Wild food plants and fungi sold in the markets of Luang Prabang, Lao PDR

Background Open air markets hold an important position for ethnobiologists. In Southeast Asia, they are seriously understudied, in spite of their incredible biocultural diversity. In order to fill this gap we recorded plants and fungi sold in the open air markets of Luang Prabang, Lao PDR. Methods The markets were visited 38 times in four seasons: the dry season, early monsoon, mid-monsoon, and end-of-monsoon, at least 8 times per season. All items were photographed and voucher specimens were collected. Fungi were identified using DNA barcoding techniques. Results We recorded 110 species of wild edible plants and 54 species of fungi, including 49 wild-collected species. The sold plants included 86 species of green vegetables, 18 species of fruits and 3 species of flowers. Products from woody species constitute around half of all taxa sold. These include the young shoots of tree leaves, which are used for salads—an interesting feature of Lao cuisine. A large number of extremely rare Russula, with no reference sequences represented in databases or even species unknown to science is present on sale in the markets. Conclusions Luang Prabang markets are some of the richest in species of wild edible plants and fungi in Asia, and indeed in the whole world. It is worth pointing out the exceptionally long list of wild edible mushrooms which are sold in Luang Prabang (and probably elsewhere in Laos). We view the Morning Market of Luang Prabang as a cultural treasure that unites the traditions of eating a large number of living species with very diverse flora and fauna. Measures should be taken to strike a balance between local foraging traditions and nature conservation priorities.


Background
Open air markets hold an important position for ethnobiologists [1][2][3]. They are places where one can usually find the plants, animals, and fungi which are most important to a given culture, e.g., commonly eaten fruits, vegetables, or medicinal plants. Of course, some highly valued goods-plants with a sacred status (like entheogens) or illegal items, such as protected bush meat-may not be present in open air markets, but the bulk of most commonly consumed organisms usually is. Ethnobotanical studies of open air markets are a frequent topic of ethnobotanical enquiry and they have been performed in most geographical regions, including several countries of Eurasia (e.g., ). The oldest known ethnobiological market surveys were carried out by Hungarian and Polish researchers in the early twentieth century, in Budapest [38,39], Wilno (now Vilnius in Lithuania) [40] and Poznań [41,42]. Recently, return studies in the markets of Budapest and Poznań showed large changes in the list of sold plants and fungi compared to what was sold in the beginning of the twentieth century [4,43]. Bye's study from Mexico [1]was another important early work based on market surveys.
Ethnobiological studies of markets are an ethnobiologist's entry point to local food systems. This research situation allows for establishing quick contact with plant sellers (who are often responsible for their collection, or whose families collect the plants for them). The meeting in a public space enables a quick exchange of information of an ethnobiological character, concerning names of the sold organisms, their occurrence and properties. One of the disadvantages of market surveys is sometimes the difficulty of collecting classic voucher specimens, as usually only organs or parts of plants are on sale [2]. Fortunately, the difficulties in proper biological identification of these fragmentary items can be overcome by DNA barcoding [28,31,37].
Bearing in mind how easy it is to perform a market study, it is surprising that there are many parts of the world in which such studies have been made rarely or not at all. Open air markets are an important part of the eastern and south Asian rural economy, and even though Asia is the largest, most populous, and perhaps the most diverse of all continents, such studies are quite few and far between ( ;). Although surveys of wild edible plants and fungi sold in Southeast Asia are rare, some research effort has been put into studying the socioeconomic aspects of "green" open markets in general [44][45][46][47] as well as the contamination of plants with heavy metals [48] or parasites and pathogens [49][50][51]. Some studies from Southeast Asia performed in open air markets concern plant genetic resources from a single species, genus or family; the identification of the main cultivated plants (e.g., [52][53][54][55][56]); or medicinal plants [24,57].
Lao PDR is a diverse country with over 40 ethnic minorities and 11 thousand vascular plants species. Due to the very turbulent political and economic situation in Southeast Asia in the twentieth century, the biological diversity of Laos is still poorly described, in spite of increasing efforts to document its Traditional Knowledge and identify its non-timber forest products [58][59][60][61][62][63][64][65][66][67][68][69].
No lists of plants or fungi sold in particular markets have ever been published in Laos apart from a list of wild vegetables collected in rice fields and sold locally in Houaphan Province [23]. More surveys concerning animals, both vertebrates [70][71][72][73] and insects [74], have been performed. Some new species have been found in Lao markets, e.g., a new species of Impatiens (a dicot plant) [75] and a new species of rodent belonging to a new family, which was found being sold for meat in a local market [76]. Wildlife-both flora and fauna-is present in most Lao markets. Wild plants and animals are an important part of national cuisine, both because of the country's low economic status, but also because of low human density, easy access to nature and widespread beliefs about the health benefits of eating wild foods. In a recent quick survey of 7 Lao markets, mammals from as many as 12 families were recorded for sale as bush meat [72].
All the towns in Laos have one or a few markets where both cultivated and wild products are sold. Luang Prabang is one of the largest towns in Laos, with a population of 90,000 people. It is the biggest tourist hotspot of the country. Luang Prabang used to be the capital of Laos until 1975 and hosts many monuments important to the history of the country. As it is located in the center of the city, the Morning Market in the center of the historical part of Luang Prabang is probably the market most visited by foreigners. In spite of this, although some products are tourist-oriented, it mainly serves the local community. Thus, numerous vegetables, fruits, and wild and domesticated animals are sold there each day.
A few other open air markets are located in the city and its peripheries.
Lao markets are worth investigating not only in search of endangered and rare organisms. Lao cuisine is very rich in ingredients [77] and many wild vegetables and fungi are gathered. Lao PDR is undergoing deep cultural changes as it is becomes increasingly involved in the global market economy, and traditional subsistence economy is gradually being replaced by commercial agriculture and the tourist industry, especially in towns. The richness of Lao NTFP products and local traditions of plant use have attracted a lot of research attention in the last three decades, but what is sold in the local markets was never a subject of study. The traditional foods of the Luang Prabang royal court in the mid-twentieth century were documented in a unique cook book written by the king's cook, Phia Sing [77]. A provisional list of edible plants used in Laos was reported by Jaques Vidal in the mid-twentieth century, with one of the main sites of observation being Luang Prabang [78][79][80].
Mushrooms are an important part of Lao cuisine and a commonly exploited NTFP. That is why they are featured in many local rural development studies and some attempts have been made to list the fungi species most commonly sold in Laos [62-65, 81, 82].
Overall, the aim of our study was to make an inventory of wild edible plants and fungi sold in the markets of Luang Prabang, with special reference to the following issues: Documenting traditional foods. Monitoring the presence of any endangered species. The possibility of discovering taxa new to science.

Fieldwork
The most species-rich Morning Market was surveyed regularly in four different seasons (dry season 10th to 19th of February 2018, end of monsoon/beginning of dry season 10th to 18th of November 2019, early monsoon 31st of May to 10th of June 2019 and midmonsoon 31st of July to 7th of August 2019), each time for 8 to 11 consecutive days-38 days altogether. All the stalls were visited and most of them were photographed. Voucher specimens of wild vegetables and fungi were taken. Unstructured interviews about the uses and origins of each species were carried out with the market's sellers, with the help of other co-authors or translators. However, they were not recorded. Apart from the Morning Market, four other markets in Luang Prabang (Phousi, Phanluang, Navieng Kham, Sayxoumxon) and two markets 20 km south of Luang Prabang (north of Xiang Ngeun) were also occasionally monitored and visited at least three times during our research. However, the market with the greatest diversity-the Morning Market-was the main focus. A list of sold taxa was compiled for each season. If possible, plants and fungi were preserved as herbarium specimens and deposited in duplicates: in the herbarium of Warsaw University (WA) and the National Herbarium of Laos (NHL).

Plant and fungi identification
Plants were identified using local field guides and literature available in our institutions and internet resources, taking into consideration recent Lao plant checklists [83,84].
Altogether, 109 specimen vouchers of fungi were analyzed. They were first identified morphologically using the only available guide to the mycota of Laos [82]. The collected voucher specimens were identified with DNA barcoding [85,86] following the guidelines of accepted methods for DNA barcoding of fungi [87]. Fungal DNA was extracted from a small part of the sporocarp (ca. 1 mm 3 of dry mycelium taken from the cap) using a Plant and Fungi DNA Purification Kit (Eurx), following standard protocol. The PCR cocktail consisted of a 4 µl DNA extract, 0.5 µl of each of the primers (ITS5/ITS1f and ITS4 in 10 nmol concentration) and a 5-µl Type-it Microsatellite PCR Kit (Qiagen). PCR was carried out using the following thermocycling conditions: an initial 15 min at 95°C, followed by 35 cycles at 95°C for 30 s, 55°C for 30 s, 72°C for 1 min, and a final cycle of 10 min at 72°C. The PCR products were estimated by running a 5-ml DNA amplicon on 1.5% agarose gel for 30 min. The PCR products were sequenced using ITS4 or ITS5 primers at the Laboratory of Molecular Biology of Adam Mickiewicz University (Poznań) and at the Institute of Biology and Biotechnology of the University of Rzeszów. Obtained sequences were compared with published sequences in UNITE databases using the BLAST tool. A positive identification of a specimen was confirmed if they shared > 97% ITS region sequence identity with the reference sequence. Nuclear ITS sequences obtained in this study have been deposited in GenBank [88] (with the accession numbers listed in Table 3). Nomenclature has been accepted according to the species hypothesis described in UNITE [89].
Plant nomenclature follows the Plant List [90] and fungi names follow Index Fungorum [91].

Wild versus cultivated
It is important to bear in mind that the studied area is a complex agroforestry ecosystem-the gardens have many trees and the numerous species that surround villages often come from spontaneous regeneration; thus, it is very difficult to establish if certain products come from planted or wild specimens. This concerns for example trees growing within villages as well as plants that are both cultivated and collected from the wild or merely tolerated within the agroecosystem, being a part of incipient cultivation (see e.g., [92]). We assume that wild and cultivated plants constitute a continuum. In our study, we decided to include all plants which are at least sometimes collected from spontaneously selfseeded specimens or plants and which are considered wild by the local population even if they are also cultivated. In this, we follow the emic approach to classifying whether a wild plant is wild (for a discussion of this approach, see paper by Sõukand and Kalle [93]). A very similar problem in identifying what is wild in a Southeast Asian market was encountered by the researchers in the markets of Khon Kaen in the Isaan Province of Thailand [15]. They wrote: "Given the extent to which rural ecosystems in Northeast Thailand have been subject to continuing human interference for hundreds of years, it is often difficult to determine if a species is truly wild or not. Wild species are defined as species that normally grow under natural conditions without deliberate human management" [15]. Further they give examples of star fruit (Averrhoa carambola L.) and tamarind (Tamarindus indica L.) often selfpropagating and considered wild or numerous species transplanted to gardens from wild locations to enhance market yields.

Results
We recorded the sales of 110 species of wild plants for food purposes (           Scrophulariaceae. Woody plants (trees, shrubs, and woody vines) constitute exactly half (50%) of the plants sold, and among them 36% are trees (bamboos were not included in this calculation). The largest number of taxa was available in the early monsoon season (June), with a slightly lower number in the mid-monsoon and on the turn of the monsoon and dry season. A much lower choice of plants was available in the dry season (Table 1). However, in each season (observation period), some plants were observed which were not present in other periods.
Molecular investigation and morphological observation revealed a total of 54 fungal taxa from 17 fungi families (Tables 2 and 3; Figs. 3, 4, 5, and 6). Of these, 37 taxa were assigned down to species level and the rest to genus level. Russulaceae was best represented among fungi. The extraction of genetic material failed for some specimens; therefore, they could only be identified morphologically. The most common fungi sold in open air market were russuloid fungi, representing 16 taxa. Within this group, seven taxa were identified to species level and nine to genera. Some differences between obtained sequences were recorded in this group. The phylogenetic analysis of ITS sequences placed these taxa in separate clades. Therefore, 9 unique taxa of unidentified Russula species have been distinguished, each with low similarity to the reference sequence (Table 3). Additionally, the differences between obtained Russula' sequences was higher than 3%, which is the expected level of interspecific variation for fungi within ITS. This allows us to assume that a large number of extremely rare Russula, with no reference sequences represented in databases, or even species unknown to science may be present on sale in the markets.

Discussion
The number of food taxa sold in the studied markets is remarkable on a world scale (see e.g., a list of ethnobotanical market studies in Eurasia in a recent paper about Armenia [94]). We should especially note the long list of 54 fungi species sold, comparable to some of the markets of Mexico (over 90 species sold in 12 local markets [95] and 40 species in another market [96]) and Central Europe, e.g., Poland-32 species in Rzeszów [31], 56 species in Poznań [41] or Hungary, with 38 species in Budapest [38,39]. Such a large number of fungi taxa on sale have not yet been recorded anywhere in Asia outside Laos. Only 6 fungi species have been recorded in the Isaan Province of Thailand, which is culturally very close to Laos [15]. Two studies from Yunnan, China, both found 18 species of fungi on sale [14,32]. In Armenia, 12 species of fungi are sold in the markets of its capital city-Yerevan [94]. Some of the Russula taxa recorded on sale in Luang Prabang may potentially be new species, but, due to the extremely complex taxonomy of the genus, we did not undertake the challenge of describing them. Also, some taxa found in the markets, i.e., Pisolithus orientalis, Polyporus udus, and Calvatia sp. have not been reported as used for consumption in Laos before.
The number of wild food plants-110-is also impressive. For comparison, in Khon Kaen (Bang Lam Phu) located in the Isaan Province, a neighboring region of Thailand, Shirai et al. recorded only half as many species (54) as we found in Luang Prabang [15]. Out of these 54 species 22 were recorded in Luang Prabang as well. In Jinping, Yunnan, China, 35 species of wild food plants were sold in markets [97]; in an area of Assam, India-29 [25]; in the Ukhrul District of Manipur, India [26]-55; and only 28 species of wild vegetables (out of 132 of all the plant taxa in the market) were found in a study of 10 markets in Myanmar [98].
The diversity of wild food plants sold can only be compared to Xishuangbangna in Yunnan, China, where 146 species were recorded in 10 markets [14]; to Armenia, where in Yerevan, the capital of the country, 148 wild food species were recorded on sale [94]; or to Turkey, where 143 wild edible plants were found in Mugla, Bodrum [9].
A large group of species sold are wild vegetables: leaves, shoots, inner stems, or flowers which are ingredients of traditional dishes. They are mainly used to make a dish called soup phak, a gently boiled salad flavored with spices. Sometimes the species are sold in a mix. A previous paper from another part of Laos (Houphan) reports the use of mainly wild vegetable mixes [23], but here in Luang Prabang, most species are sold in bunches of single species. Only small rice field weeds are sold in a mix.
It is worth emphasizing that a large proportion of wild vegetables in the markets of Luang Prabang come from woody taxa. In most countries, agricultural weeds            where-similarly to Lao PDR-young shoots of many species of local trees and shrubs are used for food [99]. Even though we recorded much higher numbers of wild plants and fungi than in any other previous study from Southeast Asia, we suspect that even more species may still occasionally appear due to the extreme diversity of ingredients used by the population of Lao PDR. We hope this is only the beginning of a more detailed surveillance of Lao markets. This also applies to animals, which, as has been pointed out by Greatorex et al. [72], are a potential epidemiological hazard, as proven by the recent coronavirus epidemic [100].
The on going process of modernization of Lao society may bring a decrease in the number of taxa used. In some cases (protected animals), this may be with benefit to nature. In the case of plants and fungi, the taxa for sale are common species originating from rice fields, gardens, and nearby forests, exploited to a level which does not endanger them. Forgetting them may bring large loss to the rich Lao culinary tradition. Fortunately, all the ingredients listed in the Phia Sing's recipe book of the Lao royal court in Luang Prabang (including all the species of fungi) can still be found in markets, which demonstrate Lao cuisine's great resistance to change. We did detect some identification mistakes in Phia Sing's book: the plant listed as mak deed is not Ardisia crispa, but Solanum spirale Roxb., phak tam ling/phak tam nin listed as Melothria heterophylla is actually Coccinia grandis (L.)Voigt.
The large knowledge of forest products in Lao PDR can also serve as a model for tropical organic and permaculture movements, which advocate an increase in the number of food taxa we utilize with a minimal impact on nature [101]. However, this should be done without over-harvesting natural resources. Unfortunately, with the increasing population of Southeast Asia and the culinary popularity of "bush food", there is a danger that many species will become decimated [100]. Fortunately, our study found that it is mainly common weeds and semi-cultivated common tropical trees that are used as food sources, while the danger of overharvesting fungi is very unlikely and usually does not impact mushroom populations [102].
We hope that our study will add to the knowledge of edible fungi in SE Asia and help to distinguish them from the toxic taxa. This especially concerns the genus Amanita. There are many cases of fungi poisoning in Lao PDR. The recently published first atlas of Lao fungi [82] aims to help collectors, but many of the photographed taxa are only identified to the genus level and marked with numbers-this shows the vast need to perform mycotaxonomic and ethnomycological studies in Lao PDR.
The availability of the lists of wild food plants used in particular areas, especially those sold in markets, is very important both on a local and on a global scale. Such research on local food items allows for the recording of traditional products. Their exact taxonomic identification will make further phytochemical and nutritional research easier and enable the charcaterization of local food culture, which if properly advertized, can highly improve the livelihoods of local populations through international marketing and increasing these products' prices. The lists of plants used can also help increase existing local efforts to popularize Lao cuisine among tourists online [103] and via small ethnobotanical market guidebooks [104]. Moreover, it enables the detection of protected species sold in the markets. On a global scale, the inventorying of all wild foods is an important task for developing further strategies for improving the nutrition of the human population and food security [31,94,[105][106][107][108].

Conclusions
The markets of Luang Prabang are very rich in wild edible plants, especially in wild vegetables originating from woody species. The list of fungi sold in them is the longest ever recorded in Asia. The incredible biological diversity we found there has urged us to make similar documentation in other large market towns of Lao PDR.

Authors' contributions
The first author conceived the study and took part in all the field visits and interviews. KP and VL identified the plants. MP and MK identified the fungi. Other authors took part in some of the interviews and manuscript preparation. The authors read and approved the final version of the paper.

Funding
The research was financed by the University of Rzeszów and private funds of the first author, as well as the Institute of Dendrology (Polish Academy of Sciences).
(See figure on previous page.) Fig. 6 The diversity of brittlegills (Russula) sold in the studied markets (voucher numbers are given in brackets). Some of the unidentified specimens may be species unknown to science. a Russula alboareolata (WA0000072251). b Russula sp., (72252). c R. virescens (72253). d Russula sp. (72254). e Russula sp. (72258). f Russula sp. (72268). g R. paludosa (72270). h-k Russula species are usually sold mixed. l Antidesma acidum is only sold alongside Russula spp., as it is a special sour herb used the preparation of dishes made with Russula species