Agathis vs. Hymenaea – trapping biases to interpret arthropod assemblages in ambers

doi:10.21203/rs.3.rs-6999659/v1

Download PDF

Research Article

Agathis vs. Hymenaea – trapping biases to interpret arthropod assemblages in ambers

https://doi.org/10.21203/rs.3.rs-6999659/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 06 Nov, 2025

Read the published version in BMC Biology →

You are reading this latest preprint version

Background The genera Agathis(Coniferales: Araucariaceae) and Hymenaea(Fabales: Fabaceae) contain resin-producing tree species that are crucial for actuotaphonomic studies. While certain Cretaceous ambers likely originated from Agathis or Agathis-like trees, Hymenaeais the primary source of many Miocene ambers. Field studies were conducted in New Caledonia and Madagascar, to collect Defaunation resin (resin produced after 1760 AD (Anno Domini)). Arthropods were collected with yellow sticky and Malaise traps in New Caledonia, Madagascar and Mexico. Cretaceous and Miocene ambers, copals (2.58 Ma to 1760 AD), and Defaunation resins from various regions were analysed to compare arthropod trapping patterns.

Results Actuotaphonomic results show lower number of arthropods trapped in Agathis Defaunation resin, with a non-uniform distribution, compared to the abundant and uniformly distributed arthropods trapped in HymenaeaDefaunation resin. The lower number of arthropod inclusions in the trunk resin of the Agathistrees is attributed to the rapid polymerisation of that resin. Under the same experimental conditions, the arthropods in Agathis Defaunation resin plot far from the arthropods collected in the yellow sticky and Malaise traps, while the arthropods in Hymenaea Defaunation resin plot close to the arthropods in the yellow sticky traps.

Conclusions These findings confirm different resin trapping patterns between Agathis and Hymenaea, with significant implications for interpreting the amber record. The fauna trapped by Hymenaea resin closely resembles the arthropod biocoenosis that live in and around the trunks, indicating autochthony and close relationship with the forest ecosystem, unlike Agathisresin. These results improve our understanding of arthropod trapping biases in resin and lead us to reconsider previously proposed interpretations of Cretaceous forest biocoenoses.

amber

actuotaphonomic studies

Cretaceous

Miocene

biocoenosis

taphonomy

copal

Defaunation resin

Amber is a fossil plant resin, that sometimes preserves inorganic inclusions such as water or soil particles, and a high diversity of bioinclusions, such as phloem sap, plant remains, microorganisms, invertebrates (with arthropods being the most abundant) and vertebrate remains, providing a unique window into biocoenoses in ancient forested ecosystems [1–4]. In the Cretaceous period, conifers were globally the primary contributors to sedimentary deposits of resin (today amber) [5]. Several families have been proposed as the producers of large quantities of resin at that time, namely Araucariaceae, Cupressaceae, Pinaceae, Podocarpaceae and the extinct Cheirolepidiaceae [6–9]. Moving to the Eocene epoch, amber deposits were derived from conifers and angiosperms. For example, Eocene Baltic amber originated from a conifer tree of the family Pinaceae and/or Cupressales: Sciadopityaceae [10, 11], while Cambay (India) amber was linked to the angiospermous Dipterocarpaceae [12]. Conversely, the majority of Miocene ambers, copals, and Defaunation resins (sensu Solórzano-Kraemer et al. [13]) were derived from angiosperms, predominantly of the families Fabaceae and Dipterocarpaceae [13, 14].

The Cretaceous, Paleogene and Neogene periods are the most prolific in terms of described amber deposits [5, 13]. However, a difference has been noted in unbiased collections between the amber that originated from conifers and that from angiosperms. Of the approximately 500 Barremian amber localities known in Lebanon, only 38 contain arthropod inclusions [15, 16]. Similarly, of the more than 140 Albian amber localities in Spain, only 11 contain arthropod inclusions [17, 18]; of these, four are poor and contain only one or two arthropod inclusions. Another example is the Albian―Cenomanian French amber, with 60 reported localities, of which 14 contain arthropod inclusions, but only one, Archingeay-Les Nouillers, was reported as highly fossiliferous [19–23]. Cretaceous amber from Southern America seems to exhibit a similar circumstance, as arthropod inclusions have remained elusive despite many years of intensive searching [24–26]. Only 21 terrestrial arthropods in Ecuadorian Albian amber are known until now from South America [27]. It is important to consider that many Cretaceous amber deposits mentioned in the literature contain only an anecdotal amber record. Deposits with significant amber quantities are scarce [5], and researchers primarily focus on amber-rich deposits when searching for arthropod inclusions.

On the contrary, Cenozoic amber localities (from the Paleogene and Neogene) appear less abundant than in the Cretaceous, but nearly all contain arthropod inclusions [13,28]. However, the inclusion density, i.e. the number of specimens per volume or mass of inspected amber, has never been studied before. These observations raise question on whether the difference in arthropod inclusions is due to a different distribution of fauna in the ecosystem or a different trapping bias of the resins.

Resin has been defined as an entomological trap that works similarly to commercial yellow sticky traps [29, 30]. The fauna living in and around the resin-producing trees of Hymenaea Linnaeus, 1753 has been described as having a greater chance of being trapped and preserved in the resin (the so-called thanatocoenosis or death assemblage), than the fauna living in surrounding areas of the forest [30]. Some resins often contain inclusions that are closely related to the habitats provided by resiniferous trees. Diptera is an order that contains a high number of inclusions in amber. Most of these insects are fungivores, carnivorous, necrophagous, or hematophagous, which explains why they are commonly found as amber inclusions [30]. Agathis Salisbury, 1807 (Araucariaceae) and Hymenaea (Fabaceae) genera contain resin-producing tree species serving as pivotal models for actuotaphonomic investigations in this respect [29–32]. The genus Agathis is primarily distributed today in the Southern Hemisphere, growing in tropical environments across the Malay Archipelago (Malesia), the islands of the South-West Pacific (Melanesia), northern New Zealand, and South Queensland in Australia [33]. Agathis or Agathis-like trees have been postulated to be the source of certain Cretaceous and Cenozoic ambers from Lebanon (Barremian), Spain (Albian), France (Albian-Cenomanian), Myanmar (Cenomanian), Australia (Turonian–Campanian), and New Zealand (Oligocene and Miocene), while there is certainty that Agathis was or is involved in the origin of the copals and Defaunation resins from New Zealand and New Caledonia [3, 13, 34, 35]. Hymenaea is a tropical and subtropical resin-producing tree originating in East Africa during the early Miocene, mainly distributed throughout central and South America, and East Africa (including Kenya, Tanzania, Mozambique, Comoros, Madagascar, Mauritius and Reunion Island) [36–39]. The genus Hymenaea has been identified as the source of the ambers from Ethiopia, Mexico, Dominican Republic (all Miocene in age), Pleistocene and Holocene copals, and Defaunation resins from East Africa, and Central and South America [13, 39, 40].

Understanding why many amber deposits are poor or devoid in arthropod inclusions, and why this holds true for Cretaceous deposits more than for Paleogene and Neogene ones, would contribute to a better interpretation of oryctocoenoses (fossil assemblages) based on amber records. We have thus carried out a comparative actuotaphonomic study based on Agathis trees from New Caledonia and Hymenaea trees from Madagascar. We followed the same experimental protocol as in a previous study based on Hymenaea from Madagascar [30] in order to compare the results. Our aim was to quantify, using statistical analysis, the differences in the intensity of arthropod trapping between the resin exuded by two different resin-producing trees ("gymnosperms" exclusively conifers vs. angiosperms) in their natural environments, with a view to applying the conclusions to the reconstruction of deep-time environments based on fossil assemblages preserved in amber.

Trapping patterns of arthropods in ambers, copals and Defaunation resins from Agathis, Agathis-like, and Hymenaea trees

Fig. 1 can be placed here

In the Defaunation resin obtained from Agathis in New Caledonia (Additional file 1: Fig. S1), a total of 48 arthropod inclusions were found in 2,618.7 grams of resin (Table 1), mostly Acari, Pseudoscorpiones, Collembola and non-flying Hymenoptera, namely Formicidae, but also a Thysanoptera and a Diptera remain. This scarce number of arthropods in Agathis Defaunation resin was unexpected, particularly considering the significant number of arthropod inclusions (~4,000) found in 2,300 grams of Malagasy Defaunation resin from Hymenaea (Table 1, see Additional file 2 for a detailed list of arthropod inclusions). We observed a visible divergence in arthropod trapping efficiency (Fig. 1). Here, efficiency refers to two things: First, it refers to the increased trapping of arthropods and other organisms by both ancient and modern resins. Second, it refers to the recording of a more complete (varied) assemblage that better reflects the communities that inhabited these ancient and modern forests. We also observed a visible divergence in arthropod trapping patterns, represented in Fig. 2, between Agathis and Agathis-like, and Hymenaea samples (including ambers and copal from both types of trees). When comparing arthropod counts with resin mass, two different trends emerge: With the increase in resin mass, the pattern of Agathis and Agathis-like samples show a slow, near-flat increase. Meanwhile, Hymenaea samples show a sharp increase, indicating markedly higher trapping efficiency.

Fig. 2 can be placed here

The different arthropod trapping patterns observed in the Hymenaea and Agathis Defaunation resins are revealed not only by the number of arthropod inclusions per mass of resin (Fig. 2), but also by the probability that a collected Defaunation resin piece/lump with a certain mass does not have any inclusions. The thanatocoenosis (death assemblages) varies greatly between that of Agathis Defaunation resin and that of Hymenaea Defaunation resin, which will considerably influence the interpretation of the resulting oryctocoenosis. The probability density function (PDF) of resin pieces lacking arthropod inclusions was analysed in relation to mass for Hymenaea (Madagascar) and Agathis (New Caledonia) Defaunation resins (Fig. 3). The distributions for both types of resin followed a power-law trend, with exponents of b = -1.97 and b = -3.34 for Agathis and Hymenaea, respectively. The exponent b is a measure of distribution uniformity: values closer to zero indicate greater variability, while more negative values suggest a system approaching uniformity (an explanation of PDF and the degree of uniformity is presented in the Additional file 1). For Hymenaea, the probability decreases rapidly as the mass increases; for Agathis, however, the decrease is gradual. This reflects the fact that large Agathis pieces lack arthropod inclusions. The rate at which the probability decreases reflects the distribution of arthropod inclusions. A uniform distribution leads to an exponential decrease corresponding to an infinite exponent b, whereas a distribution in which all inclusions are concentrated in a single location would have an exponent of b = 0. Thus, the value of b serves as an indicator of the uniformity of the distribution of arthropod inclusions in amber. If the only difference were a lower arthropod density around Agathis and both resins trapped them equally, the lines would have the same shape but offset from each other and they would appear parallel. Therefore, the different values of b for each collection of pieces of Defaunation resin show that Hymenaea resin traps arthropods more efficiently and more uniformly than Agathis resin (see also Additional file 1: Fig. S1 and extended material and methods in Additional file 1). In other words, the characteristics of the thanatocoenosis are very different for each type of resin under investigation here.

Fig. 3 can be placed here

Arthropod trapping patterns including yellow sticky and Malaise traps

The significant difference in arthropod trapping patterns observed in the Defaunation resins raises questions about the trapping efficiency of other entomological traps we used, which were placed: i) on the trunk and ii) close to the same tree species. These include yellow sticky traps, which resemble the sticky resin, and Malaise traps – both of which are used for arthropod collection and actuotaphonomic studies [29, 30, 41, 42]. Therefore, to better understand the trapping patterns and to assess whether Agathis tree resin traps arthropods in a similar way to yellow sticky traps, we analysed their arthropod assemblages using multidimensional scaling (MDS) with Bhattacharyya distance, in order to visualise the relationships between the taxa based on their frequency data. An important result is that the arthropods (at the order level) in the Agathis Defaunation resin, collected from the tree trunks in New Caledonia, plots far from the arthropods (at the order level) collected in both the yellow sticky and Malaise traps placed on the trunk and close to the Agathis trees, respectively (Fig. 4A) (in the Additional file 2 we present the arthropod list). The arthropods (at order level) trapped in the yellow sticky traps in New Caledonia plotted far away from the arthropods trapped in the Malaise traps, but also far from the yellow sticky and Malaise traps from other sampling places, such as the Hymenaea forests in Madagascar and Mexico (Fig. 4C), where the Hymenaea trees were also sampled [29, 30]. In contrast, the samples from Hymenaea trees in Madagascar, analysed using the same method, plots the arthropods (at the order level) in the Hymenaea Defaunation resin from the tree trunks close to the arthropods collected in the yellow sticky traps (Fig. 4B).

Fig. 4 can be placed here

Families in Diptera deserve a special analysis below the order level, attending to their high abundance in the collected material and in general in amber [43]. In the yellow sticky traps placed on the trunkof Agathis, Diptera made up more than 60% of the total. Within the Diptera, the family Phoridae comprised nearly 90% (Additional file 1: Fig. S2). At family level within Diptera, the yellow sticky traps placed on the trunkof Agathis from New Caledonia plot far away from all other samplings whether they were related to Agathis or Hymenaea (Fig. 4D).

The assemblage of trapped arthropods includes Acari, Collembola, Hemiptera, Thysanoptera, Hymenoptera, and Coleoptera, among other taxa (see Additional file 2 for arthropod list). The differences between flying and non-flying insects present in amber, copal and Defaunation resin are not significant among the studied samples, except for the arthropods from the Malagasy Defaunation resin of Hymenaea and those from the Australian Eocene amber (Additional file 1: Fig. S3).

We also categorised the arthropods found in all the ambers, copals, and Defaunation resins originating from Agathis or Agathis-like trees and Hymenaea trees, and the arthropods trapped by yellow sticky and Malaise traps in the Agathis and Hymenaea forests (see Additional file 1 and 2 for details of the material and the lists of arthropods, respectively) using a hierarchical clustering (Fig. 5). Agathis and Agathis-like resins (Defaunation resins and ambers, respectively) cluster together at the order level and at family level in the case of Diptera (due to the high prevalence of the order Diptera in most samples, their families were analysed separately). Conversely, for New Caledonia, arthropods from both yellow sticky and Malaise traps placed on and close to the Agathis tree trunks, respectively, cluster together with those from amber, copal and Defaunation resin derived from Hymenaea, as well as with the arthropods from both yellow sticky and Malaise traps on and close to the Hymenaea tree trunks.

As an additional result, we have updated the list of unbiased amber inclusions from various localities that can be used in future investigations (Additional file 2).

Fig. 5 can be placed here

The number of amber localities lacking arthropod inclusions is unknown for most amber-bearing regions. This is due to the fact that palaeontological research focuses primarily on systematics and taxonomy, and on palaeoautecological information provided by the bioinclusions, in an attempt to reconstruct forest biocoenoses in deep time, while other aspects of the amber and amber deposits, such as taphonomy and geology, remain unstudied or poorly studied.

It is also important to mention that a large number of arthropod inclusions were yielded from some amber deposits discovered several years ago and worked for a long time. Peñacerrada I, for example, is notable for yielding the greatest number of arthropod inclusions from Spanish amber [18, 44]. Notably, it is also the locality from which the most amber has been extracted and prepared (see Table 1). The same is true for the Albian–Cenomanian amber locality from south-west France at Archingeay-Les Nouillers. Recently discovered deposits such as the Eocene amber from Australia or the Miocene amber from New Zealand, produced by Agathis, currently contain only a few arthropod inclusions [35, 45]. Thus, for some deposits, the absence of arthropod inclusions is most likely due to collection bias. However, it is clear that the majority of the Cretaceous amber deposits does not contain arthropod inclusions or are extremely poor [5]. Some authors have suggested that this circumstance for many Cretaceous amber deposits may be explained by a much more abundant resin exudation under confined conditions in the underground than in aerial conditions [46]. Araucariaceae trees, particularly Agathis spp., are known for their comparatively abundant production and exudation of resin, not only on their tree trunks, but also within their root systems [3, 37, 47, 48, 49]. In contrast to the hundreds of arthropod bioinclusions described from Malagasy copal and Defaunation resin [39], the copals found in regions like New Zealand and New Caledonia (own field. obs.) lack such abundant inclusions, and no arthropod bioinclusions have been described from this material. This can be partly explained by the fact that root resin is unlikely to trap arthropods in confined conditions; given the abundant production of root resin and the resulting mixture of aerial and root copal in these geological deposits, a large proportion of the collected pieces are of the root type. When examining the extant roots of Hymenaea trees and their resin production, a distinct contrast is evident (Additional file 1: Fig. S4). The resin in Hymenaea roots appear intricately interspersed in the sand, forming a crust around the roots ([39] fig. 8). In Madagascar, we observed large Hymenaea trees that had been uprooted by hurricanes, showing their root systems with only small amounts of resin, unlike large resin lumps commonly formed by Agathis roots [49]. In New Zealand, we have also had the opportunity to study the root systems of large Pleistocene Agathis at Waipapakauri [49] and Bayleys Beach localities. These large trees are exhumed for their trunk wood, but the roots are abandoned, allowing us to observe large amounts of copal associated with roots (Additional file 1: Fig. S4C, D and E).

To explain the exudation of resin on the tree trunk and the capability to trap arthropods, it is important to note that resin exudation, in general, archives a variety of possible defensive actions, for example, to shield trees from threats such as bark beetles and herbivores, as well as to facilitate wound healing, preventing fungal and bacterial infection, or to prevent desiccation [2, 5, 48, 50, 51]. The variation in resin composition plays a prominent role in those defensive actions. In performing those defensive actions, Hymenaea resin has been also demonstrated to accidentally act as a kind of entomological trap, more precisely as yellow sticky traps that works better for some arthropod groups than for others [29, 30]. Consequently, scientific interest in amber, copal and Defaunation resin has been primarily focused on arthropod inclusions [13]. However, the absence or fewer number of arthropod inclusions in many amber deposits, especially those of Agathis or Agathis-like origin, has never been questioned.

The very few arthropod inclusions in Agathis Defaunation resin compared to the large number of arthropod inclusions in Hymenaea Defaunation resin, along with the different values of exponent b (measure of distribution uniformity) for each resin assemblage, shows that Hymenaea resin is more efficient at trapping arthropods than Agathis resin and does so spatially more uniformly (see also Additional file 1: Fig. S1 and extended material and methods in Additional file 1). This is consistent with our observations in amber. All arthropod inclusions recorded so far in amber originated from Agathis or Agathis-like trees were found in a few pieces/lumps (non-uniform), whereas arthropod inclusions in amber originated from Hymenaea trees are more uniform (Additional file 1: Fig. S1), which means that there are more pieces with at least one inclusion.

The fewer arthropod inclusions and their non-uniform distribution in Agathis and Agathis-like resins compared with those observed in Hymenaea resin can be explained mainly through two different aspects, namely 1) resin composition and 2) arthropod attraction or repulsion, hereafter discussed more in detail.

Resin composition

The variation in resin chemical composition between plant species affects the physical properties of the different resins, namely viscosity or stickiness, and consequently the capability to trap arthropods [28, 34, 37, 52].

Resin is a complex mixture of volatile compounds, including mono- and sesquiterpenoids, diterpenoids, and sometimes triterpenoids. The diterpenoids originate mainly from conifers (gymnosperms), while triterpenoids (e.g., of the oleanane, ursane and lupane series) and sesquisterpenoids come from angiosperms [53]. These compounds contribute to the fluidity (sesquisterpenoids) of the resin, and determine also its grade of viscosity (diterpenoids) [34]. Agathis and Hymenaea resins differ in their chemical composition. Agathis produces a resin type primarily composed of terpenoids [54]. In contrast, Hymenaea produces a resin containing both terpenoids and gum components [37]. The hardening and polymerisation of resin hinge on the number of free radicals within non-volatile compounds, particularly labdatriene diterpenoids, abundant in Agathis species [34, 55, 56].

The effectiveness of resin as a defence mechanism under biotic and abiotic environmental stress depends on factors such as drying, flow rate, and viscosity. These factors determine, for example, whether arthropods are pushed out or trapped on the tree trunk, whether a microbial infection can be stopped, or whether desiccation can be prevented. In general, our observations show distinct drying and viscosity characteristics for Agathis and Hymenaea resins. This suggests that the faster polymerisation of Agathis resin, resulting in rapid drying (see Fig. 6C), likely leads to fewer arthropods being trapped. Conversely, the characteristics of Hymenaea resin imply an abundant trapping of biological remains (Fig. 6D–I). According to our observations, both resins are fluid at the time of exudation, Agathis even more than Hymenaea. However, this changes quickly as Agathis resin dries faster than Hymenaea does. Whether an arthropod gets stuck depends on it passing by just as the tree begins to produce resin and the organism walking or flying close enough to be trapped. It is quite likely that the fast-drying nature of Agathis resin prevents arthropods from sticking to it (Figs. 1 and 6).

The previous idea aligns with the sensitivity of Agathis spp. trees to the Phytophthora spp., a water- and soil-borne oomycete primarily affecting these tree species. This sensitivity prompts a substantial production and exudation of resin as a defence mechanism [3, 5, 57]. In consequence, the trees, both their trunks and roots, are producing large amount of resin. It has been hypothesized that rapid resin hardening may prevent the rapid spread of pathogens on the trunk of a tree [37, 58, 59]. However, Agathis is also attacked by the defoliating coccids, trips and boring beetles, among others [60]. Therefore, resin exudation in Agathis is not always induced by a pathogenic agent.

On the other hand, in most angiosperms, including Hymenaea, the resin is mixed with gum (a mixture of hydrophilic polysaccharides), which increases the water-holding capacity of the tissues and prevents desiccation [37, 48, 54]. Gum is rarely observed in conifers and not found in Agathis spp. [32]. The hydrophobic nature of gum increases the stickiness of the resin, as gum can form bonds and stick when it comes in contact with oily surfaces [61]. This may partly explain why the resins produced by some angiosperms remain sticky for a long time.

Fig. 6 can be placed here

How quickly the resin dries depends mainly on the polymerisation rate (low polymerisation rate implies a liquid and sticky resin for a long time), and how long the resin can trap arthropods depends on how long it takes to dry. Therefore, regardless of the amount of resin exudation, the resin will trap arthropods on the entire surface of the accumulated resin for a longer time, making the distribution of arthropod inclusions more uniform (Additional file 1: Fig. S1B), as is the case of Hymenaea resin (Fig. 6D–I). A resin with a high polymerisation rate will remain fluid and sticky for a shorter period of time, giving arthropods a shorter period to be trapped in the resin, resulting in a non-uniform distribution (Additional file 1: Fig. S1C) of arthropod inclusions.

Arthropod attraction or repulsion to resin

Attraction or repulsion toward resins may be physical or chemical in nature and are well documented for some arthropods [40]. From a physical point of view, the “water-imitating” reﬂection polarisation of resin has been proved for the presence of aquatic adult insects in amber [62]. Although more research is needed to understand how the light reflection of the resin may attract or repel other arthropods, an opaque and white surface (Figs. 1 and 6A, B, and C) may reflect less light than the transparent glue-like resin (Figs. 1 and 6D, E, and F), making Agathis resin a candidate for being less attractive.

From a chemical point of view, it has been well documented in some resin-producing conifers that (—)- pinene, a constituent of stem oleoresin, increases in response to heightened insect activity, suggesting a defence mechanism [51]. Abundant — (α)- pinene has also been reported in araucariacean resins [63]. Conifer oleoresin is a complex compound comprising various monoterpenes, sesquiterpenes, and diterpene resin acids. The turpentine portion of the oleoresin includes over 30 monoterpenes and many sesquiterpenes, serving as a defence mechanism by being toxic to some pathogens and insects. Additionally, it aids in sealing plant wounds through the hardening action of diterpenes [51]. Also, the labdanes have been a subject of research due to their potential use as natural insecticides, showcasing antifeedant properties against some Coleoptera, Diptera, and Lepidoptera [64]. Labdanes are diterpenes featuring two aromatic rings in their structure, and are notably abundant in resins derived from the Araucariaceae [65, 66].

Volatile terpenoids, including labdanes, play then a dual role in defence by directly deterring herbivores and indirectly attracting their natural predators [67]. Moreover, these compounds also contribute to attracting pollinators for some gymnosperms such as cycads [68]. In the case of Hymenaea, both in Africa and South America, different resin compounds such as caryophyllene or α-Humulene are present to defend against various caterpillars and termites [40]. The resin of some angiosperm species also attracts insect pollinators or animals (birds and mammals) that can disperse their fruits [48]. The attractiveness or repulsiveness of the resin is an important taphonomic bias in the trapping of some groups of arthropods. However, determining whether Hymenaea can attract arthropods in more abundance than Agathis, or primarily some particular arthropod groups, requires experimental investigation.

The role of yellow sticky and Malaise traps in studying Agathis resin

Different orders of arthropods typically exhibit varying proportions in assemblages within amber, copal and Defaunation resins, with Diptera, Hymenoptera, and Coleoptera being the most abundant orders. The type of organism present and its abundance are contingent upon several taphonomic and ecological variables [1,3, 40], and these determined which part of the resiniferous forest is represented in the amber record [30]. In this context, we address two primary aspects of arthropod trapping: (1) whether Agathis resin traps arthropods in a similar way to yellow sticky traps, as observed with Hymenaea resin [30], and (2) whether the arthropod assemblage preserved in Agathis Defaunation resin is representative of the fauna in the same forest.

From our actuotaphonomic studies in Madagascar, we know that the arthropod assemblage in the Hymenaea Defaunation resin is comparable to that in the yellow sticky traps placed on Hymenaea trunks. We also know that the assemblage differs notably from the arthropod fauna trapped in Malaise traps placed close to the trunks. This means that the arthropod assemblages trapped by Hymenaea resin represent mainly the fauna living in and around the trunk [30]. This pattern was reinforced by our second sampling in Sacaramy, Madagascar ([39] fig. 1), and shown in Fig. 4B, in which MDS also plots the arthropod assemblages in Defaunation resin from Hymenaea close to the arthropod assemblages in yellow sticky traps. On the contrary, MDS plots arthropod assemblages (at order level) in resin from Agathis far away from the arthropod assemblages in yellow sticky and Malaise traps in Agathis (Fig. 4). This stark contrast shows that Agathis resins exhibit a different trapping bias as an entomological trap like the yellow sticky traps.

We found that arthropods preserved in amber and in Defaunation resin from Agathis and Agathis-like trees form distinct clusters at both the order and family levels (Fig. 5) (the latter focussed on dipteran families herein). In contrast, arthropods from yellow sticky and Malaise traps that were placed on and near Agathis trees cluster with those from amber, copal and Defaunation resin of Hymenaea origin. They also cluster with the arthropods from yellow sticky and Malaise traps on and near Hymenaea trees. This reinforces the idea that the resin of Hymenaea acts as an entomological trap (i.e., it has a trapping effect similar to yellow sticky traps), and that there is difference in the ways Agathis and Hymenaea resins trap arthropods.

The arthropods trapped in Agathis Defaunation resin are mostly Arachnida or non-flying Hexapoda, except for one Thysanoptera and one Diptera remain. In contrast, the samples of Defaunation resin collected from Hymenaea tree trunks contain abundant Diptera (Additional file 2). However, as presented in the results, there was no a representative difference between flying and non-flying insects found in amber, copal, and Defaunation resin across the studied samples. Therefore, it is likely that this finding is due to the scarcity of material and needs more investigation.

The number of Diptera specimens was very high in the yellow sticky traps placed on Agathis, this may be a reason why they plot close to the Malaise traps (Fig. 4C and Additional file 1: Fig. S5C), since Malaise traps are considered a successful method to collect flies [69]. Within the Diptera, the family Phoridae, overwhelmingly dominated the yellow sticky traps placed on Agathis, comprising nearly 90% of dipterans (Additional file 1: Fig. S2). As our collection took place in November and December, the humid months in New Caledonia, seasonality probably played an important role in determining abundance. Phoridae typically exhibit higher numbers during humid periods [70]. While other dipterans, namely Chloropidae, Sciaridae, and Cecidomyiidae, were also abundant in the yellow sticky traps placed at 1m height, their numbers are small in comparison to Phoridae. This may be a reason why they plot separately from Malaise traps (Fig. 4D and Additional file 1: Fig. S5D). Although not as abundant as in yellow sticky traps on Agathis, the family Phoridae was also abundant in yellow sticky traps placed on Hymenaea, and this family is also abundant in amber, particularly in Miocene ambers (e.g., [71]). Phoridae flies can be collected using a variety of traps [29, 72, 73]; however, yellow sticky traps seem to be the most effective one for collecting these flies [74, 75]. Small vertebrates, such as lizards, can also become trapped in yellow sticky traps, drawing in phorid flies that are attracted to decaying animal matter [41, 74, 76]. This phenomenon may also partly account for the high abundance of dipterans, particularly phorid flies, observed in New Caledonia in yellow sticky traps. Surprisingly, Phoridae is not present in Agathis Defaunation resin, and Diptera is notably underrepresented in that resin (Additional file 2). Resin production is affected by various factors, including temperature, humidity, growth, carbon assimilation, soil nutrients, or injuries [3, 31]. Therefore, it is not possible to establish a correlation between seasonality and the absence of phorid flies in the resin.

Given that Agathis and Hymenaea resins trap arthropods differently, a pattern inferred for extant and deep-time ecosystems, it is crucial to consider these different trapping biases (Fig. 1) when comparing fossil assemblages preserved in ambers from these two botanical sources.

Several important factors influence the trapping of arthropods by Agathis resin. We consider that the most important factor is rapid drying, which results in a higher viscosity and the acquisition of a whitish patina on resin exuded from the trunk—even though Agathis resin is much more fluid than Hymenaea resin when it is originally exuded. Rapid drying prevents a large number of arthropods from being present as bioinclusions in the Agathis Defaunation resin, and it results in a statistically non-uniform spatial distribution of these. In contrast, Hymenaea resin has a higher viscosity when exuded, but it takes much longer to dry, and remains transparent; the slower drying allows arthropods to be abundant as bioinclusions in the resin, making their spatial distribution statistically more uniform. This leads us to conclude that Agathis resin does not act as an effective entomological trap for arthropods, unlike Hymenaea resin (Fig. 1). This suggests that the arthropods trapped in Agathis Defaunation resin (thanatocoenosis) (which eventually become amber - oryctocoenosis) do not reflect the fauna that lived in or surrounding the resin-producing trees (biocoenosis), unlike Hymenaea resin. This taphonomic pattern has critical implications for the interpretation of Cretaceous forest biocoenoses with an Agathis or Agathis-like origin. Although we discuss that Agathis resin is less effective at trapping arthropods than Hymenaea resin, we could not determine frequent ecological groups (arthropod guilds) trapped by Agathis resin using the small sample set that is available. Similarly, the fossil assemblages of arthropod bioinclusions in Eocene or Miocene Agathis ambers do not provide guild information due to the scarcity of deposits. A follow-up investigation (beyond collecting more Agathis resin with bioinclusions) could involve defining the criteria required to recognise guilds and categorising the inclusions found in Cretaceous amber, in order to determine the most prevalent types of arthropods in this amber type.

The abundance of Cretaceous amber and Quaternary copal localities featuring numerous lumps but few bioinclusions, along with the non-uniform distribution of rare arthropod inclusions, can be partly attributed to substantial resin exudation from Agathis root systems. This exudation forms both large and small lumps in confined conditions. Comparable profuse resin production has not been observed from the root system of Hymenaea trees. Several unanswered questions still require further investigation, such as how resin attracts or repels some arthropods influencing the number of arthropods trapped, and the content of bioinclusions in the oryctocoenosis. Furthermore, this influence needs to be considered with respect to whether the trees involved were from the genera Agathis or Hymenaea, or from some close relatives.

Further studies and more accurate taphonomic data are needed to identify the conditions in which arthropods were trapped in other ancient and modern resins from coniferous and leguminous trees, and to determine and how the oryctocoenosis originated.

Material

We collected living arthropods using yellow sticky and Malaise traps, following the methodology outlined by Solórzano-Kraemer et al. [30]. These traps were installed on and close to Agathis lanceolata (Lindl. ex Sebert & Pancher) Warb., 1900 trees in Bon Secours forest, adjacent to the Rivière Bleue Provincial Park (New Caledonia), in 2016 (November – December, warm and rainy season), and on and close to Hymenaea verrucosa Gaertner, 1791 in Madagascar, Mananjary region (between Nosy Varika and Ambahy), in 2013 (September- October, warm and rather dry season), and Sacaramy (near to Diego Suarez), in 2015 (April – May, warm and rainy season) (Additional file 1: Fig. S6 and extended material and methods in Additional file 1). In total, we collected arthropods around four Agathis trees and eight Hymenaea trees (four Hymenaea trees per campaign). During each collection trip, we displayed 45 yellow sticky traps at three different heights on each tree for eight days, as well as four Malaise traps, one for each of four selected trees. We added to our analyses the data extracted from Solórzano-Kraemer et al. [29], who also collected arthropods with yellow sticky and Malaise traps in 2010, 2011 and 2012 on and close to Hymenaea courbaril Linné, 1753 in La Rinconada National Park in Mexico. The list of sampled arthropods can be seen in the Additional file 2.

Throughout the manuscript we use the terms amber, copal, and Defaunation resin sensu Solórzano-Kraemer et al. [13]. In this system, amber is older than 2.58 Ma, copal is 2.58 Ma to 1760 AD (Anno Domini), and Defaunation resin is the resin produced after 1760 AD.

We sampled Defaunation resins from eight Agathis lanceolata trees, and from 28 A. ovata (C. Moore ex Veill.) Warb., 1900 trees in New Caledonia. In Madagascar, resin was collected from 11 H. verrucosa trees in the Mananjary region in 2013 [30] and from 10 H. verrucosa trees in Sacaramy (for the precise location and map see [39], fig. 1) in 2015. All Defaunation resin samples were collected from tree trunks. From Mexico, no Defaunation resin was sampled from Hymenaea courbaril trees because resin exudates were virtually absent [29].

For the purposes of this paper, we do not distinguish between the different extant species of Agathis and Hymenaea involved in the collection work, or between the different ancient tree species proposed as the origin of the resin in the past (in the case of amber derived from Hymenaea, e.g. H. protera† Poinar, 1999, H. mexicana† Poinar and Brown, 2002, and H. allendis† Calvillo-Canadell et al., 2010). We use "Agathis" to refer to both living trees and fossil resins derived from Agathis trees. "Agathis-like" refers to fossil resins derived from ancient plants of the genus Agathis or only related to it. Cheirolepidiaceous ambers are also included as they are highly similar in chemistry to Agathis resins [8, 77]. Since the proposed botanical origin of Miocene amber is not questioned, we use Hymenaea to refer to both living trees and resins, as well as fossil resins derived from Hymenaea.

We compared the arthropod assemblages obtained from yellow sticky and Malaise traps and those preserved within Defaunation resin from extant trees, with the arthropod assemblages preserved in amber from Late and Early Cretaceous and Middle Miocene from different localities around the world. All the amber, copal, and Defaunation resin collections included in the present study have been selected because they are unbiased and collected for scientific purposes; they have been collected in the field without prior selection of arthropod inclusions and/or whether the pieces/lumps containing arthropod inclusions or not. Searching for arthropod inclusions and preparing the pieces is a process that has been carried out in laboratory. With this purpose, we persuaded a detailed composition of the arthropod assemblages identified from Cretaceous amber from Spain and France, Eocene amber from Australian, Oligocene and Miocene amber from New Zealand, Miocene amber from the Dominican Republic and Mexico, copal and Defaunation resin from the Dominican Republic. Collections with the specification of the inclusions per gram of resin are mentioned in Table 1.

Table 1 can be placed here. The references in the table are numbered from this point onwards.

The term bioinclusion in amber, copal, and Defaunation resin includes all organismal remains that can be found in fossil resins. However, since arthropod inclusions are the only bioinclusions included in our analyses, we will use the term arthropod inclusions to refer to the bioinclusions we study here.

The copal and Defaunation resin samples were polished and prepared in the same manner as amber [85, 86] to correctly identify the bioinclusions. However, for some lumps, it was only necessary to create a small viewing window to observe their contents.

Imaging

The photographs were performed with digital cameras Canon EOS 40D and Canon EOS 70D. Figures were performed using Adobe Photoshop software (version 25.4 www.adobe.com).

Data processing

The data used in this study were frequency data of taxa and experimental groups (amber, copal, and Defaunation resin, and yellow sticky and Malaise traps). Multivariate analyses were used to evaluate the different hypotheses of the study, namely, multidimensional scaling (MDS), and representation of the relationships for each experimental group in the form of a correlation heat-map or correlation network.

Statistical analyses were performed using different R functions and libraries [87]. The BDbiost3 library for R [ 88, 89] was used for exploratory data analysis, and multidimensional scaling used functions LinesMDS() (see Additional file 1 for more detail on the statistical methods used [90, 91]). All data used for statistical work are available in Additional file 2.

To estimate the degree of uniformity of arthropod inclusions in Defaunation resin from Agathis and Hymenaea, we estimated the probability density function (PDF) that a piece lacks arthropod inclusions given the mass of the piece (see Additional file 1 for a discussion of this methodology [92]). For this, we measured the mass of pieces with an error of 0.01g and we counted the number of arthropod inclusions in each piece (see Additional file 2). Then, we calculated the frequencies of pieces without arthropod inclusions and with a mass in intervals of 0.01g for the Defaunation resins from the Agathis and Hymenaea trees. These frequencies were associated with the conditional probability of finding a piece of mass m, given that it does not contain arthropod inclusions. Bayes' theorem was then used to obtain the conditional probability of not having arthropod inclusions given that the piece has mass m. Finally, we fit the resulting probabilities with power laws of the form by first obtaining the logarithm of the data, then using standard least squares algorithm with the fitting model log(a) - b log(m) to obtain the values of a and b for both collections. We used b as a measurement of the uniformity of arthropod inclusions. The exponent b measures the uniformity of arthropod inclusions, while the parameter a is just the estimated value of the PDF that a piece of resin has no arthropod inclusions within one gram of resin, and was only used to fit the data correctly.

Abbreviations for Figure 4: AAES: Amber from Ariño, Spain; AAFR: Amber from Archingeay-Les Nouillers, France, mentioned in Perrichot et al. [21]; AAIFR: Amber Aix island, France; AAPES: Amber from Arroyo de la Pascueta, Spain; ABFR: Amber La Buzinie, France; ACFR: Amber Cadeuil, France; ADCO: Amber Doumanga, Congo; ADOG.P: Dominican amber collection from the private collection mentioned in Poinar [93]; ADOJ.C: Dominican amber from the Jorge Caridad private collection; AFFR: Amber Fouras, France; AFTFR: Amber Fourtou, France; AHES: Amber from La Hoya, Spain; AMX: Mexican amber collections mentioned in Solórzano-Kraemer et al. [29]; APES: Amber form Peñacerrada I, Spain; ARFR: Amber Les Renardières, France; ASES: Amber from El Soplao, Spain; ASJES: Amber from San Just, Spain; ASFR: Amber Salignac, France; CDO: Copal collected in 2019 in Cotuí, Dominican Republic; CTZ: Copal from Tanzania; MMG: Malaise trap installed close to Hymenaea in 2013 in Madagascar; MMX: Malaise trap installed close to Hymenaea since 2010 to 2012 in Mexico; MNC: Malaise trap installed close to Agathis in 2016 in New Caledonia; RMG1: Defaunation resin collected from Hymenaea tree trunks in 2013 in Madagascar; RMG2: Defaunation resin collected from Hymenaea tree trunks in 2015 in Madagascar; RNC: Defaunation resin collected from Agathis tree trunks in 2016 in New Caledonia; ST0MG, ST1MG, ST2MG: Yellow sticky traps placed at 0, 1 and 2 meters on Hymenaea tree trunks in Madagascar, respectively; ST0NC, ST1NC, ST2NC: Yellow sticky traps placed at 0, 1 and 2 meters in 2016 on Agathis tree trunks in New Caledonia, respectively; STTMG: Total yellow sticky traps placed around Hymenaea in 2013 in Madagascar; STTMX: Total yellow sticky traps placed around Hymenaea tree trunks in 2012 in Mexico; STTNC: Total yellow sticky traps placed around Agathis tree trunks in 2016 in New Caledonia.

Acknowledgements

We thank the family Caridad, owners of the Museo Mundo del Ámbar in Santo Domingo, S.B. Brower, F.A. Aquel Fernández, and Y.H. Shih (Dominican Republic) for their invaluable support, assistance during the scientific fieldwork, and donation of amber and copal/Defaunation resin pieces. Special thanks are given to R. Ravelomanana, M. Asensi, M. Madiomanana, and J. Andrianabo (Madagascar) for their dedicated assistance during fieldwork. Gratitude is expressed to Dr T. Rakotondrazafy, Director of the Départ. de Paléontologie et Anthropologie Biologique, and Dr E.M. Randrianarisoa, Director of the Départ. d’Entomologie, both at the Université d’Antananarivo (Madagascar), for their valuable help and advice in navigating administrative arrangements. The authors wish to acknowledge the contribution of the team at the Malagasy Institute for the Conservation of Tropical Environments (ICTE/MICET) for their support in the administrative development of the fieldwork in Madagascar. Thanks also go to J. Manauté and J. Delafenêtre from Parc Provincial de la Rivière Bleue, New Caledonia, for their support and assistance during fieldwork. The Direction de l’Environnement (DENV) Province Sud from Nouvelle Calédonie is acknowledged for granting the necessary fieldwork permits. Appreciation is extended to R. Kunz of the Senckenberg Research Institute (SMF) for his efforts in sorting, preparing and cataloguing the resin and copal collection deposited at the SMF and for the photos in the Figure 6H and I. Further gratitude is directed toward R. López del Valle (Spain) for their contribution to sorting, preparing, and cataloguing the Spanish amber. Thanks are due to Eduardo Espílez from Fundación Conjunto Paleontológico de Teruel-Dinópolis for cataloguing and curation of the Cretaceous Spanish amber from Teruel, and the director of that institution Dr Alberto Cobos. We would like to thank the three anonymous reviewers for their valuable feedback and corrections, which helped to improve the manuscript. We would like to extend our gratitude to José Antonio Peñas for his artistic illustration of Fig. 1.

Consent for publication

Authors agree to publication.

Funding

This work was supported by the Spanish Ministerio de Ciencia, Innovación y Universidades scientific projects CGL2017-84419/ AEI/FEDER, UE, and PID2022-137316NB, funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU; by DEGAPA-UNAM through the project PAPIIT IN113923, by the programme RYC2022-037026-I, funded by AEI/10.13039/ 501100011033 and the FSE+; by the Consejería de Industria, Turismo, Innovación, Transporte y Comercio of the Gobierno de Cantabria through the semi-public enterprise EL SOPLAO S.L. [research agreement #20963 with Universitat de Barcelona and research contract Ref. VAPC 20225428 to Instituto Geológico y Minero de España – Consejo Superior de Investigaciones Científicas, both 2022–2025]; the German VolkswagenStiftung (Project N. 90946), and the DFG project 457837041 (SO 894/6-1).

Data availability

All the data needed to evaluate the conclusions of the paper are available in the paper and in the supplementary information files. The amber from “El Valle – 7 Cañadas” and San Rafael, Dominican Republic have been acquired directly in the mine by MMS-K, XD and EP. The copal/Defaunation resin from Cotuí, Dominican Republic, have been acquired by Y.H. Shih. These amber and copal/Defaunation resin will be housed in the Museo Nacional de Historia Natural ‘Prof. Eugenio de Jesús Marcano’ in Santo Domingo, Dominican Republic. Correspondence for material related to this paper can be sent to Mónica M. Solórzano-Kraemer ([email protected]), Enrique Peñalver ([email protected]), and Xavier Delclòs ([email protected]).

Authors’ contributions

M.M.S-K., E.P. and X.D. conceived the idea, designed the methodology, analysed the data and drafted the manuscript. M.M.S-K., designed the first draft. A.M-G. and A.S.K. performed the statistical analysis. M.M.S-K., E.P., X.D., M.C.M.H., D.P., A.A., V.P. and M.P. identified the specimens, and provided the list of specimens, M.M.S-K., E.P., X.D., E.B. and R.G. carried out the field work. M.M.S.-K., E.B., E.P. and X.D. acquired and managed project funding. All authors contributed critically to the drafts and gave final approval of the manuscript.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Not applicable.

Ethics statements

Sampling materials were acquired through permits from the Government of New Caledonia and Madagascar, and the relevant permit numbers are as follows: In New Caledonia the Direction de l’Environnment de la Province Sud (DENV) permitted us to work, the sampling and exportation Permit is 3021-2016/ARR/DENV; and in Madagascar, The Ministère de l’Environnement, de l’Écologie et des Forêts gave us permission to work in the Malagasy protected areas, the sampling Permits are 192/13 and N. 060/15, with exportation permits 192/13/MEF/SG/DGF/DCB.SAP/SCB. 060/15/ MEEF/SG/DGF/DCB.SAP/SCB.

Martínez-Delclòs X, Briggs DE, Peñalver E. Taphonomy of insects in carbonates and amber. Palaeogeogr Palaeoclimatol Palaeoecol. 2004;203(1–2):19–64. https://doi.org/10.1016/S0031-0182(03)00643-6
Labandeira CC. Amber. In: LaFlamme M, Schiffbauer JD, and Darroch S A, editors. Reading and Writing of the Fossil Record: Preservational Pathways to Exceptional FossilizationBritish Columbia: Paleontol Soc.;2014. p. 163–215.
Seyfullah LJ, Beimforde C, Dal Corso J, Perrichot V, Rikkinen J, Schmidt AR. Production and preservation of resins—past and present. Biol Rev. 2018;93(3):1684–714. https://doi.org/10.1111/brv.12414
Lozano RP, Pérez-de la Fuente R, Barrón E, Rodrigo A, Viejo JL, Peñalver E. Phloem sap in Cretaceous ambers as abundant double emulsions preserving organic and inorganic residues. Sci Rep. 2020;10(1):9751. https://doi.org/10.1038/s41598-020-66631-4
Delclòs X, Peñalver E, Barrón E, Peris D, Grimaldi DA, Holz et al. Amber and the Cretaceous Resinous Interval. Earth-Sci Rev. 2023;104486. https://doi.org/10.1016/j.earscirev.2023.104486
Grimaldi DA, Beck CW, Boon JJ. Occurrence, chemical characteristics, and paleontology of the fossil resins from New Jersey. Am Mus Novit. 1989;2948:1–28.
Anderson KB, Winans RE, Botto RE. The nature and fate of natural resins in the geosphere—II. Identification, classification and nomenclature of resinites. Org Geochem. 1992;18(6):829–41. https://doi.org/10.1016/0146-6380(92)90051-X
Nohra YA, Perrichot V, Jeanneau L, Le Pollès L, Azar D. Chemical characterization and botanical origin of French ambers. J. Nat. Prod. 2015;78:1284–1293. https://doi.org/10.1021/acs.jnatprod.5b00093
Menor-Salván C, Simoneit BR, Ruiz-Bermejo M, Alonso J. The molecular composition of Cretaceous ambers: Identification and chemosystematic relevance of 1,6-dimethyl-5-alkyltetralins and related bisnorlabdane biomarkers. Org Geochem. 2016;93:7–21. https://doi.org/10.1016/j.orggeochem.2015.12.010
Goeppert HR, Berendt GC. Der Bernstein und die in ihm befindlichen Pflanzenreste der Vorwelt. In: Berendt GC, editor. Die im Bernstein befindlichen organischen Reste der Vorwelt. Vol. 1. Berlin: Commission der Nicolaischen Buchhandlung; 1845. p. 1–125.
Wolfe AP, Tappert R, Muehlenbachs K, Boudreau M, McKellar RC, Basinger JF, et al. A new proposal concerning the botanical origin of Baltic amber. Proc R Soc B Biol Sci. 2009;276(1672):3403–12. https://doi.org/10.1098/rspb.2009.0806
Rust J, Singh H, Rana RS, McCann T, Singh L, Anderson K, et al. Biogeographic and evolutionary implications of a diverse paleobiota in amber from the early Eocene of India. Proc Natl Acad Sci. 2010;107(43):18360–18365. https://doi.org/10.1073/pnas.1007407107
Solórzano-Kraemer MM, Delclòs X, Engel MS, Peñalver E. A revised definition for copal and its significance for palaeontological and Anthropocene biodiversity-loss studies. Sci Rep. 2020;10(1):1–12. https://doi.org/10.1038/s41598-020-76808-6
Wang B, Shi G, Xu C, Spicer R, Perrichot V, Schmidt AR, et al. The mid-Miocene Zhangpu biota reveals an outstandingly rich rainforest biome in East Asia. Sci Adv. 2021;7(18):eabg0625. doi:10.1126/sciadv.abg0625. https://doi.org/10.1126/sciadv.abg0625
Maksoud S, Azar D. Lebanese amber: latest updates. Palaeoentomology 2020;3(2):125–55. https://doi.org/10.11646/palaeoentomology.3.2.2
Maksoud S, Azar D. Palaeoentomological outcrops in Lebanon. The 9th International Conference on Fossil Insects, Arthropods and Amber; 2023. p. 63.
Peñalver E, Delclòs X. Spanish amber. In: Penney D, editor. Biodiversity of fossils in amber from the major world deposits. Manchester, UK: Siri Scientific Press; 2010. p. 236–70.
Peñalver E, Arillo A, López del Valle R, Delclòs X Cretaceous Spanish amber: History, research and checklist of taxa. Palaeoentomology 2025;8(2):195–234. doi: 10.11646/palaeoentomology.8.2.10
Nel A, De Ploëg G, Millet J, Menier JJ, Waller A. The French ambers: a general conspectus and the Lowermost Eocene amber deposit of Le Quesnoy in the Paris Basin. Geol. Acta 2004;2: 3–8.
Perrichot V. Environnements paraliques à ambre et à végétaux au Crétacé nord‐aquitain (Charentes, Sud‐Ouest de la France). Mém Geosci Rennes. 2005;118:1–310.
Perrichot V, Néraudeau D, Nel A, De Ploeg G. A reassessment of the Cretaceous amber deposits from France and their palaeontological significance. Afr Invertebr. 2007;48(1):213–27.
Perrichot V, Néraudeau D. Introduction to thematic volume "Fossil arthropods in Late Cretaceous Vendean amber (northwestern France)". Paleontol. Contrib. 2014;10A:1–4. https://doi.org/10.17161/PC.1808.15981
Saint Martin JP, Dutour Y, Ebbo L, Frau C, Mazière B, Néraudeau D, Saint Martin S, Tortosa T, Turini E, Valentin X, 2021. Reassessment of amber-bearing deposits of Provence, southeastern France. BSGF - Earth Sci. Bull. 2021;192(5):1-22.
Cadena EA, Mejia-Molina A, Brito CM, Peñafiel S, Sanmartin KJ, Sarmiento LB. New Mesozoic and Cenozoic fossils from Ecuador: Invertebrates, vertebrates, plants, and microfossils. J South Am Earth Sci. 2018;83:27–36.
Pereira R, de Lima FJ, Simbras FM, Bittar SMB, Kellner AWA, Saraiva Á, et al. Biomarker signatures of Cretaceous Gondwana amber from Ipubi Formation (Araripe Basin, Brazil) and their palaeobotanical significance. J South Am EarthSci. 2020;98:102413. https://doi.org/10.1016/j.jsames.2019.102413
Castaño-Cardona RF, Jaramillo C, Pardo-Trujillo A, Vento B, Quiroz-Cabascango D, Angulo-Pardo E. Palynology of Cretaceous amber deposits of the Oriente Basin-Ecuador and the Eastern Cordillera-Colombia. Bol Geol. 2023;45(3):63–77.
Delclòs X, Peñalver E, Jaramillo C, Cadena E, Menor-Salván C, Román JL, et al. Cretaceous Amber of Ecuador: Unveiling new Insights into South America's Gondwanan Forests. Commun Earth Environ. 2025; doi.org/10.1038/s43247-025-02625-2
McCoy VE, Soriano C, Gabbott SE. A review of preservational variation of fossil inclusions in amber of different chemical groups. Earth Environ Sci Trans R Soc Edinb. 2016;107(2–3):203–11. https://doi.org/10.1017/S1755691017000391
Solórzano-Kraemer MM, Kraemer AS, Stebner F, Bickel DJ, Rust J. Entrapment bias of arthropods in Miocene amber revealed by trapping experiments in a tropical forest in Chiapas, Mexico. PLoS One. 2015;10(3):e0118820. https://doi.org/10.1371/journal.pone.0118820
Solórzano-Kraemer MM, Delclòs X, Clapham ME, Arillo A, Peris D, Jäger P, et al. Arthropods in modern resins reveal if amber accurately recorded forest arthropod communities. Proc Natl Acad Sci. 2018;115:6739–44. https://doi.org/10.1073/pnas.1802138115
Seyfullah LJ, Roberts EA, Jardine PE, Schmidt AR. Experimental induction of resins as a tool to understand variability in ambers. Foss Rec. 2021;24(2):321–37. https://doi.org/10.5194/fr-24-321-2021
Seyfullah LJ, Roberts EA, Jardine PE, Rikkinen J, Schmidt AR. Uncovering the natural variability of araucariacean exudates from ex situ and in situ tree populations in New Caledonia using FTIR spectroscopy. PeerJ Anal Chem. 2022;4:e17.
Whitmore TC, Page CN. Evolutionary implications of the distribution and ecology of the tropical conifer Agathis. New Phytol. 1980;84(2):407–16. https://www.jstor.org/stable/2431735
Langenheim JH. Biology of amber-producing trees: focus on case studies of Hymenaea and Agathis. In: Anderson KB, Crelling JC, editors. Amber, resinite, and fossil resins. Washington, DC: Am Chem Soc; 1995. p. 1–31.
Stilwell JD, Langendam A, Mays C, Sutherland LJ, Arillo A, Bickel DJ, et al. Amber from the Triassic to Paleogene of Australia and New Zealand as exceptional preservation of poorly known terrestrial ecosystems. Sci Rep. 2020;10(1):1–11. https://doi.org/10.1038/s41598-020-62252-z
Lee YT, Langenheim JH. A systematic revision of the genus Hymenaea (Leguminosae; Caesalpinioideae, Detarieae). Univ Calif Publ. 1975;69:1–109.
Langenheim JH. Plant resins: chemistry, evolution, ecology, and ethnobotany. Oregon, US: Timber Press; 2003.
Fougère-Danezan M, Herendeen PS, Maumont S, Bruneau A. Morphological evolution in the variable resin-producing Detarieae (Fabaceae): do morphological characters retain a phylogenetic signal? Ann Bot. 2010;105:311–25.
Delclòs X, Peñalver E, Ranaivosoa V, Solórzano-Kraemer MM. Unravelling the mystery of “Madagascar copal”: Age, origin and preservation of a Recent resin. PLoS One 2020;15(5):e0232623.
McCoy VE, Boom A, Solórzano-Kraemer MM, Gabbott SE. The chemistry of American and African amber, copal, and resin from the genus Hymenaea. Org Geochem. 2017;113:43–54. https://doi.org/10.1016/j.orggeochem.2017.08.005
Solórzano-Kraemer MM, Peñalver E, Herbert MC, Delclòs X, Brown VB, Nyi Aung N, et al. Necrophagy by insects in Oculudentavis and other lizard body fossils preserved in Cretaceous amber. Sci Rep. 2023;13(1):2907. https://doi.org/10.1038/s41598-023-29612-x
Mizumoto N, Hellemans S, Engel MS, Bourguignon T, Buček A. Extinct and extant termites reveal the fidelity of behavior fossilization in amber. PNAS 2024;121(12): e2308922121.
Penney D. Biodiversity of fossils in amber from the major world deposits. Manchester, UK: Siri Scientific Press; 2010.
Alonso J, Arillo A, Barrón E, Corral JC, Grimalt J, López JF, et al. A new fossil resin with biological inclusions in Lower Cretaceous deposits from Álava (Northern Spain, Basque-Cantabrian Basin). J Paleontol. 2000;74(1):158–78.
Schmidt AR, Kaulfuss U, Bannister JM, Baranov V, Beimforde C, Bleile N, et al. Amber inclusions from New Zealand. Gondwana Res. 2018;56:135–46.
Álvarez-Parra S, Pérez-de La Fuente R, Peñalver E, Barrón E, Alcalá L, Pérez-Cano J, et al. Dinosaur bonebed amber from an original swamp forest soil. eLife. 2021;10:e72477.
Speranza M, Ascaso C, Delclòs X, Peñalver E. Cretaceous mycelia preserving fungal polysaccharides: taphonomic and paleoecological potential of microorganisms preserved in fossil resins. Geol Acta. 2015;13(4):363–85.
Cabrita P. A model for resin flow. In: Ramawat KG, Ekiert HM, Goyal S. editors. Plant cell and tissue differentiation and secondary metabolites. Springer International Publishing; 2019. p.17–44.
Peñalver E, Speranza M, Delclòs X. Fungal cortices in kauri resin from New Zealand as taphonomic and ecological analogues of Cretaceous amber cortices: an actualistic approach. Span J Palaeontol. 2025;1–14. doi:10.7203/sjp.29855.
Phillips MA, Croteau RB. Resin-based defenses in conifers. Trends Plant Sci. 1999;4:184–90. doi.org/10.1016/S1360-1385(99)01401-6.
McKay SAB, Hunter WL, Godard KA, Wang SX, Martin DM, Bohlmann J, et al. Insect attack and wounding induce traumatic resin duct development and gene expression of (—)-pinene synthase in Sitka spruce. Plant Physiol. 2003;133(1):368–78.
McCoy VE, Soriano C, Pegoraro M, Luo T, Boom A, Foxman B, Gabbott SE. Unlocking preservation bias in the amber insect fossil record through experimental decay. PLoS One 2018;13(4):e0195482.
Otto A, Wilde V. Sesqui-, di-, and triterpenoids as chemosystematic markers in extant conifers—a review. Bot Rev. 2001;67:141–238.
Tappert R, Wolfe AP, McKellar RC, Tappert MC, Muehlenbachs K. Characterizing modern and fossil gymnosperm exudates using micro-Fourier transform infrared spectroscopy. Int J Plant Sci. 2011;172(1):120–38.
Cunningham A, Gay ID, Oehlschlager AC, Langenheim JH. 13C NMR and IR analyses of structure, aging, and botanical origin of Dominican and Mexican ambers. Phytochemistry 1983;22(4):965–8.
Poulin J, Helwig K. Inside amber: New insights into the macromolecular structure of Class Ib resinite. Org Geochem. 2015;86:94–106.
Weir BS, Paderes EP, Anand N, Uchida JY, Pennycook SR, Bellgard SE, et al. A taxonomic revision of Phytophthora Clade 5 including two new species, Phytophthora agathidicida and P. cocois. Phytotaxa 2015;205(1):21–38.
Langenheim JH. Plant resins. Am Sci. 1990;78:16–24.
Langenheim JH. Higher plant terpenoids: A phytocentric overview of their ecological roles. J Chem Ecol. 1994;20:1223–80.
Whitmore TC. Utilization, potential, and conservation of Agathis, a genus of tropical Asian conifers. Econ Bot. 1980;34(1):1–12.
Estruch RA. Gum base. In: Fritz D, editor. Formulation and Production of Chewing and Bubble Gum. 2nd ed. Essex: Kennedy’s Publications Ltd.; 2008. p. 93–118.
Horváth G, Egri Á, Meyer-Rochow VB, Kriska G. How did amber get its aquatic insects? Water-seeking polarotactic insects trapped by tree resin. Hist Biol. 2021;33(6):846–56.
McCoy VE, Barthel HJ, Boom A, Peñalver E, Delclòs X, Solórzano-Kraemer MM. Volatile and semi-volatile composition of Cretaceous amber. Cretac Res. 2021;127:104958.
Chinou I. Labdanes of natural origin—biological activities (1981–2004). Curr Med Chem. 2005;12(11):1295–317.
Hautevelle Y, Michels R, Malartre F, Trouiller A. Vascular plant biomarkers as proxies for palaeoflora and palaeoclimatic changes at the Dogger/Malm transition of the Paris Basin (France). Org Geochem. 2006;37(5):610–25.
Lu Y, Hautevelle Y, Michels R. Determination of the molecular signature of fossil conifers by experimental palaeochemotaxonomy–Part 1: The Araucariaceae family. Biogeosciences 2013;10(3), 1943–1962.
Achotegui-Castells A. The role of terpenes in the defensive responses of conifers against herbivores and pathogens. Barcelona: Universitat Autònoma de Barcelona; 2015. [PhD dissertation]. https://ddd.uab.cat/pub/tesis/2015/hdl_10803_323095/aac1de1.pdf
Terry I, Walter GH, Moore C, Roemer R, Hull C. Odor-mediated push-pull pollination in cycads. Science. 2007;318(5847):70.
Brown BV. Malaise trap catches and the crisis in Neotropical dipterology. Am Entomol. 2005;51(3):180–3.
Disney RHL, Coluson JC, Butterfield J. A survey of the scuttle flies (Diptera: Phoridae) of upland habitats in Northern England. Naturalist 1981;106: 53–66.
Solórzano-Kraemer MM, Brown BV. Dohrniphora (Diptera: Phoridae) from the Miocene Mexican and Dominican ambers with a paleobiological reconstruction. Insect Syst Evol. 2018;49(3):299–327.
Borkent A, Brown B, Adler PH, Amorim DDS, Barber K, Bickel D, et al. Remarkable fly (Diptera) diversity in a patch of Costa Rican cloud forest: Why inventory is a vital science. Zootaxa. 2018;4402(1):53–90.
Brown BV, Borkent A, Adler PH, Amorim DDS, Barber K, Bickel D. Comprehensive inventory of true flies (Diptera) at a tropical site. Commun Biol. 2018;1(1):21.
Puckett RT, Calixto AA, Reed JJ, McDonald DL, Drees BB, Gold RE. Effectiveness comparison of multiple sticky-trap configurations for sampling Pseudacteon spp. phorid flies (Diptera: Phoridae). Environ Entomol. 2013;4:763–9.
Puckett RT, Calixto A, Barr CL, Harris M. Sticky traps for monitoring Pseudacteon parasitoids of Solenopsis fire ants. Environ Entomol. 2014;36(3):584–8.
Cornaby BW. Carrion reduction by animals in contrasting tropical habitats. Biotropica. 1974;51:51–63.
Menor-Salván C, Najarro M, Velasco F, Rosales I, Tornos F, Simoneit BR. Terpenoids in extracts of Lower Cretaceous ambers from the Basque-Cantabrian Basin (El Soplao, Cantabria, Spain): paleochemotaxonomic aspects. Org Geochem. 2010;41(10):1089–1103.
Girard V, Neraudeau D, Breton G, Morel N. Palaeoecology of the Cenomanian amber forest of Sarthe (western France). Geol Acta.2013;11(3): 321–330.
Néraudeau D, Allain R, Perrichot V, Videt B, de Broin FDL, Guillocheau F. Découverte d’un dépôt paralique à bois fossiles, ambre insectifère et restes d’Iguanodontidae (Dinosauria, Ornithopoda) dans le Cénomanien inférieur de Fouras (Charente-Maritime, Sud-Ouest de la France). Comptes Rendus Palevol 2003;2(3):221–230.
Néraudeau D, Vullo R, Gomez B, Girard V, Lak M, Videt B et al. Amber, plant and vertebrate fossils from the Lower Cenomanian paralic facies of Aix Island (Charente-Maritime, SW France). Geodiversitas 2009;31(1):13–27.
Neraudeau D, Perrichot V, Colin JP, Girard V, Gomez B, Guillocheau F. A new amber deposit from the Cretaceous (uppermost Albian-lowermost Cenomanian) of southwestern France. Cretac Res. 2008;29(5–6):925–929.
Chaler R, Grimalt JO. Fingerprinting of Cretaceous higher plant resins by infrared spectroscopy and gas chromatography coupled to mass spectrometry. Phytochem Anal. 2005;16(6):446–50.
Peñalver E, Fernández BG, del Valle RL, Barrón E, Fernández RPL, Rodrigo A, et al. Un nuevo yacimiento de ámbar cretácico en Asturias (norte de España): Resultados preliminares de la excavación paleontológica de 2017 en La Rodada (La Manjoya). Yacimientos Paleontológicos Excepcionales en la Península Ibérica: XXXIV Jornadas de Paleontología y IV Congreso Ibérico de Paleontología. Soc Esp Paleontol. 2018;289–99.
Bouju V, Perrichot V. A review of amber and copal occurrences in Africa and their paleontological significance. Bull. Soc. geol. Fr. 2020;191(17):1–11.
Corral JC, López del Valle R, Alonso J. El ámbar cretácico de Álava (Cuenca Vasco-Cantábrica, Norte de España). Su colecta y preparación. Est Mus Cienc Nat de Álava 1999;14 Núm. Espec. 2: 7–21.
Sadowski EM, Schmidt AR, Seyfullah LJ, Solórzano-Kraemer MM, Neumann C, Perrichot V, et al. Conservation, preparation and imaging of diverse ambers and their inclusions. Earth-Sci Rev. 2021;220:103653. https://doi.org/10.1016/j.earscirev.2021.103653
R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2024. https://www.R-project.org/
Monleón-Getino A, Rodríguez-Casado C, Mendez-Viera J. How to calculate number of samples in the design of pre/pro-biotics studies (metagenomic studies). III WorkShop Insa-UB 2017; 1–2. doi:10.13140/RG.2.2.27611.67367
Monleón-Getino A. T. Library for R BDSBIOST3: Machine Learning and Advanced Statistical Methods for Omic categorical analysis and others. 2020. https://github.com/amonleong/BDSbiost3
Hout MC, Papesh MH, Goldinger SD. Multidimensional scaling. Wiley Interdiscip Rev Cogn Sci. 2013;4(1):93–103. doi:10.1002/wcs.1203.
Rodríguez-Casado C, Monleón-Getino T, Alcolea M. A priori groups based on Bhattacharyya distance and partitioning around medoids algorithm (PAM) with applications to metagenomics. IOSR J Math. 2017;13(3):24–32.
Ross SM. A first course in probability. 5th ed. USA: Pearson; 1997.
Poinar GO, Poinar R. The amber forest: a reconstruction of a vanished world. New Jersey: Princeton University Press; 1999

Table 1. Number of arthropod inclusions per gram in amber, copal, or Defaunation resin that are reported in the literature and own data. The amber, copal, and Defaunation resin (collected directly from the trees) from the different areas were not pre-selected in terms of containing or not containing arthropod inclusions. *The table includes collections that are unbiased in terms of arthropods per gram; unbiased collections only in terms of no pre-selection of arthropod species are not included here.** Stilwell et al. [35] do not specify a source for this amber, but its age and provenance suggest Agathis sp.

Locality/Area of collection	Age of resin	Resiniferous tree	Weight / Pieces	N. of arthropod inclusions	Citation
Mananjary region (between Nosy Varika and Ambahy), Madagascar	Resin collected in 2013	Hymenaea verrucosa (Caesalpiniaceae)	800.5 grams	1743	Solórzano-Kraemer et al. [30]
Sacaramy (close to Antsiranana, Diego Suarez), Madagascar	Resin collected in 2015	Hymenaea verrucosa (Caesalpiniaceae)	1500 grams	2141	Own data
Col de Yaté South Province (Gramseat South), New Caledonia	Resin collected in 2016	Agathis ovata (Araucariaceae)	1627.1 grams	40	Own data
Bon Secours (close to Rivière Bleue Provincial Park, South Province, Gramseat South), New Caledonia	Resin collected in 2016	Agathis lanceolata (Araucariaceae)	991.6 grams	8	Own data
Copal / Defaunation resin from Cotuí, Dominican Republic	Unknown	Hymenaea courbaril (Caesalpiniaceae)	1875.2 grams	319	Own data
Amber from Totolapa, Chiapas, Mexico.	early Miocene	†Hymenaea mexicana Poinar and Brown, 2002 (Caesalpiniaceae)	2000 grams	107	Solórzano-Kraemer et al. [29]
Amber from El Valle 7 Cañadas, Dominican Republic	early Miocene	†Hymenaea protera Poinar, 1991 (Caesalpiniaceae)	1678.8 grams	270	Own data
Amber from San Rafael, Dominican Republic	early Miocene	†Hymenaea protera (Caesalpiniaceae)	523.5 grams	140	Own data
Amber from Southland region of the South Island, and Otago, New Zealand	late Oligocene and early Miocene	Agathis sp.(Araucariaceae)	the exact amount is unknown, estimation: 1000 to 1500 grams	78	Schmidt et al. [45] (Alexander Schmidt, pers. comm., February 2023)
Amber from Anglesea Coal Measures (ACM), Australia	early Eocene	*Agathis sp	ca. 2000 grams	47	Stilwell et al. [35], own data 2025
Amber from Macquarie Harbour Formation (MHF), Australia	early Eocene	*Agathis sp	ca. 1500 grams	2	Stilwell et al. [35], own data 2025
Amber from Fourtou, France	middle Cenomanian	†Agathoxylon sp. (Araucariaceae) or †Cheirolepidiaceae	ca. 2000 grams	40	Girard et al. [78], own data 2025
Amber from Fouras/Bois Vert, France	early Cenomanian	†Agathoxylon gardoniense(Araucariaceae) and †Cheirolepidiaceae	ca. 3000 grams	113	Néraudeau et al. [79], Perrichot et al. [21], own data 2025
Amber from Ile d’Aix, France	early Cenomanian	†Agathoxylon gardoniense (Araucariaceae) and †Cheirolepidiaceae	ca. 500 grams	6	Néraudeau et al. [80], own data 2025
Amber from La Buzinie, France	early Cenomanian	†Agathoxylon gardoniense (Araucariaceae) and †Cheirolepidiaceae	ca. 6000 grams	149	Perrichot et al. [21], own data 2025
Amber from Salignac, France	early Cenomanian	Araucariaceae	ca. 400 grams	27	Perrichot et al. [21], own data 2025
Amber from Archingeay-Les Nouillers, France	late Albian	†Agathoxylon gardoniense (Araucariaceae) and †Cheirolepidiaceae	35000 to 40000 grams	1330	Perrichot et al. [21], own data 2025
Amber from Les Renardières, France	late Albian	†Agathoxylon gardoniense (Araucariaceae) and †Cheirolepidiaceae	300 grams	4	Perrichot et al. [21], own data 2025
Amber from Cadeuil, France	late Albian	†Agathoxylon gardoniense (Araucariaceae) and †Cheirolepidiaceae	ca. 8000 grams	98	Néraudeau et al. [81], own data 2025
Amber from Peñacerrada, Álava, Spain	late Albian (Early Cretaceous)	Agathis-like (Araucariaceae) (Chaler and Gramsimalt [82])	139500 grams	3346	Own data
Amber from San Just, Teruel, Spain	late Albian (Early Cretaceous)	Agathis-like (Araucariaceae)	12900 grams	387	Own data
Amber from La Hoya, Castellón, Spain	early Cenomanian (Early Cretaceous)	Agathis-like (Araucariaceae)	5500 grams	11	Own data
Amber from El Soplao, Cantabria, Spain	middle Albian (Early Cretaceous)	Frenelopsis sp. (†Cheirolepidiaceae) and other possible Cupressaceae (Menor-Salván et al. [77])	19937 grams	1600	Own data
Amber from Arroyo de la Pascueta, Teruel, Spain	late Albian (Early Cretaceous)	Agathis-like (Araucariaceae)	7000 grams	14	Own data
Amber from La Rodada, La Manjoya, Spain	late Albian (Early Cretaceous)	Agathis-like (Araucariaceae)	500 grams	2	Peñalver et al. [83]
Amber from Ariño, Teruel, Spain	early Albian (Early Cretaceous)	Agathis-like (Araucariaceae) (Álvarez-Parra et al. [46])	1128 grams	100	Own data
Amber from Doumanga, Congo	middle Aptian	†Agathoxylon sp. (Araucariaceae) or †Cheirolepidiaceae	2550 grams	47	Bouju & Perrichot [84], own data 2025

No competing interests reported.

Download PDF

Journal Publication

published 06 Nov, 2025

Read the published version in BMC Biology →

Editorial decision: Accepted
22 Oct, 2025
Submission checks completed at journal
09 Oct, 2025
First submitted to journal
08 Oct, 2025

You are reading this latest preprint version

Agathis vs. Hymenaea – trapping biases to interpret arthropod assemblages in ambers

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Results

Discussion

Conclusion

Methods

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1