Research Article Journal of Orthoptera Research 2021, 30(1): 73-80 Aspects of the life history and ecology of two wingless grasshoppers, Eremidium armstrongi and Eremidium browni (Lentulidae), at the Doreen Clark Nature Reserve, KwaZulu-Natal, South Africa RESHMEE BRIULAL!2, AKEEL RAJAK!, ADRIAN J. ARMSTRONG!? 1 Ezemvelo KZN Wildlife, P.O. Box 13053, Cascades, 3203, South Africa. 2 8 Deltagrove Place, Grove-End Drive, Phoenix, 4068, South Africa. 3 Centre for Functional Biodiversity, School of Life Sciences, University of KwaZulu-Natal, Private Bag X01, Scottsville, Pietermaritzburg, 3209, South Africa. Corresponding author: Reshmee Brijlal (Reshmeebrijlal21 @gmail.com) Academic editor: Maria-Marta Cigliano | Received 30 September 2020 | Accepted 1 December 2020 | Published 12 May 2021 http://zoobank.org/D02304CA-0BD4-422F-BBDA-1EF4B3CD70B7 Citation: Brijlal R, Rajak A, Armstrong AJ (2021) Aspects of the life history and ecology of two wingless grasshoppers, Eremidium armstrongi and Eremidium browni (Lentulidae), at the Doreen Clark Nature Reserve, KwaZulu-Natal, South Africa. Journal of Orthoptera Research 30(1): 73-80. https:// doi.org/10.3897/jor.30.59153 Abstract Most grasshopper species have simple and similar life cycles and histo- ries; however, different environmental and ecological factors have different effects on their distribution, sexes, and developmental stages, with effects varying among species. If we are to conserve grasshoppers, we need to un- derstand their ecology and life histories. The aim of this study was to in- vestigate aspects of the life histories and ecology of two recently described co-occurring, congeneric species of wingless grasshoppers, Eremidium armstrongi (Brown, 2012) and Eremidium browni Otte & Armstrong, 2017, at the Doreen Clark Nature Reserve near Pietermaritzburg, South Africa. These two species have limited extents of occurrence, only being known from an endangered forest type in parts of the midland area of KwaZulu- Natal Province, South Africa, and therefore may need conservation action to ensure their long-term survival. No significant differences in the abun- dances of the two Eremidium grasshoppers were found, but their phenol- ogies differed, with the adults of E. armstrongi being present before the adults of E. browni, with some overlap in presence over time. The Eremid- ium grasshoppers were only found in the forest and were more abundant in the forest margin. The Eremidium grasshoppers fed on soft plants from several families. Information on dietary differences between the species is required to determine whether there is potential competition between them. An adult E. browni female kept in an ex situ terrarium laid eggs in the soil, and nymphs took approximately two months to hatch. Keywords adult densities, adult turnover, competition avoidance, microhabitat selec- tion, sympatric congeners Introduction The order Orthoptera is an important element of biodiversity, contributing significantly to the species richness on earth (Bekele 2001). Grasshoppers are considered the most important members of Orthoptera for their contribution to biomass, abundance, and diversity (Mahmood et al. 2004). Grasshoppers are epigeic inverte- brates that sometimes form compact groups comprised of several individuals, which can be hoppers and/or winged adults, or they can be polyphenic (Capinera et al. 1997, Song 2011). The eggs of a mature female are laid in egg pods or clusters in the soil, in the stems of plants, or in rotten wood (Johnsen 1985). Once the egg hatches, the nymph gradually changes into its mature form. Grasshoppers are phytophagous insects (Johnsen 1985), and they can be the primary plant consumers in grassland ecosystems (Gar- diner et al. 2005). The nymph and adult display similar feeding patterns. However, both life stages may respond differently to dif- ferent landscape types at different scales (Bekele 2001). Biotic and abiotic factors, such as host vegetation, plant diver- sity, habitat structure, predators, changes in seasonality, light in- tensity, precipitation, and elevation, influence grasshopper diversi- ty and population dynamics (Bekele 2001, Mahmood et al. 2004, Sirin et al. 2010, Branson 2011, Ely et al. 2011). Grasshoppers are ectotherms, and their body temperature strongly influences most of their physiological processes; their ability to tolerate ambient environmental temperatures and humidity has the potential to determine their richness, spread, ecology, behavior, and the over- all fitness of an individual (Willott and Hassall 1998). Therefore, the diversity and abundance of grasshoppers is relatively low in areas dominated by forest, as these habitats are not suitable for most grasshopper species (Sergeev 2011). The grasshoppers in the family Lentulidae Dirsh, 1956 are wingless. Certain genera in this family, such as Eremidum, have species with small distribution ranges that occur in forests in the province of KwaZulu-Natal and elsewhere in South Africa (e.g., Brown 2012, Otte 2015, Otte and Armstrong 2017). Since various forest types and forests are endangered in KwaZulu-Natal due to clearing, logging, other forest products extraction, fire, and alien plant encroachment (Mucina and Rutherford 2006, Jewitt et al. 2016), these endemic grasshopper species are of conservation con- cern. However, little is known about the life histories and ecology of these species, potentially hampering conservation efforts. The present study focuses on aspects of the life history and ecology of two species, Eremidium armstrongi (Brown, 2012) and E. browni (Otte & Armstrong, 2017) found in Doreen Clark Nature Reserve near Pietermaritzburg, South Africa. These grasshoppers JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) 74 inhabit the Endangered Eastern Mistbelt Forest type (Jewitt 2018) in a restricted area of the midland region of KwaZulu-Natal Prov- ince, South Africa. Although these two species may require con- servation action for their long-term survival, almost nothing is known about their life history and ecology. The main objectives of this study were to determine the time of year during which adults of each species were present, estimate densities and the number of adult individuals of each species in the sampled habitat area, identify microhabitat features that may explain microhabitat preferences of the species, investigate how far the distribution of Eremidium grasshoppers extended into grassland, and, if possible, ascertain how two very similar species can co-occur. Materials and methods Study area.—The Doreen Clark Nature Reserve (29°57.85'S, 30°28.92'E; Fig. 1) is in the suburb of Winterskloof, north-west of R. BRIJLAL, A. RAJAK AND A.J. ARMSTRONG Pietermaritzburg, in the KwaZulu-Natal province of South Africa. The vegetation in this protected area of approximately five hec- tares is the Southern Mistbelt Forest and Midlands Mistbelt Grass- land (Fig. 1; Mucina and Rutherford 2006). The greater part of the reserve is covered by forest, containing a genus of conifers known as Podocarpus as well as many other genera of angiosperms. The forest meets the grassland at the narrow forest-grassland ecotone, where many insect species can be found, including E. armstrongi and E. browni. Between the forest margin and the grassland lies a hiking trail, which contributes to the human disturbances expe- rienced by the Doreen Clark Nature Reserve. The study site was situated along the hiking trail, extending approximately 15 m on either side (Fig. 2). At the beginning of the hiking trail, grasses in the grassland were tall and green, reducing in height and becom- ing drier as the trail moved more towards the west. A small stream runs through the forest. The forest is multistory, with a largely her- baceous understory of forest grasses, sedges, ferns, forbs, etc. Legend Dorren Clark Sampling Area ——— Foot Path JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) R. BRILAL, A. RAJAK AND A.J. ARMSTRONG 75 Estimation of time of year of presence and number of adults in the sam- pled area.—A quantitative direct count method using quadrats was used (Richards and Waloff 1954, Gardiner et al. 2005, Samways et al. 2010). Twenty-seven 1-m? quadrats (Fig. 3), each made from gardening twine and secured at the corners by 3-in nails, were placed uniformly in the study area on 12 November 2018 and left out in the field for the study period. The quadrats were placed 5 m apart to minimize the probability of double-counting grasshop- pers. Twelve quadrats were placed along the forest margin, and the remainder of the quadrats extended into the forest, ending near the stream bank. Sampling was carried out on 15 days spread out over 3 months (13 November 2018 - 14 January 2019). These months were chosen because previous observations determined that E. armstrongi adults were present during October and Novem- ber, and E. browni adults were present in December and Janu- ary. Sampling days were selected based on weather conditions of temperatures greater than 20°C to ensure that the grasshoppers were active and could be easily seen. On each sampling day, the sequence in which the quadrats were sampled was reversed from that of the previous occasion to reduce bias caused by variation in sampling time. Sampling was carried out between 9 am and midday, local time. Species turnover with time of year was determined for adult males only because it was difficult to differentiate between nymphs and females of the two species in the field. These data were ob- tained from the quadrat counts on the total (15) sampling occa- sions and plotted over time to show the turnover of the species. The average density per square meter of E. armstrongi and E. browni observed over the first five sampling occasions (for E. arm- strongi) and over the last five sampling occasions (for E. browni) was calculated from the data. The sampling occasions are given in Table 1. The dates of each sampling occasion were chosen to avoid overlap in the presence of adults of both species. All adult females recorded on a particular sampling occasion were assumed to be of the same species as the males present. The area of the sampling site was calculated using a geographical information system (GIS; QGIS 3.4.15 Madeira) and geographical coordinates collected by a hand-held global positioning system (GPS; Garmin GPSMAP 64). The average number of individuals of both species at the sampling site was then calculated. Average male to female ratios for E. arm- strongi and E. browni over the five sampling occasions for each were calculated from the raw data. An independent t-test was performed using the statistical pack- age SPSS (IBM Corp. Released 2019. IBM SPSS Statistics Subscrip- tion for Windows, Trial Version) to determine if the adult densities of the two species differed significantly within the study area. The assumption of equal variance was tested by performing Levene’s Test for equality of variances, and the assumption that the data are normally distributed was tested using the one-sample Kolmogorov- Smirnov test. The assumptions of equality of variances and normal distribution were met (F = 2.327, df = 8, p = 0.166; a = 5.3177). Thereafter, an independent samples t-test was used to test the null hypothesis that the abundance for E. armstrongi did not differ sig- nificantly from that of E. browni at a significance level of p < 0.05. Identification of microhabitats.—To identify and describe the mi- crohabitats favored by each species, the Braun-Blanquet method (Mueller-Dombois and Ellenberg 1974) was used. This assess- ment was done to see how the variation in vegetation in the 27 Table 1. Number of adult males and females recorded during each sampling occasion, mean (+ one standard deviation) number of adults per day, estimated mean total number of adults in the sam- pled area (877 m’), and estimated adult sex ratio. E. armstrongi (N = 5) E. browni (N = 5) Date Males Females Date Males Females 13 Nov 2018 4 7 08 Jan 2019 5 vi 20 Nov 2018 5 13 09 Jan 2019 8 8 26 Nov 2018 6 4 10 Jan 2019 6 vi 29 Nov 2018 8 10 11 Jan 2019 6 9 01 Dec 2018 6 9 14 Jan 2019 4 7 Total 29 43 Total 29 38 Mean (+1S8.D.) 14.20 (4 3.564) Mean(+1S.D.) 13.40 (+ 2.302) Mean/m? 0.53 Mean/m? 0.50 Mean/877 m7? 468 Mean/877 m7? 435 Male:Female i alls Male:Female 11ES a Fig. 3. Sampling quadrat (A) and female Eremidium armstrongi on the boundary of the quadrat (B). JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) 76 plots relates to the abundance of E. armstrongi and E. browni in order to determine microhabitat preferences. The Braun-Blanquet scale used is given in Table 2. The species of all plants found within each quadrat, start- ing from quadrat 1, was recorded using a labelling system (e.g., P1 = Plectranthus laxiflorus Benth.). Thereafter, the vegetation cover of each plant species found within each quadrat was re- corded using the Braun-Blanquet scale. To ensure that the feed- ing and other behavior of the grasshoppers were undisturbed, a specimen of each representative plant species in the sampling quadrats was collected using a hand spade and labelled accord- ing to the code assigned to it once sampling was completed. These plants were then pressed and dried and identified to the closet known taxonomic group using two plant field guides (Pooley 1998 and Oudtshoorn 1999) and by comparison with specimens in the Bews Herbarium at the University of KwaZulu- Natal, Pietermaritzburg. The degree of constancy for each plant species was calculated by counting the number of quadrats the plant species occupied. A simple observation method was undertaken in the field to identify what plant species the grasshoppers fed on. The grasshop- pers were observed from a distance. If the plant species was un- known, it was allocated a number and a sample taken for later identification in the laboratory. To determine where eggs were laid and the incubation period, a pair of mature grasshoppers from each species, E. armstrongi and E. browni, were captured and kept in captivity. A terrarium was created using a fish tank (61 cm x 32 cm x 33 cm) in which soil and plants from the Doreen Clark Nature Reserve were added. Soil was placed at the bottom of the tank to a depth of approximately 5 cm, and the plants were placed in the soil to provide food and shelter. The tank was covered with shade cloth in such a way as air could circulate and placed near a window for sunlight and heat. Water was added to the tank reg- ularly to prevent plants from dying, and fresh food plants were collected as needed from the study site. The grasshoppers were monitored daily. Distribution of Eremidium grasshoppers into the grassland.—The dis- tribution of the two grasshopper species into the grassland was in- vestigated using a transect sampling method. Strip transects were created using a GPS, starting at the hiking trail and extending into the grassland, with each transect approximately 5 m apart. On each transect, three people sampled: one in the middle and two approximately 1 m away on either side. Sampling was carried out over three days, and only the adults of each species were captured and identified along transects. After identification, each grasshop- per was released behind the samplers to avoid resampling the same individuals. The coordinates of each grasshopper identified along the transects were uploaded into GIS software and used to map the extent of their distribution into the grassland. Table 2. Braun-Blanquet scale. Symbol/Scale Vegetation cover r Some individuals + Many individuals but < 1% 1 1-5% 2 6-25% 3 26-50% 4 51-75% 5 >75% R. BRIJLAL, A. RAJAK AND A.J. ARMSTRONG Results Period of presence of adults.—On the first five sampling occasions (during the period of 13 November to 01 December 2018), only E. armstrongi adult males were present, and during the last five sam- pling occasions (during the period of 08 to 14 January 2019), only E. browni adult males were present. On the intervening five sam- pling occasions (during the period of 12 to 18 December 2018), both E. armstrongi and E. browni adult males were present (Fig. 4). The data for both the males of both species show an asymmetrical bell-curve distribution, with the peak value for E. armstrongi of 8 males on the fourth sampling occasion (29 November 2018) and with the peak value for E. browni of 8 males on the twelfth sam- pling occasion (09 January 2019). Estimated number of adult individuals.—The total number of E. arm- strongi adults recorded over the five days of sampling was 72 indi- viduals, with an average of 14 adult individuals per day, and the total number of E. browni adults was 67 individuals, averaging 13 individuals per day (Table 1). The calculated area for the sampling area in the forest margin habitat, as indicated in Fig. 1, was 877 m2. Eremidium armstrongi had a higher estimated number of adult individuals in the sampled habitat area than E. browni, with a dif- ference of 33 individuals (Table 1). The male to female ratio is biased towards females in both species (Table 1), with E. armstron- gi having a higher female ratio compared to E. browni. There is a greater variation of values in the E. armstrongi dataset than that of E. browni (Table 1). The mean number for E. armstrongi is greater than that of E. browni (Table 1), but not significantly so (t = 0.422, df = 8, p = 0.684). Microhabitats.—Over the five sampling occasions, a total of 59 E. armstrongi (both sexes combined) were recorded in the 12 quadrats at the margin of the forest and 13 E. armstrongi in the 15 quadrats in the forest interior; the respective numbers for E. browni were 52 and 15 (Table 3). E. armstrongi was observed to occur signifi- cantly more than expected in the quadrats at the forest margin (x7, 9) = 13.614, p < 0.001), as was E. browni (x’, ,, = 9.743, p < 0.002). The two quadrats with the highest total mean number for both species were quadrats 5 and 17, and two plant species were re- corded with high degrees of constancy (Table 3). Quadrat 5 had no bare ground and had more than 75% cover of Poaceae and 6-25% cover of Centella asiatica (L.) Urb. This quadrat was on the N an wn B @ E. armstrongi Ww @ E. browni Number of male individuals Sampling occasions Fig. 4. Variation in number of male Eremidium armstrongi and Eremidium browni counted in quadrats over time. JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) R. BRILAL, A. RAJAK AND A.J. ARMSTRONG 77 Table 3. Mean number (over five sampling occasions) of adult Eremidium armstrongi and Eremidium browni counted and Braun-Blan- quet plant cover-abundance in each quadrat. Forest margin quadrat numbers and data are italicized. Refer to Table 1 for explanation of the symbols. 1 2 3 4 5 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Degree of Quadrat number 6 7 080604 0 4 020204 0 061.2 0 0204 0 O Mean no. E. armstrongi Mean no. E. browni Plant species Plectranthus laxiflorus Monopsis stellarioides Miscanthus capensis Prosphytochloa prehensis Cheilanthes viridis vars viridis Impatiens hochstetteri Hypoestes forskaolii Asparagus plumosus Piper capense Isoglossa cooperi Justicia campylostemon Desmodium repandum Dioscorea sylvatica Tricalysia lanceolata Polystichum trankeiense Poaceae Centella asiatica Sanicula elata Cyperus sphaerospermus Crassula c.f. pellucida Achyranthes aspera Oplismenus hirtellus Tradescantia fluminensis Diclis reptans Thunbergia alata Selaginella kraussiana 0:4-0:8.0°4- -0°43.370.2 0.2) 2-032 072:-04- 05 -076c11.6. 0 0 PORN + + NR +t RW RS Rt RRR onl 4 3 0 040406 0 0.2 0402 0 0 O 0.2 1 + + 1 1 s+ Dealt + 1 + 2 + + + 2 2 Tt Dig Ps aT ee. Tt 2 ©=—l oo constancy 0 0.6 0.2 0 0.8 0 6 2 1 1 2 r+ 18 I -2 9 2 13 2 + Tr 9 2 5 7 3 oD 2 5 2 3 3 1 2. 6 1 4 7 + 3 2 1 + 5 5 2 17 Bare ground 5 Fig. 5. Eremidium armstrongi female feeding on Hypoestes forskaolii (A), Eremidium armstrongi male feeding on Diplocyclos palmatus (B), Eremidium species female feeding on Impatiens hochstetteri (C), Eremidium browni mating (D), Eremidium browni female laying her eggs in the terrarium (E), and Eremidium browni nymph hatched in terrarium (F). JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) 78 R. BRULAL, A. RAJAK AND A.J. ARMSTRONG A Legend CC) Forest | om | Grassland = Foot Path = Transect @ Grasshoppers Fig. 6. Extent of the distribution of Eremidium grasshoppers into the grassland. forest margin and received a fair amount of sunlight. It was noted that quadrat 17 was similar to quadrat 5, which was located on the forest margin and had no bare ground. Between 25-50% of the quadrat was covered by Plectranthus laxiflorus Benth. and 6-25% by Thunbergia alata Bojer ex Sims (Table 3; Fig. 5), which received a fair amount of sunlight. Most of the remaining quadrats in which either E. armstrongi or E. browni were observed contained the food plants Impatiens hochstetteri Warb. or Hypoestes forskaolii (Vahl) R.Br. or both (Table 3), while quadrats 1 and 2 received a fair amount of sunlight. Another recorded food plant species, Diplocyclos palmatus (L.) C.Jeffrey, was not found in any of the quadrats. Most of the quadrats where no Eremidium species were observed had a relative- ly high percentage cover of bare ground (Table 3). The female E. browni in the terrarium was observed laying eggs in the soil (Fig. 5). Distribution into the grassland.—Fig. 6 indicates the 12 transects that were sampled and the positions of Eremidium grasshoppers that were found along these transects over the three sampling days. The sampling area included both grassland and forest. Only one male individual was observed along each of transects 1 and 4, two along transect 5, three along each of transects 7 and 8, four along transect 12, six along transect 6, and seven along transect 9. No individual was found along transects 3, 10, and 11. Transect 9 was located at the margin of the forest. The Eremidium grasshoppers recorded along transect 6 were at the margin of the forest, and those along transect 12 were in the for- est. No individual was found in the grassland interior (Fig. 6). Discussion Grasshopper diversity and populations in any area are influ- enced by topography, vegetation, and soil (Lockwood and Lock- wood 2008). They respond to a combination of interacting abiotic and biotic factors that vary over time and space (Branson 2008). However, the direct, indirect, and interacting effects of host vegeta- tion, competition, weather conditions, and other factors on grass- hopper population dynamics are still poorly understood (Skin- ner and Child 2000, Branson 2008). There has been much debate between ecologists about the role of intrinsic and extrinsic factors controlling population dynamics (Ritchie 1996). Grasshoppers show variations in their life history, with each species responding differently to these factors (Branson 2004). According to Latch- ininsky et al. (2011), general grasshopper distribution trends for different regions have been described. However, the processes and factors affecting grasshopper species richness patterns at the differ- ent scales are still being elucidated. Local distribution trends have been discussed with regards to grasshopper diversity in relation to vegetative species composition, habitat structure (Latchininsky et al. 2011), and the overall microhabitat of the species (e.g., Joern 1982). Complex interactions between competing necessities influ- ence habitat selection behavior (Ahnesj6 and Forsman 2006), and the use of resources, such as food and microhabitats, among grass- hopper species is influenced by biotic associations (Joern 1979). Most grasshopper species have specific microhabitat preferenc- es. These preferences are based on the multiple abiotic and biotic factors that make up the microhabitat. Some of these factors in- clude resource availability (e.g., food or nutrients), microclimate variations (e.g., light intensity, temperature, humidity, and precip- itation), structural qualities, suitable hiding places, predation, and competition (Joern 1982, Ahnesj6 and Forsman 2006). For grass- hoppers, some abiotic factors that influence microhabitat utiliza- tion and population size are microclimate, plant structure, plant species richness and abundance, soil characteristics, availability of acceptable oviposition sites and food plants, and suitable hiding places. Hemp and Hemp (2003) used phytosociological relevés, applying the Braun-Blanquet method to ascertain the grasshopper coenoses in the plant communities distinguished in the high-alti- tude grasslands on Mount Kilimanjaro. They could also determine the microhabitat preferences of species from the data. Only a few studies have shown evidence that biotic factors, such as predation and competition, influence microhabitat selection (Joern 1982). According to Joern and Klucas (1993), wherever there are her- bivorous insects such as grasshoppers, food may become limited, resulting in competitive interactions, something which is of inter- est yet remains poorly understood. The congeners E. armstrongi and E. browni are similar in their appearance but also in their mi- JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) R. BRIJLAL, A. RAJAK AND A.J. ARMSTRONG crohabitat preferences and densities. Since these two species share the same microhabitat and possibly food source, they may avoid potentially adverse interaction through differences in phenology, as indicated by the adult male numbers over time (Fig. 4). Com- petition for resources is a possible reason why E. armstrongi and E. browni illustrate a peak number of adult male individuals (and, by extension, adult females) when only a single species is present, allowing sympatry of two very similar species. The adults of E. arm- strongi were observed in the field between October and December 2018, while the adults of E. browni were observed between Decem- ber and February 2019. The number of adult individuals of E. arm- strongi and E. browni were similar in the sampled habitat area (Ta- ble 1), with a slight difference in the recorded male to female ratio. General feeding behavior in grasshoppers, such as food plant specificity, number of taxa in the diet, and the type of vegetation they feed on, is varied (Joern 1983), with most grasshopper species feeding selectively on a variety of plant species from different fami- lies (Sword and Dopman 1999), but preference is evident (Joern 1979). Such grasshoppers are referred to as polyphagous, and this has been seen at an individual, population, and species level (Sword and Dopman 1999). Work done by Rowell (1978) suggested that this view is true for grasshoppers found in the temperate zone, and further studies show a greater degree of specificity in grasshoppers of the tropic zones. Eremidium species are temperate zone grass- hoppers that feed on plants from different families (Fig. 5). At Do- reen Clark Nature Reserve, they displayed some level of preference by feeding on soft green vegetation while apparently avoiding veg- etation with waxy and sticky surfaces. They were observed feeding on Impatiens hochstetteri Warb. (family: Balsaminaceae), Hypoestes forskaolii (Vahl) R.Br. (family: Acanthaceae), and Diplocyclos palma- tus (L.) C.Jeffrey (family: Cucurbitaceae). Observations of feeding by E. armstrongi and E. browni were too few to determine whether they differ in the plant species in their diets, and the method used does not allow extrapolation to a general conclusion. Eremidium grasshoppers were only found along the forest- grassland edge and into the forest interior, probably as a result of particular microhabitat requirements. The perceived differences in microhabitat between the forest interior, its margin, and grassland include vegetation composition and structure, light intensity, tem- perature, and soil compactness. The grassland interior consists of, inter alia, tall, hairy grasses containing relatively large amounts of silica in the body structure and more compacted soils exposed to direct weather conditions (light intensity, temperature, and pre- cipitation). The forest floor, in contrast, consists of soft green veg- etation, moist soft soils, and dappled sunlight. The microhabitat in the sampled area was suitable for Eremidium grasshoppers, but more suitable towards the margin of the forest (Table 3), perhaps owing to the greater availability of sunlight. This area may be ther- mally suitable for egg production, as mating has been observed at the edge on many occasions. Temperature plays an important role in the fitness and survival of grasshoppers; since grasshoppers are ectotherms, their body temperature influences the activities, ecol- ogy and, eventually, overall fitness and development of an individ- ual (Kemp 1986, Willott and Hassall 1998, Ahnesj6 and Forsman 2006). Therefore, grasshoppers need to select thermally suitable mi- crohabitats (Ahnesj6 and Forsman 2006). In the forest, the canopy cover provides shade with patches of sunlight that filters through to the forest floor and that the Eremidium grasshoppers can use to regulate their body temperatures. Grasshoppers select areas that best suit all their requirements because some microhabitats may provide one factor (e.g., shel- ter from predators) but may lack other important factors (e.g., 79 suitable temperatures, foodplants) needed by the species for sur- vival (Ahnesj6 and Forsman 2006). Eremidium grasshoppers were found in quadrats that had food plants and bare ground and that received some sunlight for thermoregulation. Depending on the species, grasshopper eggs are laid in the soil where conditions are suitable for growth (Dempster 1963, Lockwood and Lockwood 2008), with a female producing one or more pods containing 3 to 200 eggs each (Lockwood and Lockwood 2008). A female E. browni laid eggs in the soil of an ex situ terrarium, which were de- termined from the date of appearance of nymphs in the terrarium (Fig. 5) to take approximately two months to hatch. Conclusion Eremidium armstrongi and E. browni are two recently described species of grasshoppers; to conserve them for future generations, it is important to understand their life histories and ecology. The two species are very similar, with no significant difference in abun- dances. Both species occupy the same specific microhabitat with a short period of overlap. Both E. armstrongi and E. browni are selec- tive for microhabitat and were found to be most abundant along the margin of the forest, but also occurred in the forest interior. Food plants included one species in each of three families in the sampled area, but more observations on feeding are needed to de- termine whether the two Eremidium species differ in diet. Further research needs to be done to improve understanding of the two species, especially in terms of diet and reproductive behavior. Di- agnostic features of the adult females of E. armstrongi and E. browni that can be easily seen in the field should be elucidated to enable researchers to tell them apart more easily. Acknowledgements Assistance was provided by Wandile Thwala in the field and was greatly appreciated. We extend special thanks to Dr Clinton Carbutt from Ezemvelo KZN Wildlife and Christina Potgieter from the Bews Herbarium of the University of KwaZulu-Natal in Piet- ermaritzburg for assistance with the plant identifications. We are particularly grateful to Ezemvelo KZN Wildlife for the opportunity to work on this project and for the financial support to make it possible. We thank Claudia Hemp and Maria Marta Cigliano for comments on the manuscript. References Ahnesj6 J, Forsman A (2006) Differential habitat selection by pygmy grasshopper colour morphs; interactive effects of temperature and predator avoidance. Evolutionary Ecology 20: 235-257. https://doi. org/10.1007/s10682-006-6178-8 Bekele SG (2001) Grasshopper ecology and conservation in the Nama- Karoo. PhD thesis, University of Natal, Pietermritzburg. Branson DH (2004) Relative importance of nymphal and adult resource availability for reproductive allocation in Melanoplus sanguinipes (Orthoptera: Acrididae). Journal of Orthoptera Research 13: 239- 246. https://doi.org/10.1665/1082-6467(2004)013[0239:RIONAA| Pan VAG Dp Branson DH (2008) Influence of individual body size on reproductive traits in melanopline grasshoppers (Orthoptera: Acrididae). Journal of Orthoptera Research 17: 259-263. https://doi.org/10.1665/1082- 6467-17.2.259 Branson DH (2011) Relationships between plant diversity and grasshopper diversity and abundance in the Little Missouri National Grassland. Psyche 2011: 748635. [7 pp.]| https://doi.org/10.1155/2011/748635 Brown HD (2012) Additional species of Eremidium Karsch, 1896 with de- scription of a new allied genus from South Africa (Orthoptera: Acrid- JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1) 80 oidea: Lentulidae). Annals of the Ditsong National Museum of Natu- ral History 2: 47-68. Capinera JL, Scherer CW, Simkins JB (1997) Habitat associations of grasshoppers at the MacArthur agro-ecology research center, Lake Placid, Florida. Florida Entomologist 80: 253-261. https://doi. org/10.2307/3495558 Dempster JP (1963) The population dynamics of grasshoppers and lo- custs. Biological Reviews 38: 490-529. https://doi.org/10.1111/j.1469- 185X.1963.tb00791.x Ely SO, Njagi PG, Bashir MO, El-Amin SET, Ahmed H (2011) Diel behav- ioral activity patterns in adult solitarious desert locust, Schistocerca gregaria (Forskal). Psyche: A Journal of Entomology 2011: 459315. [9 pp.] https://doi.org/10.1155/2011/459315 Gardiner T, Hill J, Chesmore D (2005) Review of the methods frequent- ly used to estimate the abundance of Orthoptera in grassland eco- systems. Journal of Insect Conservation 9: 151-173. https://doi. org/10.1007/s10841-005-2854-1 Hemp C, Hemp A (2003) Saltatoria coenoses of high-altitude grasslands on Mt. Kilimanjaro, Tanzania (Orthoptera: Saltatoria). Ecotropica 9: 71-97. Jewitt D (2018) Vegetation type conservation targets, status and level of protection in KwaZulu-Natal in 2016. Bothalia 48: a2294. https://doi. org/10.4102/abc.v48i1.2294 Jewitt D, Goodman PS, O'Connor TG, Erasmus BE, Witkowski ET (2016) Mapping landscape beta diversity of plants across KwaZulu-Natal, South Africa, for aiding conservation planning. Biodiversity and con- servation 25: 2641-2654. https://doi.org/10.1007/s10531-016-1190-y Joern A (1979) Resource utilization and community structure in assem- blages of arid grassland grasshoppers (Orthoptera: Acrididae). Trans- actions of the American Entomological Society 150: 253-300. Joern A (1982) Vegetation structure and microhabitat selection in grass- hoppers (Orthoptera, Acrididae). The Southwestern Naturalist 27: 197-209. https://doi.org/10.2307/3671144 Joern A (1983) Host plant utilization by grasshoppers (Orthoptera: Acridi- dae) from a sandhills prairie. Journal of Range Management 36: 793- 797. https://doi.org/10.2307/3898212 Joern A, Klucas G (1993) Intra-and interspecific competition in adults of two abundant grasshoppers (Orthoptera: Acrididae) from a sandhills grassland. Environmental Entomology 22: 352-361. https://doi. org/10.1093/ee/22.2.352 Johnsen P (1985) Suborder Caelifera. In: Scholtz CH, Holm E (Eds) In- sects of southern Africa. University of Pretoria, Pretoria, 86-95. Kemp WP (1986) Thermoregulation in three rangeland grasshopper species. The Canadian Entomologist 118: 335-343. https://doi. org/10.4039/Ent118335-4 Latchininsky A, Sword G, Sergeev M, Cigliano MM, Lecog M (2011) Lo- custs and grasshoppers: behavior, ecology, and biogeography. Psy- che: A Journal of Entomology 2011: 578327. [4 pp.| https://doi. org/10.1155/2011/578327 Lockwood D, Lockwood J (2008) Grasshopper population ecology: catas- trophe, criticality, and critique. Ecology and Society 13: e34. https:// doi.org/10.5751/ES-02496-130134 R. BRIJLAL, A. RAJAK AND A.J. ARMSTRONG Mahmood K, Abbas K, Shah WH (2004) A Preliminary Study of grasshop- pers (Acrididae: Orthoptera) of Baltistan, Azad Jammu and Kashmir, Pakistan. Pakistan Journal of Zoology 36: 21-25. Mueller-Dombois D, Ellenberg H (1974) Aims and methods of vegetation ecology. John Wiley & Sons, New York. Mucina L, Rutherford MC (2006) The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity In- stitute, Pretoria. Otte D (2015) Revision of the genus Eremidium Karsch with descrip- tions of sixteen new species (Acridoidea: Lentulidae). Transactions of the American Entomological Society 141: 499-544. https://doi. org/10.3157/061.141.0310 Otte D, Armstrong A (2017) Additional Eremidium species from Eastern Cape Province and KwaZulu-Natal Province in South Africa and a new genus, Zulutettix (Acridoidea: Lentulidae). Transactions of the American Entomological Society 143: 305-433. https://doi. org/10.3157/061.143.0211 Oudtshoorn FV (1999) Guide to grasses of southern Africa. Briza Publica- tions, Pretoria. Pooley E (1998) A field guide to wildflowers: KwaZulu-Natal and the east- ern region. Natal Flora Publications Trust, Durban. Richards OW, Waloff N (1954) Studies on the biology and population dy- namics of British grasshoppers. Anti-locust Bulletin 17: 1-182. Ritchie ME (1996) Interaction of temperature and resources in population dynamics and experimental test of theory. Frontiers of population ecology. CSIRO Publishing, Melbourne, 79-91. Rowell HF (1978) Food plant specificity in neotropical rainforest acridids. Entomologia experimentalis et applicata 24: 651-662. https://doi. org/10.1111/j.1570-7458.1978.tb02829.x Samways MJ, McGeoch MA, New TR (2010) Insect conservation: a hand- book of approaches and methods. Oxford University Press, Oxford. Sergeev MG (2011) Distribution patterns of grasshoppers and their kin in the boreal zone. Psyche: A Journal of Entomology 2011: 324130. [9 pp.] https://doi.org/10.1155/2011/324130 Sirin D, Eren O, Ciplak B (2010) Grasshopper diversity and abundance in relation to elevation and vegetation from a snapshot in Medi- terranean Anatolia: Role of latitudinal position in altitudinal dif- ferences. Journal of Natural History 44: 1343-1363. https://doi. org/10.1080/00222930903528214 Skinner KM, Child RD (2000) Multivariate analysis of the factors influenc- ing changes in Colorado grasshopper abundances. Journal of Orthop- tera Research 9: 103-109. https://doi.org/10.2307/3503640 Song (2011) Density-dependent phase polyphenism in nonmodel locusts: a minireview. Psyche: A Journal of Entomology 2011: 741769. [16 pp.] https://doi.org/10.1155/2011/741769 Sword GA, Dopman EB (1999) Developmental specialization and geo- graphic structure of host plant use in a polyphagous grasshopper, Schistocerca emarginata (= lineata) (Orthoptera: Acrididae). Oecologia 120: 437-445. https://doi.org/10.1007/s004420050876 Willott SJ, Hassall M (1998) Life-history responses of British grasshoppers (Orthoptera: Acrididae) to temperature change. Functional Ecology 12: 232-241. https://doi.org/10.1046/j.1365-2435.1998.00180.x JOURNAL OF ORTHOPTERA RESEARCH 2021, 30(1)