What are the lacewings, and why are they interesting?

We live in a world which is lacewing-blind. Most people would not be able to identify a flying insect as a lacewing, let alone distinguish between species (Figures 1 and 2). But almost everyone has encountered an artefact created by the larvae of lacewings. They recognize the distinctive funnel-shaped pits in sandy areas (Figure 3), and they have been told that there is a beast called an antlion lying in wait below to consume any insect that slips down the side of the funnel. But, few people grasp that the antlion is to the lacewing what the caterpillar is to the butterfly. They are blind to the existence, and value, of lacewings, the adults of the creatures that live in the sand.

13 of the 16 recognised families of Neuroptera occur in southern Africa, and both families of Megaloptera. This report focuses mainly on the Neuroptera, popularly known as lacewings. The Afrotropics (i.e. Africa south of the Sahara Desert) has an especially rich and varied fauna of lacewings and approximately 500 species occur in southern Africa alone, defined as the region south of the Kunene and Zambezi Rivers (Mansell 2002). Furthermore, about half of these are endemic to this area.

Neuroptera are excellent indicators of environmental and habitat transformation, and include key species for signifying areas and faunas that require priority protection. They are vulnerable to habitat fragmentation and pesticide contamination (Mansell 2002, Winterton et al. 2010).

The larvae of the lacewings are all specialised predators with unique, highly evolved mouthparts. As predators, lacewing larvae have the potential to have a major impact upon populations of other insects and small Arthropoda, and especially aphids. They have therefore, long been considered an attractive option as biological control agents in greenhouses, orchards and fields (New 1975, Mansell 2002). The recommendation is to augment species native to an area by means of mass rearing, and not to introduce new lacewing species (New 1985).

Only one of the families, the Myrmeleontidae, includes species whose larval stage consists of antlions that construct funnel-shaped pits in sand (Figure 3). The larvae of the other families take on a diverse variety of forms; they range from aquatic to semi-aquatic, and there are species with larvae which live freely in sand, under rock ledges, small caves, holes in trees, and as free-living ambush predators on vegetation. Some are parasites in spider nests, and inquilines in ant nests. Nothing is known about the larvae of some species (Mansell 2002, Winterton et al. 2010).

The Neuroptera are model subjects for scientific research because they have a wide diversity of lifestyles. Adults of several families are key pollinators of indigenous flora; especially the family Nemopteridae (the thread-wing and the spoon-and ribbon-wing lacewings) (Mansell 2002).

What is the objective of LacewingMAP?

Given that the lacewings are important, the long-term objective of the LacewingMAP project is to develop an atlas of the distributions of the Neuroptera and Megaloptera in Africa, focusing initially on southern Africa, then the Afrotropics, and ultimately the African continent. The project is loosely modelled on the “butterfly atlas” and the “reptile atlas” (Mecenero et al. 2013, Bates et al.2014). For both those projects, the foundational data were the historical specimen record data, supplemented by photographic data uploaded to the “Virtual Museum” by citizen scientists. The Virtual Museum is described by Mecenero et al. (2013) and Bates et al. (2014). The lacewing atlas uses the same strategy. Specimen records were (and continue to be) assembled by us, photographic records are collected by citizen scientists, and the combined database is curated by the Virtual Museum.

This report reviews progress up to March 2019. The first image of a lacewing was uploaded to the LacewingMAP section of the Virtual Museum on 19 September 2014. This report is based on the specimen database, plus photographic records assembled over four and a half years, up to 31 March 2018.

What is the volume of records in the LacewingMAP database?

The total number of records in the LacewingMAP database on 31 March 2019 was 15,781 (Table 1). They are split into two components in this database, seamlessly merged as a single entity. The largest component consists of 12,898 records, mainly based on museum specimens, assembled by us, and recorded in a Palpares Relational Database (Mansell & Kenyon 2002). This is supplemented by 2,883 photographic records, submitted to the LacewingMAP section of the Virtual Museum (http://vmus.adu.org.za) by citizen scientists (Table 1). Each photographic record uploaded to the Virtual Museum contains either one, two or three images of the live animal; each record is evaluated by us, and we allocate it to family, genus or species.

The majority of the 2,883 photographic records, uploaded to the Virtual Museum were submitted from South Africa (2,225, 77%) (Table 1). A total of 658 records were submitted from 20 other African countries; six countries had more than 40 records: Malawi (170), Botswana (100), Namibia (90), Swaziland (87), Zambia (87) and Mozambique (43) (Table 1).

In the overall database, 10,917 records are from South Africa (Table 1). Countries with totals more than 500 records are Namibia (1,020), Democratic Republic of Congo (708) and Zimbabwe (644) (Table 1). 50% of Malawi’s 342 records are photographic, as are 44% of Swaziland’s 197 records, and 30% of Mozambique’s 144 (Table 1).

Within the nine provinces of South Africa, the largest contributions of photographic records have come from Northern Cape (484, 21.7% of total of 2,222 for South Africa), Limpopo (456, 20.5%) and KwaZulu-Natal (420, 18.5%) (Table 2). Within the database as a whole, Limpopo has the most records (2,606, 24.6% of 10,594 records for South Africa) and the Northern Cape has 1,688 (15.9%) (Table 2). Three of the photographic records and 323 of the total records from South Africa did not have “province” assigned (Tables 1 and 2).

The average rate of submission of photographic records for LacewingMAP for the four years 2015 to 2018 was 566 per year (Table 3). This rate can be compared with the annual collection rate for the specimen section of the database (Table 4). The photographic rate generated by citizen scientists is 64% above the “best” decade (the 1980s), 5.5 times more than the 20th century as a whole (102 per year), and three times more than the second half of the 20th century (176 per year) (Table 4).

The monthly pattern of submissions shows a minimum in the winter months from May to August, and a peak in the summer months from December to April (Figure 4). This plot confirms the general pattern of seasonality of conspicuous occurrence of lacewings.

Each record is georeferenced as accurately as feasible. For mapping purposes each record is allocated to a quarter degree grid cell. This 15-minute grid system has been widely used by biodiversity atlas projects in southern Africa (e.g. Mecenero et al. 2013, Bates et al. 2014). The 15-minute (quarter degree) grid generates 2025 quarter degree grid cells in South Africa, Lesotho and Swaziland. Of these, 835 grid cells (41.2%) have at least one species of lacewing recorded (Figure 5). 230 grid cells have a single species recorded in them. On the other hand, there are only two degree cells with no records at all, one in the Northern Cape and one in North West Province. At this stage, the patterns of species richness still reflect observer effort rather than the true distribution of species richness.

What species are in the LacewingMAP database?

The taxonomy upon which LacewingMAP is based contained 1,249 species in March 2019 (Table 5); this taxonomic spine, which is pivotal for the project, is updated from time to time, as necessary. This taxonomy is of Afrotropical species; 18 of these species are from the order Megaloptera (two families Corydalidae and Sialidae), and the remaining 1,231 species are Neuroptera, classified into 13 families (Table 5). By far, the largest family is the Myrmeleontidae, containing 461 species. 415 species of Neuroptera are currently known from South Africa (Mansell & Oswald 2018), and 834 from the remainder of the Afrotropical Region, i.e. species that do not occur in South Africa.

Of the 1,249 species in the taxonomy, the overall LacewingMAP database (specimens and photographs) contained records for 952 on 31 March 2019. 20 species had 148 or more records, of which 18 were members of the family Myrmeleontidae (Table 6). The two species with the most records were Myrmeleon obscurus (518) and Hagenomyia tristis (516) (Figures 1 and 2). The distribution maps for these two species within South Africa, Lesotho and Swaziland (Figures 6 and 7) show distinctly different patterns: it seems probable that Myrmeleon obscurus occurs throughout South Africa (Figure 6), but that Hagenomyia tristis is confined to the eastern half of the country (Figure 7).

All 12,898 records in the specimen database are identified to species. Species level identification from photographs is not always possible because many lacewings, and especially the species of “green lacewings” of the family Chrysopidae, can only be identified by dissection.

By 31 March 2019, we had undertaken identifications of 2,853 of the 2,883 photographic records submitted by citizen scientists. This provides a large sample of records from which we can attempt to quantify the extent of the identification issues. 1,885 of the 2,853 records (66.1%) were identified to species level, 613 (21.5%) to genus level only, and 355 (12.4%) to family level only (Table 5). Of those identified to family level only, 217 records (61%) were Chrysopidae (green lacewings), 50 records (14%) were Myrmeleontidae (antlions) and 49 records (14%) were Mantispidae (mantidflies) (Table 5).

Of the 613 records identified to genus level only (Table 5), 348 belonged to five genera: 105 in the genus Chrysoperla in the family Chrysopidae, and 83, 79, 67, and 55 in the genera Centroclisis, Cueta, Myrmeleon and Creoleon, respectively, of the family Myrmeleontidae (antlions). In summary, the green lacewings, i.e. the family Chrysopidae and especially the genus Chrysoperla within this family, and four genera within the family Myrmeleontidae (antlions) present the largest identification challenges from photographs.

In the photographic database, of the 22 species with more than 20 records (Table 7), 15 are also in Table 6, the top 20 species overall. There is one species in Table 7 for which more than half of all records are photographic: Dichochrysa tacta (recently renamed Pseudomallada tactus) has 43 photographic records and 41 specimen records. The distribution map (Figure 8) demonstrates how the photographic records are helping to “fill in” the range suggested by the specimen records.

The genus Dichochrysa (now Pseudomallada) is part of the family Chrysopidae, the green lacewings, for which identifications are generally difficult. However, along with the genus Italochrysa, most photographic records for both genera were identified to species (88% and 86%, respectively) (LacewingMAP database).

Who are the main contributors of photographic records to the LacewingMAP database?

By March 2019, 234 people had submitted records to LacewingMAP; 36 had submitted more than 20 records (Table 8). It is true to state that none of these 36 people have a primary interest in the lacewings (in the way that people have primary interests in a particular taxon, such as birds, butterflies, reptiles, dragonflies and damselflies, or even spiders or scorpions). 90 people had submitted a single record, and the median number of submissions per observer was three. The Virtual Museum had a total of 2,256 observers on 31 March 2019. Only eight of the 234 participants in LacewingMAP had submitted records only to this section of the Virtual Museum (seven had submitted one record, and one person had submitted 12, the only specialist LacewingMAPper). For 98.8% of the 2,256 Virtual Museum participants, submissions to LacewingMAP were less than 10% of their total numbers of records submitted. These observations suggest that photographic records are submitted to LacewingMAP opportunistically, as they are encountered. The lacewings are an extremely valuable by-catch.

What are some of the interesting photographic records in LacewingMAP?

LacewingMAP has contributed many interesting and valuable locality records. It has added a vast number of new locality records and has contributed to our overall knowledge of the distribution of Afrotropical lacewings. Thus it is difficult to single out individual records.

Two records, both from 2018, are outstanding. They highlight the value of the contribution being made by citizen scientists.

LacewingMAP record 15379 is a specimen from Lüderitz Peninsula, southwestern Namibia, on 24 July 2018 (Figure 9). It belongs to the the genus Palmipenna. It is doubtless an undescribed species, remarkable for its early appearance (July) and its close proximity to the sea. This record was a total surprise. It is the farthest north that this genus has ever been recorded, and the second record of this genus from Namibia. Previous records of this genus were almost exclusively from the Western Cape, South Africa.

LacewingMAP record 10583 is a specimen of a new antlion (Myrmeleontidae), either in the genus Fadrina or the genus Centroclisis (Figure 10). It cannot be placed with certainty; it has characteristics of both, and also remarkable for its small size. Provisionally, it is placed in Fadrina because of the double costal series in the forewings. This lacewing was found in the Cederberg area on 22 January 2018. This photographic record alerts us to the existence of a previously unknown taxon. It also emphasizes the exceptional lacewing diversity of the Cederberg.

What are the priorities for fieldwork for LacewingMAP?

The answer to this is simple. At this stage in the life-cycle of the LacewingMAP project every record, from anywhere in Africa, is valuable.

How do I participate in LacewingMAP?

In a nutshell, the protocol is simple. Take photographs of lacewings, and upload them to the LacewingMAP section of the Virtual Museum website. There is no need to identify the species in the photograph. This gets done by the expert panel for LacewingMAP.

The easiest way to take photographs of lacewings is to be aware that they are attracted to light at night, in exactly the same way that moths are, although usually in far smaller numbers. The entire spectrum of cameras are used to take photographs of lacewings; the most versatile for this type of photography are the new generation of “compact” cameras

Before you can upload into the Virtual Museum you need to register as a citizen scientist. The procedure for doing this is described here: (https://www.slideshare.net/Animal_Demography_Unit/how-to-register-as-a-citizen-scientist-with-the-animal-demography-unit)

Once you are registered, you log on to the website using your email address and password. A “Data upload” section now becomes visible. The critical information that needs to be uploaded into the database is date, place and a series of one to three photographs of a single species, usually different angles on the same individual. Guidance on the upload process is provided in this slide show: https://www.slideshare.net/Animal_Demography_Unit/how-to-submit-records-to-the-virtual-museums

We do our best to identify each record to species level. As described earlier, this is difficult to achieve for several of the lacewing families, and especially for the green lacewings. But this should not deter you from submitting photographs. As a beginner participant, the best strategy for a positive confirmed identification is to take lots of photos of a specimen, and to select the best one, two or three photographs for submission, preferably from different angles. It is helpful to try to get different parts of the specimen in sharp focus in the three pictures.

Acknowledgements

We thank all the contributors to the LacewingMAP project for their photographs, and all those who collected specimens over the years, upon which the original dataset is based. We gratefully acknowledge the South African Biodiversity Information Facility (SABIF), the Global Biodiversity Information Facility (GBIF) and especially the JRS Biodiversity Foundation, Seattle, USA, for supporting the databasing of Afrotropical lacewings, which underpins this project. Museum specimen records were (and continue to be) assembled by MM. The expert panel for LacewingMAP is lead by MM, who evaluates all photographic submissions and attempts to assign records to species level.

References

Bates MF, Branch WR, Bauer AM, Burger M, Marais J, Alexander GJ, de Villiers MS (eds) 2014. Atlas and Red List of the reptiles of South Africa, Lesotho and Swaziland. Suricata 1. Pretoria: South African National Biodiversity Institute.

Erasmus BFN, Kshatriya M, Mansell MW, Chown SL, Van Jaarsveld AS 2000. A modelling approach to antlion (Neuroptera: Myrmeleontidae) distribution patterns. African Entomology 8: 157-168.

Freitag S, Mansell M 1997. The distribution and protection status of selected antlion species (Neuroptera: Myrmeleontidae) in South Africa. African Entomology 5: 205-216.

Mansell MW 2002. Monitoring lacewings (Insecta: Neuroptera) in southern Africa. Acta Zoologica Academiae Scientiarum Hungaricae 48 (Suppl. 2): 165-173.

Mansell MW, Kenyon B 2002. The Palpares relational database: an integrated model for lacewing research. Acta Zoologica Academiae Scientiarum Hungaricae 48 (Suppl. 2): 185-195.

Mansell MW, Oswald JD 2018. Neuropterida of South Africa. Available online at http://lacewing.tamu.edu/Faunas/SouthAfrica.

Mecenero S, Ball JD, Edge DA, Hamer ML, Henning GA, Krüger M, Pringle EL, Terblanche RF, Williams MC (eds) 2013. Conservation assessment of butterflies of South Africa, Lesotho and Swaziland: Red List and atlas. Johannesburg: Saftronics and Cape Town: Animal Demography Unit.

New TR 1975. The biology of Chrysopidae and Hemerobiidae (Neuroptera), with reference to their usage as biocontrol agents: a review. Ecological Entomology 127: 115-140.

Winterton SL, Hardy NB, Wiegmann BM 2010. On wings of lace: phylogeny and Bayesian divergence time estimates of Neuropterida (Insecta) based on morphological and molecular data. Systematic Entomology 35: 349-378.