Research Article |
Corresponding author: Daniela Matenaar ( daniela.matenaar@gmail.com ) Academic editor: Martin Husemann
© 2024 Daniela Matenaar.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Matenaar D (2024) Application of extended depth of field 3D imagery to tackle the challenges of cryptic species: a use case in the genus Betiscoides Sjöstedt, 1924 (Orthoptera, Caelifera, Lentulidae) and its taxonomic implications. Evolutionary Systematics 8(1): 65-90. https://doi.org/10.3897/evolsyst.8.117735
|
Discovering and handling cryptic diversity among species challenges taxonomists around the world. This is particularly true for the most diverse animal class – the insects – with cryptic diversity, apart from vast species numbers, being one of the main factors that hamper the description of new species. The biodiversity hotspot Cape Floristic Region of South Africa harbors many endemic and yet undescribed insect species, inter alia, Orthoptera. In this study, extended depth of field and 3D imagery enabled for a novel assessment of the external morphological characteristics used for defining and describing the genetically highly diverse genus Betiscoides Sjösdtedt, 1924, leading to a new definition of the genus’ characteristics as well as a revision of character traits of the known species. Two new species are described and a key to all five recognized Betiscoides species is provided. Application standards are derived to enable replicable and reliable image acquisition and measuring. These findings shall contribute to promote efforts being made to establish image based taxonomic identification for researchers worldwide. High-resolution images provide the basis to train deep learning algorithms/ tools, to detect the smallest differences in highly morphologically alike species, and to implement this knowledge in global species monitoring and conservation action to prevent further species loss.
Image based taxonomy, molecular genetics, divergent evolutionary lineage, deep learning, photogrammetry
Perhaps it is in the nature of things that with an estimate of around 10 million species (
However, up to date one issue remains unsolved: the integration of cryptic diversity (in image based deep learning algorithms), as discrimination among cryptic species challenges taxonomists around the world (
Up to date, the drivers of cryptic diversity are not fully understood and consequently the global distribution of cryptic diversity remains unclear. For instance, it is not proven, that cryptic diversity is greater in tropical than in temperate regions (
In this study, I aim at finding and defining fine scale morphological characteristics and diagnostic characters suitable for distinguishing the DELs of Betiscoides by applying 3D and extended depth of field (EDOF) image based analysis. These findings shall contribute to identify external morphological structures relevant for taxonomic differentiation in invertebrates and, in general, to derive application standards (best practice) to support the development of deep learning algorithms suitable for rare and cryptic species.
Specimens were collected by hand and via sweep netting in the years 2011–2013, 2016 and 2020 (permit numbers AAA007-00374-0035 (28/11/2011), AAA007-00197-0056 (15/03/ 2016, CN44-59-11619 (11/11/2019)) on Jonaskop and on the plateau of Groot Winterhoek Wilderness Area, South Africa (Fig.
Map of the southwestern area of South Africa including the type localities of the Betiscoides species. Star = Table Mountain, type locality of Betiscoides sjostedti Key, 1937. Circle = type locality of B. parva Key, 1937 and the collection site of B. meridionalis specimens,
DNA extraction of specimens HLMD-Cael-4HT, B168; HLMD-Cael-18AT, B166; HLMD-Cael-8PT, B162; HLMD-Cael-10PT, B163; HLMD-Cael-14PT, B165; HLMD-Cael-15PT, B167; HLMD-Cael-364HT, B156; HLMD-Cael-368AT, B155; HLMD-Cael-362PT, B153; HLMD-Cael-367PT, B154, HLMD-Cael-379PT, B169 was conducted in March 2022 using the Quiagen DNAEasy Blood and Tissue Kit. DNA was isolated from the specimens’ hind leg muscle. Further processing (PCR, purification and sequencing) of gDNA took place at Macrogen, Netherlands. Three mitochondrial genes were sequenced: 12S rRNA, NDS (16S rRNA, t-LEU and NADH-Dehydrogenase subunit1), ND5 (NADH-Dehydrogenase subunit 5). Numbers added to the museum collection numbers starting with B, e.g. B168, are running, internally used, numbers of Betiscoides specimens, for further information see Suppl. material
Sequences were inspected and aligned using MEGA X (
External morphological characters were analyzed and measured using Leica M165 C and Keyence VHX 7000 microscope. Morphometrics (in millimetres) taken with Leica M165 C were measured with a measuring eyepiece. Pictures were taken with Keyence VHX 7000 microscope. Plane 2D and 3D Volume measurements were digitally conducted using the Keyence VHX 7000 microscope and its integrated software. Settings were adapted to the specific requirements and are detailed in figure legends. Profile analysis in 3D (eyes, prosternal process, female ovipositor valves) was conducted on 100 or 150× HDR images of the end of the abdomen. Sufficient quality of the images was checked prior the analysis by assessing the “point height”. I attempted to achieve an automated counting of hairy structures on the deep focus images and used the automated counting of extracted shapes tool integrated in the Keyence VHX 7000 software. In order to accomplish that, I took an ×150 3D depth composition image in HDR mode. Hereby, I stopped the image stacking when the area had been fully captured, which I wanted to analyze. This procedure should ensure that interferences were prevented. Then, I used the “auto area measurement” function, the extraction method was set to “Brightness” 113 to 255, and shaping was done by eliminating grains in the size of an area of > 1 mm2. In the last step, “Inversion” was used to achieve that the end of the abdomen was extracted except for the hairy structures framing it. Eye volume measurement was done by manually specifying the Z-axis according to the visible base of the eye. A polygon was drawn along the visible callus line framing the eyes and the volume was automatically calculated. 3D Images and outputs of the profile and volume measurements are exemplarily shown in Fig.
The outputs of 3D eye analyses in Keyence VHX 7000 using the example of Betiscoides specimen HLMD-Cael-379PT, B169. A. 3D image of the eye, ×150; B. The output and measurement of the longitudinal height and length of the eye; C. The manually drawn polygon and result window of the 3D volume measurement of the eye; D. The output and measurement of the cross-sectional height and width of the eye.
Terminology applied for morphological characters is based on
The electronic version of this article in Portable Document Format (PDF) will represent a published work according to the International Commission on Zoological Nomenclature (ICZ), and hence the new name contained in the electronic version is effectively published under that Code from the electronic edition alone. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix http://zoobank.org/. The LSID for this publication is: 6B534FD5-542C-4636-87D1-1738CE33D9F8.
Phylogenetic reconstruction with Bayesian interference showed the same relations as results obtained from Maximum Likelihood analyses. In total, sequences of 123 specimens were included in the phylogenetic reconstruction and 24 DELs were derived confirming the lineage division revealed by
Consensus tree of the Bayesian Inference for genes 12S, NDS, ND5, ITS2 and H3, 2402 bp in total, 10 million generations with sample frequency of 1000. The tree is divided into 24 divergent evolutionary lineages, five of them representing valid species. The respective species are Betiscoides meridionalis, DEL 3, B. sjostedti, DEL 8, B. parva, DEL 9, B. nova sp. nov., DEL 19 and B. muris sp. nov., DEL22. Collection areas are signaled on the right-hand site. Jonkershoek is part of the Hottentots Holland Mountains; modified from Matenaar et al. 2017.
The morphological examination of the collected specimens represented in the phylogenetic reconstruction revealed shared diagnostic characters when compared to the nominal species, confirming their overall allocation to the genus Betiscoides (body slender; head from above conical; pronotum cylindrical; anterior and middle legs short; cerci very small, straight and conical; subgenital plate elongate; female ovipositor valves recurved at the tips, not toothed, (
The genus Betiscoides Sjöstedt, 1924 was described by
Type species: Betiscoides meridionalis Sjöstedt, 1924, type locality: South Africa. Body slender or thin, elongate, stick-like, smooth or hairy, with a prominent ventral line of hairs along the abdomen or at the end of tergits five to eight. Antennae thin or thick ensiform, segments strongly separated, finely and evenly rugose. Head in dorsal view conical. Frontal ridge compressed between antennae. Eye longitudinal or ovate and prominent. Pronotum cylindrical, with very weak or no median carina. Prosternal process trapezoid, T-shape, or lamelliformly compressed, laterally either flat or raised in the anterior or posterior part. Mesosternal interspace reduced, mesosternal lobes connected. Abdomen with fine longitudinal, dorsal median carina. Tympanum absent. Anterior and middle legs short; sometimes hairy, hind femur slender. External apical spine present. Tarsus shorter or half the length of the tibia. Arolium large, sometimes white. Male supra-anal plate triangular in dorsal view. Cerci very small, straight and conical. Subgenital plate elongate and conical. Female ovipositor valves slightly or strongly recurved at the tips; not toothed.
A phylogenetic reconstruction revealed that the sister taxon of the genus Betiscoides is Gymnidium Karsch, 1896 (
Specimens of DEL 3 from Cape Point, Table Mountain NP, are hereby considered as Betiscoides meridionalis as they coincide with its overall morphological diagnostic characters. The large and slender body, the 24-jointed, ensiform, antennae, the very elongate head with the Fv nearly twice the length of the eye as well as the extremely elongate and acutely pointed subgenital plate, reason this allocation. The specimens occur in sympatry with the specimen from DEL 8.
Male. HLMD-Cael-380, LB11, Genbank Acc: KU206332, KU214602, 34°19'2.30"S, 18°25'12.66"E, 42 m above sea level, South Africa, Western Cape, Table Mountain National Park, Cape Point, Restio wetland, November 2013, L. Bröder leg.
Vol: 88.3974 mm3 or 8.83974e-08 m3; Vol. eye: 1002540337.66 µm3, ratio eye length/ fastigium length: 1.30, Subgenital plate degree: 31.6°
Female. HLMD-Cael-383, B24, Genbank Acc: MG243742.1, MG244116.1, MG243841.1, MG243940.1, MG244036.1, 33°57'32.18"S, 18°23'16.06"E, South Africa, Western Cape, Table Mountain National Park, Twelve Apostle, Restio wetland, March 2012, S. Wirtz leg.
Upper valve degree: 52°, upper valve distance = 182 µm, upper valve width = 201 µm, protrusion degree = 129°, Pr line 1 = 146 µm, Pr line 2 = 75 µm, Pr distance = 205 µm, Pr distance to tip of the dorsal valve = 280, 11 µm, Pr vertical distance = 78 µm, Pr straight-lined, apex acute.
Specimens of DEL 9 from the 12 Apostles, Table Mountain NP, are hereby considered as Betiscoides parva as they coincide with its overall morphological diagnostic characters. The small body size, the thick ensiform shape of the antennae and the large, prominent and ovate eyes, as well as the relatively large feet (half the length of the hind tibiae), reason this allocation. In order to assess the hairiness, the specimens of DEL 9 were compared with images of the holotype (male, NHMUK Cape Peninsular, South Africa, December 1930), both showed fairly hairy integument. In order to assess the female genital characters, I analyzed images of the allotype (female, NHMUK, Cape Peninsular, South Africa, December 1930).
Male. HLMD-Cael-381, B103, Genbank Acc: MG243794.1, MG243893.1, MG243991.1, MG244078.1, MG244168.1, 33°57'32.18"S, 18°23'16.06"E, South Africa, Western Cape, Table Mountain National Park, Twelve Apostle, Restio wetland, March 2012, S. Wirtz leg.
Vol: 42.080 mm3 or 4.20806e-08 m3; Vol. eye: 973202045.68 µm3, ratio eye length/ fastigium length: 0.54, Subgenital plate degree: 51° rounded.
The specimen of DEL 8 from Cape Point, Table Mountain NP, is hereby considered as Betiscoides sjostedti as it coincides with its overall morphological diagnostic characters. The intermediate body size, the almost equal length of the antennae and head, the fastigium about 2/3 the length of the eye, as well as the comparatively short subgenital plate, reason this allocation. In order to assess the hairiness, the specimen of DEL 8 was compared with images of the holotype (male, NHMUK, Table Mountain, South Africa, December 1929–1930), both showed fairly smooth integument. This specimen occurs in sympatry with specimens from DEL 3. In order to assess the female genital characters, I analyzed images of the allotype (female, NHMUK, Table Mountain, South Africa, December 1929–1930).
Male. HLMD-Cael-382, B21, Genbank Acc: MG243739.1, MG243838.1, MG243937.1, MG244033.1, MG244113.1, 34°19'2.30"S, 18°25'12.66"E, 42 m above sea level, South Africa, Western Cape, Table Mountain National Park, Cape Point, Restio wetland, March 2012, S. Wirtz leg.
Vol: 58.9269 mm3 or 5.89269e-08 m3; Vol. eye: 773244697.52 µm3, ratio eye length/ fastigium length: 0.78, Subgenital plate degree: 40.8°. The three nominal species are shown in Fig.
The type material is deposited in the invertebrate collection of Hessisches Landesmuseum Darmstadt (HLMD). Holotype. Male, pinned. HLMD-Cael-4HT, B168, Genbank Acc. PP417675, South Africa, Western Cape province, Jonaskop, 33°58'9.75"S, 19°30'17.99"E, 1553 m above sea level, Plot 07, Restio wetland, 13 Dec 2013, leg. L. Bröder, E. Heym & D.Matenaar, Paratypes. Allotype. Female, pinned. HLMD-Cael-18AT, B166, paratypes HLMD-Cael-5-19PT, all same data as holotype. In total, five female and ten male paratypes are designated.
The specific epithet is a Latin adjective (feminine form) meaning ‘extraordinary’ or ‘new’. It refers to the species’ prominent habitus and to the fact that it is a new species characterized with the help of novel techniques.
The described species is assigned to the genus Betiscoides Sjöstedt, 1924 due to its phylogenetic position and overall morphological characteristics. The species is defined by the elongate, stick-like medium to large sized body, ensiform antennae, which are triangular in cross-section; its conical head, which is in profile almost equal in length and height, large eyes, the trapezoid shape of the prosternal process with two kinds of hairs and little variation in general. The end of the abdomen has only few and short hairs. The arolium is very large. The subgenital plate is elongate and conical; the ovipositor is short with robust, recurved valves.
This new species differs morphologically from the described ones mostly through its robustness and higher body volume and the almost identical HL and HH. The body is less elongate than Betiscoides meridionalis. Genital valves are less acute compared to Betiscoides meridionalis. Furthermore, the new species represents a divergent evolutionary lineage (DEL 19) as stated in
Lateral views of female allo- and paratypes of B. nova sp. nov. A. HLMD-Cael-8PT, B162; B. HLMD-Cael-16PT; C. HLMD-Cael-12PT; D. HLMD-Cael-14PT, B165; E. HLMD-Cael-18AT, B166. HLMD-Cael-18AT is the allotype.
Body robust, of medium size (Figs
Holotypes of Betiscoides nova sp. nov. and Betiscoides muris sp. nov. in lateral view. A. Shows B. nova, HLMD-Cael-4HT, B168; B. Shows B. muris, HLMD-Cael-364HT, B156.
Holotypes of Betiscoides nova sp. nov. and Betiscoides muris sp. nov. in dorsal view. A. Shows B. nova, HLMD-Cael-4HT, B168; B. Shows B. muris, HLMD-Cael-364HT, B156.
Larger than the male (Fig.
Image of the prosternal process of the allotype of Betiscoides nova sp. nov., HLMD-Cael-18AT, B166, ×200, captured in ventral view, with a tilt angle of 60°.
Specimens show little differences in the shape of the head in profile, varying from described morphology of the holotype to a slightly more convex form. By contrast, the face differs concerning the distinction of the lateral carinae below the sulcus and the shape of the sulcus itself. In some specimens, the lateral carinae are distinct throughout, also beneath the sulcus. The sulcus is shaped as a mustache sometimes. Median Carina/ Central ridge of the body: Whereas the median carina is present in all specimens, the central ridge of the body is clearly distinct in males on the abdomen, starting on the third segment. This pattern is truly sex specific. Females show a distinct median carina throughout, but in three females (HLMD Cael-18AT, B166; HLMD-Cael-14PT, B165; and HLMD-Cael-16PT) the median carina culminates in a black spot at the verge of the fastigium of vertex. The prosternal process was described as laminate or lamelliform compressed by
3D image profile analysis of the prosternal process of specimen HLMD-Cael-10PT, B163 (B. nova sp. nov.). A. Shows the 3D image turned to profile view and the two points set to measure the distance and height progression curve of the selected part; B. Shows the depth image composition of the 3D image and the two points set to measure the distance; C. Shows the profile analysis results, including the length and height and the curve progression.
Plane and 3D measurements of the morphological characters for males and females of Betiscoides nova sp. nov. and B. muris sp. nov.
Measurements | B. nova males | B. nova females | B. muris males | B. muris females |
---|---|---|---|---|
Body length mm | 22.5–27.2 | 29.3–30.02 | 21.2–24.3 | 26.15–29.5 |
Pronotum length mm | 3.0–3.3 | 3.5–3–7 | 2.2–2.65 | 2.85–3.4 |
Hind femur length mm | 8.7–10 | 9–11 | 6.9–8.1 | 8.9–10.4 |
Antenna segment no. | 21 | 21 | 21–22 | 22–23 |
Head length mm | 3.6–4.5 | 4.2–5 | 3.5–4.3 | 4.3–4.7 |
Head diagonal length mm | 5.51–6.32 | 6.48–7.02 | 4.73–5.52 | 5.65–6.62 |
Head height mm | 3.26–3.77 | 3.69–4.19 | 2.14–2.59 | 2.85–3.2 |
Subgenital plate length mm | 2.3–3.0 | 2.2–3.0 | ||
Subgenital plate degree | 46.3–58.3 | 30.7–35.8 | ||
3D measurements of the eyes | ||||
Longitudinal height µm | 482.72–709.17 | 458.61–760.13 | ||
Cross-sectional height µm | 532.59–657.29 | 598.04–646.17 | 460.35–577.58 | 422.86–696.96 |
Length µm | 2254.37–2462.66 | 2500.83–2737.3 | 1943.74–2137.79 | 2131.68–2469.35 |
Width µm | 1321.6–1454.8 | 1412.78–1524.26 | 1055.54–1213.7 | 1162.68–1344.2 |
Volume µm3 | 1071906358.87–1812003789.7 | 1566466980.82–1798593679.03 | 587729028.76–865630822.96 | 894156739.30–1428871850.17 |
upper valve degree | 146–163 | 134–140 | ||
upper valve distance µm | 165–240 | 115–208 | ||
upper valve width µm | 160–311 | 106–180 | ||
degree of protrusion | 106–135 | 57–100 | ||
protrusion line 1 µm | 171–235 | 118–158 | ||
protrusion line 2 µm | 104–137 | 86–118 | ||
protrusion distance µm | 236–327 | 174–199 | ||
protrusion vertical distance µm | 161.57–180.92 | 126.31–154.28 |
The longitudinal height cannot be safely derived from the measurements since most of specimens eyes have been subject to certain degree of decay, which caused the highest part of the eyes to collapse at the highest point. Specimens HLMD-Cael- 6, 8, 9, 13, and 15PT were excluded from the cross sectional height analysis since the eyes had collapsed.
Described from 11 males and five females from Jonaskop, Western Cape, South Africa, 2013. The species occur on top of Jonaskop in restio-dominated wetland. Abundant in patches with water covering the bare ground between the single restio plants. As the three formally recognized species of Betiscoides are considered endangered (
The type material is deposited in the invertebrate collection of Hessisches Landesmuseum Darmstadt (HLMD). Holotype. Male, pinned, HLMD-Cael-364HT, B156, (Genbank Acc. PP411598, PP417668), South Africa, Western Cape province, Groot Winterhoek Wilderness Area, 32°59'21.52"S, 19°3'26.84"E, 991 m above sea level, Plot 134, Restio wetland, 18 April 2016, leg. D. Matenaar.
Paratypes. Allotype. Female, pinned. HLMD-Cael-368AT, B155, South Africa, Western Cape, Groot Winterhoek Wilderness Area, 32°59'53.06"S, 19°4'12.40"E, 929 m above sea level, Plot 133, Restio wetland, 18 April 2016, D. Matenaar. Paratypes. HLMD-Cael-362PT, B153; HLMD-Cael-366PT, South Africa, Western Cape, Groot Winterhoek Wilderness Area 32°59'35.34"S, 19°3'32.90"E, 969 m above sea level, Plot 65, Restio wetland, 18 April 2016, D. Matenaar. HLMD-Cael-363PT; HLMD-Cael-367PT, B154, South Africa, Western Cape, Groot Winterhoek Wilderness Area, 32°58'54.76"S, 19°3'18.37"E, 1001 m above sea level Plot: 66, Restio wetland, 18 April 2016, D. Matenaar. HLMD-Cael-365PT South Africa, Western Cape, Groot Winterhoek Wilderness Area, 32°59'53.06"S, 19°4'12.40"E, 929 m above sea level, Plot 133, Restio wetland, 18 April 2016, D. Matenaar. HLMD-Cael-360PT, HLMD-Cael-361PT; HLMD-Cael-369PT, South Africa, Western Cape, Groot Winterhoek Wilderness Area, 134: 32°59'21.52"S, 19°3'26.84"E, 991 m above sea level, Plot 134, Restio wetland, 18 April 2016, D. Matenaar. HLMD-Cael-379PT, B169, South Africa, Western Cape, Groot Winterhoek Wilderness Area, Restio wetland 33°0'11.81"S, 19°4'20.91"E, 908 m above sea level, Plot 95, Restio wetland, 06 March 2020, D. Matenaar. In total, five female and five male paratypes are designated.
The specific epithet is a Latin noun, meaning mouse. It refers to the species’ very slender and delicate habitus.
The described species is assigned to the genus Betiscoides Sjöstedt, 1924 due to phylogenetic and overall morphological characteristics. The very slender, elongate, stick-like and medium sized body defines the species. The antennae are ensiform and triangular in cross-section. The head is acutely conical elongated and the end of the abdomen shows dense and long hairs. The arolium is large. The subgenital plate is elongate and acutely conical; the ovipositor is short with well-curved valves and sharp tips.
This new species differs morphologically from the described ones mostly by its overall delicate habitus, the comparatively densly and longhaired end of the abdomen of the males. The body is smaller and more delicate than Betiscoides meridionalis. The head is more elongate than in B. nova and B. parva. The genital valves are strongly acute, sharply pointed, upper valve width narrower than in Betiscoides meridionalis. Just as B. nova, the new species represents a divergent evolutionary lineage as stated in
Images of the end of the abdomen of specimens A. HLMD-Cael-364HT, B156 (Betiscoides. muris sp. nov.); B. HLMD-Cael-4HT, B168 (B. nova sp. nov.); C. B. parva (HLMD-Cael-381, B103) and D. B. sjostedti (HLMD-Cael-382, B21).
Body of medium length, slender, delicate (Figs
Larger than the male (Fig.
High resolution DOF images of the prosternal process of Betiscoides muris sp. nov. clustering in DEL 22, Groot Winterhoek, revealing four different shapes of Pp. A. Image of HLMD-Cael-379PT, B169; B. Image of HLMD-Cael-366PT; C. Image of allotpye HLMD-Cael-368AT, B155; D. Image of HLMD-Cael-367PT, B154.
The antennae are 22–23-jointed and the segments are covered with fine, black dots and shiny appearance in all specimens. Posterior margins of the antennae segments of one male specimen are matt beige, while the rest is of the usual common shiny brown color. Specimens show little differences in the shape of the head in profile, varying from described morphology of the holotype to a slightly more concave form. Similar to B. nova, the face varies concerning the degree of distinction of the lateral carinae below the sulcus and the shape of the sulcus itself. In some specimens, the lateral carinae are distinct throughout, also beneath the sulcus. The sulcus is shaped as a mustache sometimes. This intraspecific variation is not sex-specific. Variation in spines on the hind tibia: In male, the hind tibiae had 9–10 outer and 10–14 (!) inner spines. In females, the hind tibia showed 9–10 outer spines and 11–13 inner spines. In contrast to the other known species of Betiscoides, there seems to be a intraspecific variability in the shape of the prosternal process, in general the Pp is about twice as long as broad, its margins and angles rounded; the anterior end little broader than the posterior. However, four different shapes can be distinguished within the B. muris species, making the Pp itself a variable character within this species (see Fig.
3D image of the prosternal process of B. muris sp. nov. specimen HLMD-Cael-379 PT, B169. The output of the 3D profile image analysis is shown, including the curve progression from the anterior to the posterior part in profile, and total length and height.
Male individual of Betiscoides muris sp. nov. from Groot Winterhoek Wilderness Area, South Africa in life; image taken on 06 March 2020, 32°58'54.76"S, 19°3'18.37"E.
Specimen HLMD-Cael-379PT, B169 is not included in the subgenital plate degree measurement due to damage.
Described from six males and five females from Groot Winterhoek, Western Cape, South Africa, collected in 2016 and 2020. The species occurs on the plateau of the Groot Winterhoek Nature reserve in restio-dominated wetland. It is abundant in patches with water covering the bare ground between the single restio plants. As the three formally recognized species of Betiscoides are considered “endangered” (
1 | Body elongate, head in profile conical, fastigium of vertex shorter than twice the length of eye | 2 |
– | Body very elongate, head in profile acutely conical, fastigium of vertex nearly twice the length of eye | B. meridionalis |
2 | Body smooth | 3 |
– | Body finely but densely haired | 4 |
3 | Body slender, head conical, eyes ovate, male subgenital plate comparatively short, apex pointed in profile | B. sjostedti |
– | Body robust, head in profile as long as high, eyes large and ovate, male subgenital plate elongate, apex rounded in profile | B. nova |
4 | Body short, head conical but less elongate, eyes ovate and very prominent, tarsus half the length of the hind tibia | B. parva |
– | Body of medium length, head conical and elongate, its diagonal length twice its height, eyes ovate but not protruding, tarsus shorter than half the length of the tibia | B. muris |
An overview of the main morphological differences between the species is provided in Table
Description of the shape of the morphological characters with the main differences of the five Betiscoides species.
Morphological character | B. meridionalis | B. parva | B. sjostedti | B. nova sp. nov. | B. muris sp. nov. |
---|---|---|---|---|---|
Body | very elongate, very slender | slender, short | slender, medium length | robust, medium to large | delicate, slender medium length |
Head | acutely conical, very elongate | conical, less elongate | conical, elongate | conical, in profile as long as high | conical, very elongate |
Eyes | ovate, flat | ovate, large prominent | ovate | ovate, large, less prominent than in B. parva | ovate, not protruding |
Prosternal process | strongly lamelliform, compressed | slightly lamelliform, anterior end broader and higher | slightly lamelliform, anterior end a little broader | trapezoid, flat, two kinds of hairs | very variable, one kind of hair |
Subgenital plate | extremely elongate and acutely pointed | elongate, hairy, apex rounded | comparatively short, smooth, pointed | elongate, smooth, in profile: apex rounded | very elongate and acutely pointed, hairy |
Female ovipositor valves | recurved, apex acute, Pr apex acute | short, recurved, apex rounded | long, recurved, apex acute | recurved, apex rounded, Pr and Pr apex rounded | recurved, apex and protrusion strongly acute, Pr apex rounded |
Revealing morphological traits among cryptic species or divergent evolutionary lineages is a challenging but important task to firstly, obtain an overview of the global biodiversity and its distribution and secondly, to provide for the basis of species conservation actions. The application of deep learning and image based identification tools increased strongly in the last decade (
Head
The ratio of the eye length to the fastigium length were diagnostic characters used for discriminating B. meridionalis, B. parva and B. sjostedti and was provided in addition to the description by
Hair – quantitative analysis
The DELs of Betiscoides differ, inter alia, by the amount of hairs at the end of the male abdomen. In this study, I was able to show that B. parva is overall very hairy which led to the redefinition of the genus characteristics. Differences in this character state are particularly evident in sympatrically distributed lineages. For instance, DEL 11 and DEL 16 that occur sympatrically in Limietberg/ Du Toitskloof, show differences in the amount and length of the hairs at the end of the abdomen (see Fig.
End of abdomen of Betiscoides sp. specimens B134 from DEL 16 (A) and B128 from DEL 11 (B), collected at Limietberg, Du Toitskloof, study site 34.
Female genital characters
Detailed measurements of the external genital characters of the females resulted in consideration of additional character states, which can be used to distinguish among species. There was no overlap between the measured values except for “upper valve width” and “protrusion line 2”. However, the combination of the female genital characters serves as a solid base to distinguish between the species. “Vertical distance of protrusion” was also an efficient measure to distinguish between the species, despite some intraspecific variations. However, images need to be of very good quality and need to be controlled prior the subsequent image analysis. Specimen HLMD-Cael-12PT either showed high variability or was in a different state of decay, either way; the protrusion itself is folded towards the body center, which could only be recognized in 3D mode. Thus, measuring from the tip of the protrusion to the tip of the valve resulted in a lesser vertical distance. In such cases, I advise to set the point on the tip of protrusion at a slightly higher point to ensure the actual level of the protrusion is met.
Profile analysis of prosternal process
The integrative taxonomic approach revealed that the Pp is a plastic character within B. muris, which can be classified in four different states some of which are described for Betiscoides for the first time. Results of the phylogenetic reconstruction made clear that these states are not diagnostic to delimitate species. The 3D analysis of the specimens’ Pps revealed that the first impression of a proximal concave Pp turned out to be false. Only by turning to the side of the process and the profile analysis, it became clear that the surface was not sunken in at all but there is a slight gradient towards the posterior end of the Pp. This procedure worked also for even finer structures and lines, which seemed to be small sulci in 2D view but were in fact not existent when analyzed in profile. Unfortunately, in order to conduct the profile analysis the Pp needs to be recorded in straight top ventral view as different angles falsify the results. However, sufficient Pps could be analyzed and revealed no variation regarding the curve progression of the Pps in B. nova and could precisely show that this diagnostic character is highly variable in B. muris. The gradient of the Pp towards either the anterior or posterior edge is a diagnostic character for B. parva and B. sjostedti (
Volume – Body and eyes
Using fully calibrated and referenced EDOF based 3D models of insects is rather novel in species descriptions (
3D measurement of the eyes revealed that the eyes’ volume is sexually dimorphic within B. muris but not in B. nova. It remains challenging to calculate the volume and other 3D eye characters because of the degree in decay of dried specimens. However, the advantage is that it is easy to evaluate the specimens’ state and to decide to not consider it if conditions appear inappropriate. Species like Betiscoides, which feature very prominent eyes and are regarded as morphologically cryptic, benefit from integrating this new evaluation of character states into future descriptions as they possibly very often will turn out to be diagnostic. In eyes of B. muris, the longitudinal height correlated negatively with length, meaning the shorter the eye was, the higher was the longitudinal height. Regarding the body volume, it interesting for species like Betiscoides sp. to compare the body volume with the equivalent piece of blade of the respective host restio plant to look for a certain morphological adaptation or mimicry.
The 3D measurement approach applied provided evidence for the existence of morphological differences between the cryptic divergent evolutionary lineages of Betiscoides and further collecting and analyses is needed to resolve the taxonomy of this highly diverse group. The integrative taxonomic approach allows to conclude that all DELs can be assigned to the genus Betiscoides, due to the the phylogeny and overall morphology.
The application of standardized, calibrated and high-resolution digitization is a novel technique to enable the detection of fine scale morphological differences, which requires access to specific equipment and technology. At first, this seems to be a barrier to achieve an effective and repetitive procedure in taxonomic work. However, on the long term exactly these procedures shall enable an even more rapid identification as deep learning [AI] tools become more common. Applying 3D image technology enables you to describe what you actually see and to put this into numbers and objective relations. You can measure what you really see, or you can finally see how it really is! Furthermore, this is a noninvasive approach, which does not damage or even destroy the specimens and might even work on living individuals in the field. Thus, it is a worthwhile investment to achieve simpler, comparable and reliable identification in the future.
Similar to other digital morphometric analyses and attached imaging techniques, one advantage in photogrammetry is that measurements are often very exact since the high resolution makes it easy to set measuring points accurate and, if measuring lines are inserted in the image, the methods of measuring a certain distance remains traceable. Especially on dried and pinned material, photogrammetric methods are often the more suitable ones compared to micro-tomography as the needle likely causes artifacts in the latter and inner organic structures have often decomposed anyway (
As important as imaging analysis and deep learning are, it still requires (human) taxonomic expertise to avoid misinterpretation of plastic characters, especially in almost cryptic species, or physical changes during life span. As also suggested before (
In order to gain robust und replicable photogrammetric measurements, certain standards should be considered. If applicable, the specimen should be digitally processed as soon as possible after killing/ preparation. This avoids misleading measurements due to decay, which of course is especially important if referring to fine scale traits. Specimens should be prepared or dissected as carefully as possible to conserve possible diagnostic characters and assure that these remain visible. When taking the images bound for volume and further 3D measurements, magnification to full screen or stitching is necessary to achieve reproducible results. Images should be taken in top view, meaning the image degree should be 0° and the specimen should be placed in dorsal, ventral, or lateral position; and checked for the point height prior to further analysis to gain replicable and reliable results. When conducting 2D plane measurements, I advise to zoom into or very close to the area where the point marks will be set and to describe in detail where the ruler was placed.
I wish to thank SANParks, CapeNature, and Eastern Cape Park and Tourism Agency for providing the permits and for their continuous support over the last 13 years. I thank Trier University and Stuttgart Statemuseum of Natural History for their financial support during the respective collecting trips. I also thank Nico Blüthgen and Chris von Beeren for providing access to the molecular lab of Ecological networks (TU Darmstadt). Special thanks to Antonia Späth for building the 3D models and to Jörn Köhler for the valuable comments on the manuscript. I would also like to thank the reviewers for their constructive feedback and time.
Specimens collections sites and Genbank Acc. numbers
Data type: pdf
Explanation note: Information on the collection locality (localitly, site no., coordinates) and Genbank Accession numbers of genes 12S rRNA, 16S rRNA + tRNA-Leu + NADH dehydrogenase subunit I (ND1) (NDS), NADH dehydrogenase subunit 5 (ND5), internal transcribed spacer 2 (ITS2) and Histone 3 (H3) of the sequenced Betiscoides specimens.