, with only a few studies examining their chemical composition, let alone variation (Basden 1965; Basden 1966; Basden 1970; Woinarski et al. 1989). Therefore, the aim of this review is to bring together the existing literature on genetic and environmental variation in sugar-based plant traits including nectar, honeydew, manna and lerp, which are important resources to a wide range of animals. Focussing on the evidence for environmental and genetically-based phenotypic variation in these traits, this review aims to identify key research gaps by summarising the literature on nectar, honeydew, manna and lerp, and then apply nectar and honeydew as “model systems” to manna and lerp, as they have received substantially more investigation. Knowledge of how these traits vary and respond to genetic and environmental constraints will provide information to aid future predictions and hypothesis about the resource variation, and contribute to predictions in restoration, conservation and management.

Nectar

Nectar is a sugary secretion produced by the nectary glands of plants. It contains primarily sugars, such as sucrose, glucose and fructose as well as a small amount of carbohydrates, amino acids and volatiles. Nectar is an obligatory part of the diet of a diverse range of animals, including bees, ants and many small birds, such as honeyeaters and hummingbirds, as well as some reptiles and marsupials (Bond and Brown 1979; Bradshaw and Bradshaw 2001; Oliver 2001; Olsson et al. 2000). Nectar also has large economic value to humans, being the main sugar resource for honey (Ball 2007). One characteristic of nectar is that it experiences diurnal variations (Bond and Brown 1979; Pleasants 1983), most likely to follow the activity of the pollinators it has co-evolved with (Canto et al. 2008). Nectary glands are present in various parts of the flower (floral nectaries), as well as outside of the flower (extrafloral nectaries), and the location of the nectaries on the plant is directly associated with their function. For example, the floral nectaries serve an important role during reproduction, providing a nutritional reward (Parachnowitsch et al. 2019) commonly at the base of the flower enticing pollinators to transport reproductive pollen between flowers. Extrafloral nectaries, however, serve a different but equally important function. They are features of plants that either distract insects that aren’t their target pollinators (Wagner and Kay 2002), or encourage more predatory species, such as ants, to control foraging herbivores that may otherwise feed on reproductive or floral parts (Bentley 1977; Ness et al. 2009). Indeed, this strategy for nectar defence other than nectar toxins (i.e. the investment in alkaloid content) may constitute a less energetic pathway to deter non-specific pollinators (Adler 2000; Adler and Irwin 2005).

Box 1: Nectar traits

Some nectar traits discussed within this review include:

Nectar production rates (NPR): the rate in which nectar is produced and/or replenished after removal from a flower.

Nectar concentration: The relative amounts (%) of different chemicals that make up nectar.

Sugar concentration: The concentration of sugars existing within the nectar.

Total sugar: the total sum of sugar present within the nectar. Genetic variability Nectar volume: the mean volume of nectar over a 24 hour period. There has been extensive investigation in nectar production rates (NPR) and sugar composition (see box 1: Nectar traits), but outside of these nectar traits, nectar variation has

52 been studied very little. Studies have found that plants with higher rewards per flower may receive more visits from pollinators (Boose 1997; Real and Rathcke 1991; Thomson et al. 1989; Zimmerman 1983), resulting in pollinators either spending more time at each flower or spending less time at each flower but visiting more flowers (Campbell 1996; Zimmerman 1983), potentially increasing the diversity of pollen donors and reproductive success. However, for plant traits to evolve by selection, there must be sufficient phenotypic variation for pollinators to choose between, and an underlying heritability to these traits (Boose 1997; Falconer 1960; Via and Lande 1985). Within nectar, there is evidence of genetic-based variation of nectar traits, with medium estimates of heritability (Mitchell 2004; Parachnowitsch et al. 2019).

Studies have reported low heritability in nectar traits and speculated whether this reflected a true genetic signal. For example, Thomson et al. (1989), who posited that the strong differences observed in their study was due to pollinator visitation and subsequent revisits speculated the slightest variation in pot watering may play a factor as well. Campbell (1996), who conducted a field study, produced low heritability results, but noted results were swamped with environmentally induced phenotypic variation. The limited information on genetic variation within nectar traits, as mentioned in Mitchell (2004) review is likely due to the often large effects of environmental variation seen in the field that clouds genetic signals (Boose 1997; Campbell 1996; Mitchell and Shaw 1993). The difficulty in assessing nectar traits, unwitting ability for cross contamination in green house studies, and the diurnal variability pattern in nectar production, together constrain studying genetic variability in nectar (Mitchell 2004). However, as shown in figure 2, studies have overcome this constraint. In a field study, Liess et al. (2004) showed that high and low nectar production was partly under genetic control. Within glasshouse experiments, heritability’s have also been estimated. Mitchell and Shaw (1993) found significant differences between narrow-sense and broad- sense heritability estimates, indicating additive genetic variation (see box 2) of both nectar volume but not nectar concentration.

A review published earlier this year (Parachnowitsch et al. 2019), which reported the total of all heritability estimates for nectar production within the literature, included just eight studies that listed nectar trait heritability’s. The eight studies noted heritability estimates (figure 2), and although the paper calls for more research to make synthetic analysis of nectar heritability, there is suitable evidence to suggest at least NPRs are often heritable (with an average NPR heritability of 0.31) (Boose 1997; Campbell 1996; Klinkhamer and van der Veen- van Wijk 1999; Leiss et al. 2004; Vogler et al. 1999). Outside of NPR, as can be seen in figure 2, there only a few studies that have presented heritability of other nectar traits (Campbell 1996; Kaczorowski et al. 2008; Vogler et al. 1999; Zu and Schiestl 2017). Therefore, the evidence suggests that the nectar trait of NPR may be heritable , but there is not enough information about other nectar traits to see if this applies to other traits. (Mitchell 2004; Parachnowitsch et al. 2019).

Figure two: All heritability in nectar traits reported in the literature (figure 2 of Parachnowitsch et al. 2019). Estimates by crosses (circles, n = 8), clones (diamonds, n = 8) and response to selection (triangles, n = 6). Symbols in black are statistically significant at P < 0.05. NPR = nectar production rate. Data from a total of eight studies are include results from (Mitchell and Shaw, 1993; Campbell, 1996; Boose, 1997; Klinkhamer and Wijk, 1999; Vogler et al., 1999; Leiss et al., 2004; Kaczorowski et al., 2008; Zu and Schiestl, 2017)

Box 2: Glossary

Additive genetic variance involves the inheritance of certain alleles from parents and this allele has independent effect on the specific phenotype, which will cause the phenotype deviation from the mean phenotype.

Broad-sense heritability is the ratio of total genetic variance to total phenotypic variance. H2 = VG/VP.

Narrow-sense heritability is the ratio of additive genetic variance to the total phenotypic variance.

Heritability includes both broad-sense and narrow-sense heritability and is the genetics an individual has gained from parents.

Environmental variability

Environmental effects are the largest source of variation within nectar traits, as nectar traits show high environmental sensitivity (Boose 1997). Nectar traits respond plastically when exposed to varying environments, such as differing levels of soil nutrients, precipitation and humidity. This response to varying conditions can translate to enhanced fitness (Chambel et al. 2005; Nicotra et al. 2010), which is particularly important for long-lived plants such as trees (Bradshaw 1965). NPR and nectar volume has been shown to increase with increasing water availability (Boose 1997; Carroll et al. 2001; Leiss and Klinkhamer 2005; Waser and Price 2016; Zimmerman 1983) and soil nutrient levels (Baude et al. 2011). Conversely, decreased nectar production has been associated with shading (Zimmerman and Pyke 1988) and the droughting of plants (Carroll et al. 2001). The fungal composition of the soil communities shows varied effects. In one instance on Polemonium viscosum, fungicide treatments yielded greater volumes of nectar (Becklin et al. 2011), whereas another study found that mycorrhizal fungi increase nectar sugar content in Tagetes erecta, but not in two other annual plant species in the study (Gange and Smith 2005). Humidity has also documented positive correlation with increasing nectar volume (Bertsch 1983). An increase in CO2 is observed to have complex effects on nectar traits, with nectar volume observed decreasing in the presence of elevated CO2 (Rusterholz and Erhardt 1998). Contrastingly, NPR has been observed increasing in Epilobium angustifolium when exposed to elevated CO2 (Erhardt et al. 2005), and had no effect on legume plants (Rusterholz and Erhardt 1998). On top of that, increased temperature

55 has been noted to have positive correlations with NPR, between 10°C and 18°C (Jakobsen and Kritjansson 1994), and up to 38°C in Mediterranean plants (Petanidou and Smets 1996).

Pollinator interaction with plants influences some nectar traits. For example, pollinator sounds may influence an increase nectar sugar concentration (Veits et al. 2019), where certain frequencies induced plant responses in as little as three minutes (Veits et al. 2019). Age and flower shape influences pollinator choice. For example, Bond and Brown (1979) found in Eucalyptus impressor and honeybee interactions a preference of old flowers foraging effecting pollinator choice, as flower age has an effect over nectar traits. A similar result was seen by Southwick and Southwick (1983), who noted peak nectar volume on the second day of the flowering period and a decrease in nectar sugar concentration as the flowering period continued (Southwick and Southwick 1983).

The production of nectar additionally has a temporal component. Ipomopsis aggregate also showed variation over seasons with significantly higher NPR during summer compared to later in the season, however, sugar content between seasons did not differ (Pleasants 1983). Such seasonal patterns in the production of nectar is believed to be linked with the timing of pollinator activity levels, as well as pollinator visitation and subsequent revisits (Thomson et al. 1989).

Genetic by environmental (GxE) variation

Boose (1997) was among the first to demonstrate a genetic by environment (GxE, figure 1) interaction in nectar production rates. Nectar production traits had significant responses to changes in water and light availability, but these responses were not equal, despite being grown in a glasshouse and having a large control over external features. Boose (1997) concluded that a high-nectar producing genotype in one environment may not be high when that genotype is tested in another environment, potentially indicating a link between plasticity and genetics.

The interaction between genetic and environment shaping variation in nectar traits is underappreciated and possibly has a much larger influence over nectar variation, despite the huge phenotypic variation seen due to the environment. Nectar production traits had significant responses to changes in water and light availability, but that they were not equal in their responses, despite being gown in a glasshouse and having a large control over external features. Boose (1997) concluded that a high-nectar producing genotype in one environment may not be in another, indicating that genotypes may be suited to different environments. Leiss and Klinkman (2005) supported this, hypothesising that there were selective advantages to having both an increased plasticity to water availability and nectar production, as well as a higher nectar producing type of genotype within the same environment.

Honeydew

Honeydew is made up of primarily sugars, such as glucose, sucrose and xylose (Völkl et al. 1999), and is a sugary secretion of nymphal staged aphids, coccids and psyllids, which feed on tree sap (Gaze and Cloud 1983). It is a particularly important food resource for ants, other insects, birds and even some marsupials (Holland et al. 2007; Paton 1980). In particular, honeydew is important as it is a carbohydrate present all year round and suggested to be a substitute for nectar in the off seasons (Murphy and Kelly 2003), and also shows diurnal patterns as noted in nectar. Honeydew is produced by aphids and scale insects, which consume sap directly from phloem transport, and excrete honeydew as a bi-product from their anus (Tjallingii 1995). The end product either falls from the insect onto bark or the forest floor, or is plucked straight from their body, an action formed by ants in a mutualistic relationship (Detrain et al. 2010). The honeydew traits of production rate, quantity and quality (I.e. sugar concentration and composition) can all covary. Honeydew traits are known to vary in three ways; due to insect species (producing honeydew, and interactions), host tree species, and environmental variables (Dungan and Kelly 2003; Fischer and Shingleton 2001; Kelly et al. 1992; Murphy and Kelly 2003; Völkl et al. 1999).

Host tree species variation

The effect of host tree may be linked to photosynthetic capacity or specific variations within host tree traits such as sap. Fischer and Shingleton (2001) investigated how honeydew varied based on a poplus species, noting changes in sugar compositions with different tree species. Some trees were able to support more production of honeydew (more insects) than others in study by Dungan and Kelly (2003) on New Zealand Beech trees. There has been no studies that have investigated host-tree species genetic variation of honeydew on trees. A plausible hypothesis for host plant variation on honeydew may have to do with the nutritional quality of phloem sap, and therefore, the flow on effects this has on the aphid producing honeydew (Douglas 1993; Fisher et al. 1984).

Abiotic variation

Honeydew varies greatly with season, not unlike nectar. Murphy and Kelly (2003) documented seasonal variation in honeydew production in New Zealand on trees from the Fagaceae, or beech family, seeing the lowest quantity of honeydew in winter and the highest in spring. This result is un-surprising and correlates with increased foraging of honeyeater and similar birds, as well as honeybees, wasps and further pollinator activity. Moller and Tilly (1989) confirm seasonality of honeydew, seeing high quantity in late winter to spring and then a decline. Gaze and Clout (1983) also saw this trend but noted a peak in honeydew production again in Autumn, which isn’t always documented and could be due to tree species. Moller and Tilly (1989) also noted that the quality of honeydew drops on anal tubes were greater in winter/spring, indicating sugar concentrations may also be impacted by season, and Glaze & Clout (1983) noted a similar trend. This seasonal fluctuation is also seen in nectar, and could be either an indicator of environment, or pollinator activity.

Honeydew, Similarly to nectar, experiences significant diurnal variation. This variation is seen in both honeydew quality and quantity. Honeydew can become more concentrated as the day progresses (Gaze and Cloud 1983), which has been documented not only as an effect on honeydew quality but also quantity. When sugar concentrations were high, honeydew became thicker, and hung from the insect possibly inhibiting further flow, but when sugar concentration was low, honeydew flowed freely (Moller and Tilley 1989) these results of daily variation in honeydew standing crop are supported by Kelly et al. (1992).

Environmental effects such as home-site altitude and aspect may have large influences of honeydew quantity. Studies comparing insect infestations and resulting honeydew production at two sites of Nothofagus solandri saw higher quantities at lower altitudes (Kelly et al. 1992). Dungan and Kelly (2003) saw similar, site dependent results that were hypothesised to be linked with altitude. Aphids consume sap directly from phloem transport, and therefore there is importance in cell turgor and aphid uptake and production of sap (Tjallingii 1995). Crozier (1981) hypothesized aspect to be a contributing factor of high densities of honeydew threads, noting that sunnier aspects promote a more productive

59 environment and are favoured by honeydew producing insects (Crozier 1981). However, other studies studying the relationship between aspect and honeydew production did not see evidence of an association of aspect on honeydew traits (Kelly 1990; Kelly et al. 1992). This could be due to a range of factors, but most probably due to insect or host species presence and abundance.

Other environmental factors such as weather, temperature, humidity and CO2 have all showed effects on honeydew traits and honeydew producers. An increased temperature can lead to a more concentrated sugar content of the honeydew, and more humid conditions may dilute honeydew droplets (Gaze and Cloud 1983). Bark temperature has been shown to be a strong determinant of the behaviour of bark-dwelling beetles (Schmid et al. 1991), and Dungan and Kelly (2003) hypothesized it is possible that beech scale insects are similarly temperature regulated. Studies about the impact of carbon dioxide (CO2), and the resulting expectations of climate change, are beginning to emerge. Phloem sap-suckers such as aphids have a complex response to elevated CO2 (Newman 2004; Sun et al. 2010). Aphids feed exclusively on phloem sap and are very sensitive to changes in its composition (Himanen et al. 2008; Pritchard et al. 2007). In a study conducted by Sun et al. (2009), a greater quantity of honeydew was produced by aphids reared under elevated CO2 conditions than by those reared under ordinary CO2 conditions. However, climate change is associated with more than just increased CO2, and it is important to note that despite the indication of increased productivity, increase temperature appears to have negative effects on aphid productivity (Blanchard et al. 2019). Temperature shows positive effects on aphid productivity, to a certain extent, as different aphid species respond differently to environment (Özder and SAGLAM 2013). In a study on four aphid species, half the aphid species responded negatively in developmental speed to increased temperature, whereas the other half expressed increased developmental growth in a given timeframe and showed similar positive and negative associations in fecundity and adult longevity. In a general trend, insects respond positively (to a threshold) with increased temperature, showing increased growth and reproduction in warmer temperatures (Gilbert and Raworth 1996). Therefore, we would expect to see an increase in honeydew production in response to increased temperature and warmer weather. However, despite this probable link, there is limited literature on the effect of temperature

60 on honeydew production. A single study on ant-aphid-plant interaction due to temperature, suggested a threshold of 29°C and above temperatures where honeydew standing crop relative abundance decreased significantly compared to 23 and 27°C (Sagata and Gibb 2016). This response to temperature is quadratic, suggesting honeydew production has an optimum temperature depending on aphid species, showing low productivity at high and low temperatures.

Biotic environment

An important biotic factor that influences honeydew production is of course, the insects themselves. The effect of the honeydew producing insect is possibly the most studied variation in honeydew. However, honeydew producing insects can have a mutualistic relationship with ants, adding another biotic factor into the complex relationships that may lead to honeydew variation. The aphids that produce honeydew, invest effort into making it more favourable to ants, as ants can provide protection from predation (Völkl et al. 1999; Yao and Akimoto 2002). This indicates that both aphid and ant species can lead to variation within honeydew (Fischer and Shingleton 2001). Studies have found glucose tends to be lower in aphids raised under an ant mutualism than without (Fischer and Shingleton 2001), as ants responds positively with greater concentrations of melezitose (Kiss 1981; Völkl et al. 1999), which is a type of sugar. The gut microbial composition of aphids may be one of the reasons for aphid alteration of sugar content for ants. Davidson et al. (1994) determined the gut microbia in the digestive tract of aphids controlled for variations in sugar composition, however, this hypothesis was later refuted by Wilkinson et al. (1997), claiming it was simply the species involved that lead to variation in sugar content, not its internal gut microbial community (Davidson et al. 1994; Wilkinson et al. 1997).

This interaction between ants and aphids must be significant, considering the magnitude of this interaction, as ants and aphids co-occur in almost all environments. This relationship is also associated with greater yields of honeydew (Fischer and Shingleton 2001; Völkl et al. 1999), and can even lead to competition for ants between aphid individuals (Del-Claro and Oliveira 1993). This effect on honeydew quality and quantity being influenced by ant interaction has been widely replicated (Blanchard et al. 2019; Del-Claro and Oliveira 1993;

Detrain et al. 2010; Völkl et al. 1999), and therefore indicates its importance in influencing honeydew variation.

Manna and Lerp

Manna and lerp are sugar rich resources found frequently on and around Eucalyptus trees (Paton 1980). Manna is a sugary secretion that exudes from wounds on tree branches and stems (Paton 1980), and lerp is a protective covering of many psyllids, composed of carbohydrates (Basden 1970). They vary in both the mechanisms in which they arise, as well as their chemical composition. Lerp, similarly to honeydew, is produced by passing through the alimentary system of an insect. Honeydew is a waste product, and lerp is produced to cover the insect from external weather. Interestingly, it has been suggested that lerp covered psyllids can produce honeydew whilst underneath lerp, implying that the function of honeydew production and lerp production can co-occur. It is unknown what mechanism controls how manna arises, but it appears to be a reaction of the tree in response to wounds.

Manna

Manna, as defined by Basden (1965), is the saccharine secretion of trees themselves, are the product of tree wounds, and are an attractive resource to ants, birds, sugar gliders and other animals (Holland (Carthew et al. 1999; Holland et al. 2007; Paton 1980; Recher et al. 1985; Recher 2016; Smith 1982; Steinbauer 1996). Manna exudes as a clear-whiteish liquid and crystalize into small white globs (around 0.5mm- 5mm diameter in general) upon contact with the air. From the reviewed literature manna has only been observed within the Myrtaceae (Myrtle) family, on several species of Eucalyptus (including, but not limited to, E. viminalis, E. Paucifora, E. panctata, E. maculate, E. manifera) and two Angophora species, A. floribunda and A. costata. Although manna’s chemical composition is known to vary between tree species (see table 1), manna consists of about 60% sugars, 16% water, 20% pectin and uronic acids, and a small amount of ash (Basden 1965). Little of manna has been investigated. Except for a few past studies on manna chemical composition and occurrence, nothing of the genetic and/or environmental factors relating to manna variation are known.

Manna and lerp have been confused in past by biologists, and up until the last half of the 20th century, the differences between manna and lerp had not clearly been defined. However, manna and lerp, which have markedly different chemical compositions and mechanisms of

63 arrival. Basden determined manna only occurs through wound inflicted by insect, as attempts at inducing a manna response from the tree were unsuccessful. However, recently manna has been shown to be produced by the wounds on E. Viminalis trees by the Forty-spotted pardalote, Pardalotus quadragintus, in Southern Tasmania (Case and Edworthy 2016)(see box 3). The mining of manna by the Forty-spotted pardalote is a unique case and is one of the only instances in which manna has been produced by another animal except for small insects. Case and Edworthy (2016) could replicated the action in which the pardalote creates wounds, and created wounds using a sharp scalpel blade by hand. The ability to create manna by scalpel blade can occur in E. viminalis and E. pauciflora (Swart 2017) and therefore possibly in more trees, but this has yet to be documented. Basden (1965) showed that manna is more than just tree sap. It consists of sugars such as raffinose, stachyose, Melibiose and Sucrose (see table 1) (Basden 1965). Basden hypothesized an enzyme in saliva may be influencing the production of manna. However, now that manna can be created with a scalpel blade, this is unlikely. Basden also hypothesized with less conviction that it may not be saliva but the presence of a microorganism indirectly to the wound site. However, Basdens hypothesis of insect saliva causing the manna response has been generally accepted (Smith 1982), we now know that this is untrue. Based on Basdens (1965) and Case and Edworthys (2016), manna may only be produced with very fine wounds, possibly only a few cells thick.

Although there is little research on manna, manna can vary in form of production rates, as well as quantities varying based on season (Basden 1965). Spring and early summer saw greater quantities of manna than winter, coinciding with rapid tree leaf growth (Basden 1965). Aspect may also have effected manna quantity, seeing higher densities of manna on the side of the tree facing the sun than away, possibly relating to sap flow.

Box 3: Manna

1 2

Picture 1 shows a Forty-spotted Pardalote eating manna from a tree which it possibly induced (Sourced from http://www.graemechapman.com.au/library/viewphotos.php?c=722&pg=1). Picture 2 shows manna exuding from a tree stem (indicated by white arrow) (photography by Wing (2019)).

Lerp

Lerp is the protective coating found over many Australian psyllids (Paton 1980). Lerp, and the insects producing it, are an important part of many animal diets, particularly small birds (Oliver 1998; Paton 1980; Recher et al. 1985). Lerp like honeydew, has passed through alimentary tract of an insect consuming phloem sap and excreted from the anus (Basden 1966; Basden 1970). Some species that create lerp coverings also excrete honeydew (Basden 1970). Lerp and manna have similar chemical compositions (see table 1), and both lerp and manna are soluble in water and completely breaks down in the stomach of animals consuming it (Basden 1966) making investigation of the plant resource difficult.

Lerps can vary in a few ways. Notably the biggest variation is in physical shape, where lerps can either be round, cotton like, and even fern shaped (See figure 3) (Basden 1970) depending on the insect involved in their arrival. Due to the vast diversity of psyllid species (Yang and

Raman 2007), it is most probable that lerp shape is controlled by the insect producing it. They also have varied chemical composition, particularly sugar composition, and it has been noted that birds show preferences to different lerps based on size, shape and chemical composition (Woinarski et al. 1989).

A B C

D E F

Figure three: Variation in lerp shape: A) Shows a basket lerp, with identified (Cardiaspina spp.). This leaf may be red in colour due to the lerp consuming chlorophyll while sucking out sap B and C) fern shaped lerp of a psyllid nymph; D & E) shows common lerp (also referred to as cotton-like or fibrous lerp). E shows ant guarding lerp and possibly harvesting honeydew from developing lerp. F) shows scallop lerps. A and D sourced from http://oneminutebugs.com.au/gum-tree-lerps/; B and C sourced from https://aussiebugs.wordpress.com/2016/02/15/psyllids-lerps/ E sourced from http://oneminutebugs.com.au/gum-tree-lerps/ F sourced from https://aussiebugs.wordpress.com/2016/02/25/scallop-shell-lerps/

Table 1: Manna and lerp percentage composition: Lerp specific to lerp of Eucalyptolyma Maidenii Froggatt from table 1 in (Basden 1966). An average of manna composition referred to in Table 1 of Basden (1965), listing manna composition of two Eucalyptus species (E. maculate and E. punctate) and angophorous costata. Manna composition of E. viminalis from table 1 in Swart (unpublished data). Note that information from Basden (1965) and Swart (unpublished data) had to be converted from total sugar concentration to total chemical composition, with sugars equalling 60% of chemical composition.

MANNA LERP E. viminalis E. maculata E. punctata Angophorous sp. Eucalyptus sp. STACHYOSE 2.3 - 1.2 6 Present RAFFINOSE 41.2 48 48 39 ~70 SUCROSE 15.6 3.6 12 Present GLUCOSE & <1 2.4 4.8 3 Present FRUCTOSE MELIBIOSE - 6 - - ~10 MOISTURE 16 16 16 16 3.2 PECTIN AND 20 20 20 20 9.4 URONIC ACIDS ASH 4 4 4 4 <1 SOURCE Swart Basden Basden (1965) Basden (1965) Basden (1966) (unpublished (1965) data)

Further study

The importance of manna and lerp greatly exceeds being ‘substitutes’ in animal’s diets when nectar sources are scares (Paton, 1980), and should be considered as primary food sources (Case and Edworthy 2016; Holland et al. 2007; Oliver 1998; Oliver 2001; Steinbauer 1996). Therefore, it is imperative to gain a deeper understanding of these resources, how they vary and what influences this variation. Investigation in manna and lerp has had little attention in recent literature. Significant improvements to their identification and chemical composition was made 50-60 years ago, but scientific studies have slowed since then (Basden 1965; Basden 1966; Basden 1967; Basden 1970; Woinarski et al. 1989). An important aspect of investigating important manna and lerp traits involves their digestibility and solubility in water. They are completely digested within the stomachs of animals, and therefore traditional faecal studies cannot be implemented (Basden 1965; Basden 1966; Basden 1970; Smith 1982;

Woinarski et al. 1989). Despite very little research focussing solely on manna or lerp, a great extent of literature has mentioned the presence of these resources in animal diets (Holland et al. 2007; Oliver 2000; Paton 1980; Paton 1982; Recher et al. 1985; Woinarski 1984). Therefore, given its prevalence in the diets of animals, investigation into these resources and the mechanisms controlling variation in the sources is important in providing a greater understanding of their role in animal diets, and flow-on effects in the ecosystem.

Within nectar and honeydew, we see trait variation of quality and quantity is influenced by both environmental and genetic factors. When comparing nectar and manna, both food sources are secreted from plants and have high levels of sugars. Nectar is highly sensitive to environmental conditions, but variation is also influenced by genetics, particularly in heritability, and interactions of GxE. Due to their similarity, it is likely that further research into manna production may find comparable reactions to the environment and genetics. Previous studies have found that manna composition varies between species, however it may also vary within species (Basden 1965). As with nectar, environmental conditions, such as humidity, precipitation or seasonality, may show plasticity within manna. Considering the large seasonality component of many traits other than nectar (such as honeydew, pollen, and plant growth investment), investigation of seasonality in manna would also be important in understanding how and why manna varies.

Much like honeydew, lerp is made by sap passing though the alimentary system of aphid or psyllid insects. This similarity in the production of lerp and honeydew makes them reasonable candidates for comparison, and both traits may have similar mechanisms controlling their variation. Honeydew traits are largely controlled by the biotic environment with insects responsible for producing the food source; lerp has also been shown to vary due to biotic factors such as aphid species. The abiotic environment was also found to influence honeydew traits, and whilst this abiotic influence is not known in lerp, it is plausible that lerp is also influenced in a similar way.

Conclusion

Plant trait variation in its simplest forms can occur due to one of three factors; the environment, genetics, or a combination of the two. This review identifies how variation in nectar, honeydew, manna and lerp occurs and discusses where literature is sparse. Nectar has been studied extensively in some areas (including production rates and sugar composition) and has been shown to vary due to genetic and environmental factors. Honeydew, however, sees very little research in trait variation underpinned by host plant genetic factors. Honeydew does experience large levels of variability due to the insect producing it and environmental factors, and despite any evidence of host tree genetic variation of honeydew, it is plausible to hypothesise that at least some variation may be underpinned by genetics or GxE. Similarly, very little about manna and lerp variation has been investigated, apart from some host-tree species and insect producing induced variation. We can use what has been learned about nectar and honeydew as model systems and apply this knowledge to how manna and lerp may potentially vary.

It is important to have a wholistic understanding of the underlying mechanisms of traits variations and the flow-on effects these resources have on the ecosystem, so that predictions can be made about future impacts. Factors controlling manna and lerp variation is currently under-investigated and applying the knowledge of how nectar and honeydew vary (considering the similarities seen between the traits) may be a good area for investigation in the future. From here, studies need to consider the flow on effects trait variation can have; not only on the fitness of individuals, but the whole ecosystem. With the increased pressures due to climate change expected to rise as this century progresses, being able to predict how crucial and foundation resources may respond will give conservative and restorative practices the necessary knowledge to make proactive, not reactive, management decisions.

References

Adler LS (2000) The ecological significance of toxic nectar. Oikos 91, 409-420.

Adler LS, Irwin RE (2005) ECOLOGICAL COSTS AND BENEFITS OF DEFENSES IN NECTAR. Ecology 86, 2968-2978.

Aitken SN, Yeaman S, Holliday JA, Wang T, Curtis-Mclane S (2008) Adaptation, migration or extirpation: climate change outcomes for tree populations. Evolutionary Applications 1, 95- 111.

Allen CD, Macalady AK, et al. (2010) A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest ecology and management 259, 660-684.

Altshuler DL (2003) ECOLOGY AND EVOLUTION: Enhanced: Darwin's Hummingbirds. 300, 588-589.

Bailey JK, Schweitzer JA, Rehill BJ, Lindroth RL, Martinsen GD, Whitham TG (2004) BEAVERS AS MOLECULAR GENETICISTS: A GENETIC BASIS TO THE FORAGING OF AN ECOSYSTEM ENGINEER. Ecology 85, 603-608.

Bailey JK, Wooley SC, Lindroth RL, Whitham TG (2005) Importance of species interactions to community heritability: a genetic basis to trophic-level interactions. Ecology letters 0, 051122062725008.

Ball DW (2007) The chemical composition of honey. Journal of chemical education 84, 1643.

Basden R (1965) The occurrence and composition of manna in Eucalyptus and Angophora. 90, 152-6.

Basden R The composition, occurrence and origin of lerp, the sugary secretion of Eurymela distincta (Signoret). In 'Proc. Linn. Soc. NSW', 1966, pp. 44-6

Basden R (1967) The occurrence and composition of the sugars in the honeydew of Eriococcus coriaceus (Mask.). Proc Linnean Soc New South Wales 92, 222-226.

Basden R A note on the composition of the lerp and honeydew of Eucalyptolyma maidenii Froggatt. In 'Proceedings of the Linnean Society of New South Wales', 1970, pp. 9-10

Baude M, Leloup J, Suchail S, Allard B, Benest D, Meriguet J, Nunan N, Dajoz I, Raynaud X (2011) Litter inputs and plant interactions affect nectar sugar content. Journal of Ecology 99, 828-837.

Becklin KM, Gamez G, Uelk B, Raguso RA, Galen C (2011) Soil fungal effects on floral signals, rewards, and aboveground interactions in an alpine pollination web. American Journal of Botany 98, 1299-1308.

Bennett A (2010) The role of soil community biodiversity in insect biodiversity. Insect Conservation and Diversity.

Bentley BL (1977) Extrafloral Nectaries and Protection by Pugnacious Bodyguards. Annual Review of Ecology and Systematics 8, 407-427.

Bertsch A (1983) Nectar production of Epilobium angustifolium L. at different air humidities; nectar sugar in individual flowers and the optimal foraging theory. Oecologia 59, 40-48.

Blanchard S, Lognay G, Verheggen F, Detrain C (2019) Today and tomorrow: impact of climate change on aphid biology and potential consequences on their mutualism with ants. Physiological Entomology 44, 77-86.

Bluthgen N, Verhaagh M, Goitia W, Jaffe K, Morawetz W, Barthlott W (2000) How plants shape the ant community in the Amazonian rainforest canopy: the key role of extrafloral nectaries and homopteran honeydew. Oecologia 125, 229-240.

Bond HW, Brown WL (1979) The exploitation of floral nectar in Eucalyptus incrassata by honeyeaters and honeybees. Oecologia 44, 105-111.

Boose DL (1997) Sources of variation in floral nectar production rate in Epilobium canum (Onagraceae): Implications for natural selection. Oecologia 110, 493-500.

Boucher DH, James S, Keeler KH (1982) The ecology of mutualism. Annual Review of Ecology and Systematics 13, 315-347.

Bradshaw AD (1965) Evolutionary significance of phenotypic plasticity in plants. In 'Advances in genetics. Vol. 13. pp. 115-155. (Elsevier).

Bradshaw F, Bradshaw S (2001) Maintenance nitrogen requirement of an obligate nectarivore, the honey possum, Tarsipes rostratus. Journal of Comparative physiology B 171, 59-67.

Campbell DR (1996) Evolution of floral traits in a hermaphroditic plant: Field measurements of heritabilities and genetic correlations. Evolution 50, 1442-1453.

Canto A, Herrera CM, Medrano M, Pérez R, García IM (2008) Pollinator foraging modifies nectar sugar composition in Helleborus foetidus (Ranunculaceae): an experimental test. American Journal of Botany 95, 315-320.

Carroll AB, Pallardy SG, Galen C (2001) Drought stress, plant water status, and floral trait expression in fireweed, Epilobium angustifolium (Onagraceae). American Journal of Botany 88, 438-446.

Carthew SM, Goldingay RL, Funnell DL (1999) Feeding behaviour of the yellow-bellied glider Petaurus australis at the western edge of its range. Wildlife Research 26, 199-208.

Case SB, Edworthy AB (2016) First report of ‘mining’ as a feeding behaviour among Australian manna-feeding birds. Ibis 158, 407-415.

Chambel MR, Climent J, Alía R, Valladares F (2005) Phenotypic plasticity: a useful framework for understanding adaptation in forest species. Forest Systems 14, 334-344.

Chen I-C, Hill JK, Ohlemüller R, Roy DB, Thomas CD (2011) Rapid range shifts of species associated with high levels of climate warming. Science 333, 1024-1026.

Close D, McArthur C, Paterson S, Fitzgerald H, Walsh A, Kincade T (2003) Photoinhibition: a link between effects of the environment on eucalypt leaf chemistry and herbivory. Ecology 84, 2952-2966.

Crozier L (1981) Beech honeydew: forest produce [New Zealand]. New Zealand Journal of Forestry (New Zealand) 1981.

Davidson EW, Segura BJ, Steele T, Hendrix DL (1994) Microorganisms influence the composition of honeydew produced by the silverleaf whitefly, Bemisia argentifolii. 40, 1069-1076.

Del-Claro K, Oliveira PS (1993) Ant-Homoptera interaction: do alternative sugar sources distract tending ants? Oikos, 202-206.

Detrain C, Verheggen FJ, Diez L, Wathelet B, Haubruge E (2010) Aphid–ant mutualism: how honeydew sugars influence the behaviour of ant scouts. Physiological Entomology 35, 168- 174.

Douglas A (1993) The nutritional quality of phloem sap utilized by natural aphid populations. Ecological Entomology 18, 31-38.

Dungan RJ, Kelly D (2003) Effect of host-tree and environmental variables on honeydew production by scale insects (Ultracoelostoma sp.) in a high elevation Nothofagus solandri forest. New Zealand journal of ecology, 169-177.

Ellison AM, Bank MS, et al. (2005) Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3, 479-486.

Erhardt A, Rusterholz H-P, Stöcklin J (2005) Elevated carbon dioxide increases nectar production in Epilobium angustifolium L. Oecologia 146, 311-317.

Falconer DS (1960) Introduction to quantitative genetics. Introduction to quantitative genetics.

Feeley KJ, Hurtado J, Saatchi S, Silman MR, Clark DB (2013) Compositional shifts in Costa Rican forests due to climate‐driven species migrations. Global change biology 19, 3472- 3480.

Fischer MK, Shingleton AW (2001) Host plant and ants influence the honeydew sugar composition of aphids. Functional Ecology 15, 544-550.

Fisher DB, Wright JP, Mittler TE (1984) Osmoregulation by the aphid Myzus persicae: a physiological role for honeydew oligosaccharides. Journal of Insect Physiology 30, 387-393.

Ford HA, Paton DC (1982) Partitioning of nectar sources in an Australian honeyeater community. Austral Ecology 7, 149-159.

Gange AC, Smith AK (2005) Arbuscular mycorrhizal fungi influence visitation rates of pollinating insects. Ecological Entomology 30, 600-606.

Gaze P, Cloud M (1983) Honeydew and its importance to birds in beech forests of south island, New Zealand. New Zealand journal of ecology 6, 33-37.

Gilbert N, Raworth D (1996) Insects and temperature—a general theory. The Canadian Entomologist 128, 1-13.

Guillaumet A (2017) Altitudinal migration and the future of an iconic Hawaiian honeycreeper in response to climate change and management. Ecological monographs 87, 410-428.

Himanen SJ, Nissinen A, DONG WX, NERG AM, Stewart Jr CN, Poppy GM, Holopainen JK (2008) Interactions of elevated carbon dioxide and temperature with aphid feeding on transgenic oilseed rape: Are Bacillus thuringiensis (Bt) plants more susceptible to nontarget herbivores in future climate? Global Change Biology 14, 1437-1454.

Hoffmann AA, Rymer PD, et al. (2019) Impacts of recent climate change on terrestrial flora and fauna: Some emerging Australian examples. Austral ecology 44, 3-27.

Holland GJ, Bennett AF, van der Ree R (2007) Time-budget and feeding behaviour of the squirrel glider (Petaurus norfolcensis) in remnant linear habitat. Wildlife Research 34, 288- 295.

Jakobsen H, Kritjansson K (1994) Influence of temperature and floret age on nectar secretion in Trifolium repens L. Annals of Botany 74, 327-334.

Kaczorowski RL, Juenger TE, Holtsford TP (2008) Heritability and correlation structure of nectar and floral morphology traits in Nicotiana alata. Evolution: International Journal of Organic Evolution 62, 1738-1750.

Kelly D (1990) Honeydew density in mixed Nothofagus forest, Westland, New Zealand. 28, 53-58.

Kelly D, Stirling DJ, Hunt GR, Newell CL, Jarvis CE (1992) Honeydew standing crop and production over 24 hours in Nothofagus solandri forest in Canterbury. New Zealand Journal of Ecology 16, 69-69.

Kiss A (1981) Melezitose, Aphids and Ants. Oikos 37, 382.

Klinkhamer PGL, van der Veen-van Wijk CAM (1999) Genetic variation in floral traits of Echium vulgare. Oikos 85, 515-522.

Leiss K, Klinkhamer P (2005) Genotype by environment interactions in the nectar production of Echium vulgare. Functional Ecology, 454-459.

Leiss KA, Vrieling K, Klinkhamer PGL (2004) Heritability of nectar production in Echium vulgare. Heredity 92, 446-451.

Litteral J, Shochat E (2017) The role of landscape-scale factors in shaping urban bird communities. In 'Ecology and conservation of birds in urban environments. pp. 135-159. (Springer).

Macior LW (1971) Co-evolution of plants and animals. Systematic insights from plant-insect interactions. Taxon, 17-28.

Mitchell RJ (2004) Heritability of nectar traits: Why do we know so little? Ecology 85, 1527- 1533.

Mitchell RJ, Shaw RG (1993) HERITABILITY OF FLORAL TRAITS FOR THE PERENNIAL WILD FLOWER PENSTEMON-CENTRANTHIFOLIUS (SCROPHULARIACEAE) - CLONES AND CROSSES. Heredity 71, 185-192.

Moller H, Tilley J (1989) Beech honeydew: seasonal variation and use by wasps, honey bees, and other insects. New Zealand journal of zoology 16, 289-302.

Murphy DJ, Kelly D (2003) Seasonal variation in the honeydew, invertebrate, fruit and nectar resource for bellbirds in a New Zealand mountain beech forest. New Zealand Journal of Ecology, 11-23.

Mutikainen P, Walls M, Ovaska J, Keinänen M, Julkunen-Tiitto R, Vapaavuori E (2000) Herbivore resistance in Betula pendula: effect of fertilization, defoliation, and plant genotype. Ecology 81, 49-65.

Ness JH, Morris WF, Bronstein JL (2009) For ant-protected plants, the best defense is a hungry offense. Ecology 90, 2823-2831.

Newman JA (2004) Climate change and cereal aphids: the relative effects of increasing CO2 and temperature on aphid population dynamics. Global change biology 10, 5-15.

Nicotra AB, Atkin OK, et al. (2010) Plant phenotypic plasticity in a changing climate. Trends in plant science 15, 684-692.

O'Reilly-Wapstra JM, McArthur C, Potts BM (2004) Linking plant genotype, plant defensive chemistry and mammal browsing in a Eucalyptus species. Functional Ecology 18, 677-684.

O’Reilly-Wapstra JM, Potts BM, McArthur C, Davies NW (2005) Effects of nutrient variability on the genetic-based resistance of Eucalyptus globulus to a mammalian herbivore and on plant defensive chemistry. Oecologia 142, 597-605.

Oliver DL (1998) The importance of insects and lerp in the diet of juvenile regent honeyeaters, Xanthomyza phrygia: Implications for the conservation of an endangered woodland bird. Wildlife Research 25, 409-417.

Oliver DL (2000) Foraging behaviour and resource selection of the Regent Honeyeater Xanthomyza phrygia in northern New South Wales. Emu 100, 12-30.

Oliver DL (2001) Activity budget of the regent honeyeater, Xanthomyza phrygia, in northern New South Wales. Australian Journal of Zoology 49, 695-712.

Olsson M, Shine R, Ba'K-Olsson E (2000) Lizards as a plant's ‘hired help’: letting pollinators in and seeds out. Biological Journal of the Linnean Society 71, 191-202.

Özder N, SAGLAM Ö (2013) The effects of temperature for development time, fecundity and reproduction on some ornamental aphid species. Journal of Central European Agriculture 14, 0-0.

Parachnowitsch AL, Manson JS, Sletvold N (2019) Evolutionary ecology of nectar. Annals of Botany 123, 247-261.

Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421, 37-42.

Paton DC (1980) The importance of manna, honeydew and lerp in the diets of honeyeaters. Emu 80, 213-226.

Paton DC (1982) The diet of the New Holland honeyeater, Phylidonyris novaehollandiae. Australian Journal of Ecology 7, 279-298.

Petanidou T, Smets E (1996) Does temperature stress induce nectar secretion in Mediterranean plants? New Phytologist 133, 513-518.

Pleasants JM (1983) Nectar production patterns in Ipomopsis aggregata (Polemoniaceae). American Journal of Botany 70, 1468-1475.

Pritchard J, Griffiths B, Hunt E (2007) Can the plant‐mediated impacts on aphids of elevated CO2 and drought be predicted? Global Change Biology 13, 1616-1629.

Prober SM, Potts BM, et al. (2016) Climate adaptation and ecological restoration in eucalypts. Proceedings of the Royal Society of Victoria 128, 40-53.

Rausher MD (2001) Co-evolution and plant resistance to natural enemies. Nature 411, 857.

Real LA, Rathcke BJ (1991) INDIVIDUAL VARIATION IN NECTAR PRODUCTION AND ITS EFFECT ON FITNESS IN KALMIA-LATIFOLIA. Ecology 72, 149-155.

Recher H, Holmes R, Schulz M, Shields J, Kavanagh R (1985) Foraging patterns of breeding birds in eucalypt forest and woodland of southeastern Australia. Australian Journal of Ecology 10, 399-419.

Recher HF (2016) The winter foraging behaviour of birds in a mixed eucalypt forest and woodland on the Southern Tablelands of New South Wales. Australian Zoologist 38, 1-15.

Rusterholz HP, Erhardt A (1998) Effects of elevated CO 2 on flowering phenology and nectar production of nectar plants important for butterflies of calcareous grasslands. Oecologia 113, 341-349.

Sagata K, Gibb H (2016) The Effect of Temperature Increases on an Ant-Hemiptera-Plant Interaction. PLos one 11, e0155131.

Schmid JM, Mata SA, Schmidt RA (1991) Bark temperature patterns in ponderosa pine stands and their possible effects on mountain pine beetle behavior. Journal of Forest Research 21, 1439-1446.

Serra‐Diaz JM, Franklin J, Dillon WW, Syphard AD, Davis FW, Meentemeyer RK (2016) California forests show early indications of both range shifts and local persistence under climate change. Global ecology and biogeography 25, 164-175.

Smith AP (1982) Diet and Feeding Strategies of the Marsupial Sugar Glider in Temperate Australia. The Journal of Animal Ecology 51, 149.

Southwick AK, Southwick E (1983) Aging effect on nectar production in two clones of Asclepias syriaca. Oecologia 56, 121-125.

Steinbauer MJ (1996) A note on manna feeding by ants (Hymenoptera: Formicidae). Journal of Natural History 30, 1185-1192.

Sun Y, Su J, Ge F (2010) Elevated CO2 reduces the response of Sitobion avenae (Homoptera: Aphididae) to alarm pheromone. Agriculture, ecosystems & environment 135, 140-147.

Suppiah R, Hennessy K, Whetton P, McInnes K, Macadam I, Bathols J, Ricketts J, Page C (2007) Australian climate change projections derived from simulations performed for the IPCC 4th Assessment Report. Australian Meteorological Magazine 56, 131-152.

Swart I (2017) Forty-spotted pardalote (Pardalotus quadragintus) tree use and the influence of tree and manna variation. University of Tasmania, Hobart

Temeles EJ (1996) A new dimension to hummingbird-flower relationships. Oecologia 105, 517-523.

Thomas CD (2010) Climate, climate change and range boundaries. Diversity and Distributions 16, 488-495.

Thomson JD, McKenna MA, Cruzan MB (1989) Temporal patterns of nectar and pollen production in Aralia hispida: implications for reproductive success. Ecology 70, 1061-1068.

Thuiller W, Lavergne S, Roquet C, Boulangeat I, Lafourcade B, Araujo MB (2011) Consequences of climate change on the tree of life in Europe. Nature 470, 531.

Thuiller W, Lavorel S, Araujo MB (2005) Niche properties and geographical extent as predictors of species sensitivity to climate change. Global Ecology and Biogeography 14, 347-357.

Tjallingii W (1995) Regulation of phloem sap feeding by aphids. In 'Regulatory mechanisms in insect feeding. pp. 190-209. (Springer).

Veits M, Khait I, et al. (2019) Flowers respond to pollinator sound within minutes by increasing nectar sugar concentration. Ecology letters 22, 1483-1492.

Via S, Lande R (1985) Genotype‐environment interaction and the evolution of phenotypic plasticity. Evolution 39, 505-522.

Virgós E (2002) Role of isolation and habitat quality in shaping species abundance: a test with badgers (Meles meles L.) in a gradient of forest fragmentation. Journal of Biogeography 28, 381-389.

Vogler DW, Peretz S, Stephenson AG (1999) Floral plasticity in an iteroparous plant: The interactive effects of genotype, environment, and ontogeny in Campanula rapunculoides (Campanulaceae). American Journal of Botany 86, 482-494.

Völkl W, Woodring J, Fischer M, Lorenz MW, Hoffmann KH (1999) Ant-aphid mutualisms: the impact of honeydew production and honeydew sugar composition on ant preferences. Oecologia 118, 483-491.

Vourc'H G, Vila B, Gillon D, Escarre J, Guibal F, Fritz H, Clausen TP, Martin J-L (2002) Disentangling the causes of damage variation by deer browsing on young Thuja plicata. Oikos 98, 271-283.

Wagner D, Kay A (2002) Do extrafloral nectaries distract ants from visiting flowers? An experimental test of an overlooked hypothesis. Evolutionary Ecology Research 4, 293-305.

Waser NM, Price MV (2016) Drought, pollen and nectar availability, and pollination success. Ecology 97, 1400-1409.

Whitham TG, Bailey JK, et al. (2006) A framework for community and ecosystem genetics: from genes to ecosystems. Nature Reviews Genetics 7, 510-523.

Wilkinson T, Ashford D, Pritchard J, Douglas A (1997) Honeydew sugars and osmoregulation in the pea aphid Acyrthosiphon pisum. Journal of Experimental Biology 200, 2137-2143.

Woinarski JCZ (1984) Small Birds, Lerp-Feeding and the Problem of Honeyeaters. Emu - Austral Ornithology 84, 137-141.

Woinarski JCZ, Cullen JM, Hull C, Nayudu R (1989) Lerp-feeding in birds: A smorgasbord experiment. Austral Ecology 14, 227-234.

Yang M-M, Raman A (2007) Diversity, richness, and patterns of radiation among gall- inducing psyllids (Hemiptera: Psylloidea) in the Orient and Eastern Palearctic. Oriental Insects 41, 55-65.

Yao I, Akimoto SI (2002) Flexibility in the composition and concentration of amino acids in honeydew of the drepanosiphid aphid Tuberculatus quercicola. Ecological Entomology 27, 745-752.

Zhang J, Nielsen SE, Stolar J, Chen Y, Thuiller W (2015) Gains and losses of plant species and phylogenetic diversity for a northern high‐latitude region. Diversity and Distributions 21, 1441-1454.

Zimmerman M (1983) Plant reproduction and optimal foraging: experimental nectar manipulations in Delphinium nelsonii. Oikos, 57-63.

Zimmerman M, Pyke GH (1988) Experimental manipulations of Polemonium foliosissimum: effects on subsequent nectar production, seed production and growth. The Journal of Ecology, 777-789.

Zu P, Schiestl FP (2017) The effects of becoming taller: direct and pleiotropic effects of artificial selection on plant height in Brassica rapa. The Plant Journal 89, 1009-1019.