TY - JOUR
AU - Roininen,, H
AB - Abstract The ability to mass-rear insects in high densities is a precondition for the edible insect industry but the space requirement has to be determined specifically for each species. Mass-rearing methods for Ruspolia differens Serville (Orthoptera: Tettigoniidae), one of the most consumed edible insect species in East Africa, are currently lacking. Though, these methods are urgently needed to enhance the food security in the region and to reduce the pressure on the wild populations. Here, we experimentally evaluated the effect of rearing density and rearing environment on the survival of R. differens nymphs. We conducted two experiments; in Experiment 1 we used small 0.15-liter rearing containers and in Experiment 2, larger 0.75-liter containers. The rearing densities ranged from 4 to 300 individuals per liter and we used three different rearing environments (‘net’, ‘spikes’, and ‘oat sprouts’). We found that the survival of R. differens nymphs is strongly density-dependent. The suitable rearing density for young R. differens nymphs should be ≤36 nymphs per liter, as in higher densities the mortality of nymphs increases rapidly over the course of time. With rearing densities ≤36 nymphs per liter, a survival rate of 60% can be expected up to 28 d after rearing. The studied environments only had a minor effect on the survival. These results create the basis for the efforts to upscale the rearing of R. differens in the future. the African edible bush cricket, edible insects, nsenene, rearing density, rearing environment Ruspolia differens Serville, commonly known as ‘nsenene’, ‘the edible grasshopper’, or ‘the African edible bush cricket’, is an economically highly valued edible insect species in East Africa (Agea et al. 2008, Okia et al. 2017). It can be an important supplement to the carbohydrate-rich diets in this region, as it is rich in fat, proteins, vitamins, and minerals (Kinyuru et al. 2010, van Huis et al. 2013, Siulapwa et al. 2014). The species also has industrial potential as a source of new food products (Agea et al. 2008). However, the utilization of R. differens is currently entirely based on wild harvesting (van Huis et al. 2013). Over the past years, this practice has commercialized and intensified with powerful light-harvesting stations (Mmari et al. 2017), possibly leading to overexploitation of wild populations in the long term. Recently, there have been attempts to develop methods for the mass-rearing of this species, to improve food security in this region and to protect the wild populations (Lehtovaara et al. 2017; Malinga et al. 2018a,b; Rutaro et al. 2018). Recent studies have improved our understanding of the optimal rearing temperature (Lehtovaara et al. 2018), potential feeds (Lehtovaara et al. 2017; Malinga et al. 2018a,b; Valtonen et al. 2018), and conditions for egg rearing (Ssepuuya et al. 2018). However, the optimal rearing density for R. differens is currently unknown. Better understanding of the rearing density on survival and growth of different-aged nymphs would profoundly enhance the future rearing efforts of this species. In the wild, R. differens occurs in two phases, swarming and nonswarming, which differ in their density and behavior (Bailey and McCrae 1978). The species forms highly aggregated swarms twice a year in the end of the rainy seasons, e.g., in Uganda typically during April–May and November–December (Bailey and McCrae 1978). Matojo and Njau (2010) reported densities of over 10,000 individuals per ha during the swarming season in Tanzania. Densities of the nonswarming phase in the wild are currently unknown. In the laboratory, R. differens has been reared in low densities; four to five first and second instar nymphs per 1-liter, or one later instar nymph per 2- to 3-liter (Hartley 1967, Brits and Thornton 1981), or one adult per 1.7-liter container (Robinson and Hartley 1978). The reason for the low rearing density is the high prevalence of cannibalism, due to which long-horned grasshoppers (Tettigoniidae) are considered more difficult to rear than, for example, crickets (Gryllidae) (Hartley 1967). However, when fresh food and water is provided, cannibalism occurs only during molting (Hartley 1967). It is therefore possible that cannibalism occurs primarily to acquire fluid rather than nutritional benefit (Hartley 1967). In order to maximize the production rate, the rearing efforts should optimize rearing density and insect performance (e.g., development and growth rate), while minimizing mortality (Cortes Ortiz et al. 2016). Rearing density strongly influences the mortality which starts increasing as density reaches a certain species-specific level (Peters and Barbosa 1977). The maximal density that can be supported by a given volume, while keeping the mortality low, is species specific and should be established for each species (Kok 2017). The space requirement per individual also increases during development as the insects grow in size (Mott 2017). Also, the behavior of individuals can change during development. For example, young nymphs of R. differens have been reported to be gregarious (Bailey and McCrae 1978), and therefore might need to be reared in higher densities than later developmental stages. Information on optimal rearing density for each developmental stage would reduce costs related to facilities and materials as well as labor costs per unit of output (Mott 2017). For the successful mass-rearing of R. differens, information on the optimal structure of rearing unit, i.e., the rearing environment, is urgently needed. The structure determines the surface area in the rearing unit that is available for hiding, molting, and resting, and it could allow for the rearing of a larger population of insects per volume (Cortes Ortiz et al. 2016). The surface area of the rearing unit can be modified by the addition of horizontal and vertical structures. For example, in cricket rearing, vertically stacked egg cartons are used to increase individual crawling space (Clifford and Woodring 1990, Cortes Ortiz et al. 2016). When molting, R. differens requires structures where it can firmly hold on to, and hide (as a refuge), due to the high prevalence of cannibalism. Brits and Thornton (1981) reported that nymphs responded to vertical patterns and shapes, and they offered first and second instar nymphs grass as hiding places to prevent cannibalism. Indeed, live shoots of grasses or grains could offer nymphs both refuge and feed. However, for mass-rearing purposes the rearing unit should be built using low-cost and/or easy-to-clean materials (Mott 2017). Vertical, spiky plastic structures could mimic the grass shape but allow for a more maintenance-friendly surface for the nymphs. On the other hand, a netted structure could offer both vertical and horizontal hiding spaces. The purpose of this work was to evaluate the survival and growth of R. differens nymphs reared along a gradient of different densities, and in different environments (‘net’, ‘spikes’, and ‘oat sprouts’). We conducted two experiments: Experiment 1 in small 0.15-liter containers and Experiment 2 in larger 0.75-liter containers. Our specific study questions were: 1) how does the rearing density or 2) environment modify the survival of nymphs and 3) what is the optimal density for rearing nymphs, while keeping the mortality low? We hypothesized that in all rearing densities, the proportion of live R. differens will decrease over time during rearing, but it will decline more steeply in higher densities. We also hypothesized that the survival would be higher in environment ‘net’ than in ‘spikes’ due to the higher number of hiding spaces, which is likely to reduce the likelihood of contact between individuals and thus cannibalism (Fox 1975). We also expected the environment ‘oat sprouts’ to be beneficial for nymphs as the germinating oats provide both natural refuge and extra feed. Materials and Methods Study Organisms and Experimental Design In Experiments 1 and 2, we used newly hatched nymphs (1- to 2-d-old) produced by females of the laboratory population at the University of Eastern Finland. The population originated from individuals collected at Makerere University Agricultural Research Institute (Kabanyolo, Uganda) in 2015 and 2016. Experiment 2 was carried out to complement Experiment 1, so that the rearing densities of Experiment 2 were fitted to the lower end of the densities used in Experiment 1 (Table 1). Furthermore, the rearing environment ‘oat’ was dropped out of the second experiment as maintaining optimal conditions would have been difficult in such a small scale. The experiments were conducted at 28 ± 2°C in 12:12 (L:D) h photocycle. In Experiment 2, the RH was 70 ± 5% (in Experiment 1, we were unable to measure it due to the small size of the experimental containers). During the experiments, nymphs were sedated with carbon dioxide (CO2) gas whenever they were moved to new containers. Table 1. Density treatments of the two experiments scaled to the same volume (individuals per 1 liter) Experiment 1: number of nymphs per unit (0.15 liter) Experiment 2: number of nymphs per unit (0.75 liter) Nymphs per liter 3 4 1 7 13 17 5 33 27 36 40 53 15 100 23 153 30 200 45 300 Experiment 1: number of nymphs per unit (0.15 liter) Experiment 2: number of nymphs per unit (0.75 liter) Nymphs per liter 3 4 1 7 13 17 5 33 27 36 40 53 15 100 23 153 30 200 45 300 Open in new tab Table 1. Density treatments of the two experiments scaled to the same volume (individuals per 1 liter) Experiment 1: number of nymphs per unit (0.15 liter) Experiment 2: number of nymphs per unit (0.75 liter) Nymphs per liter 3 4 1 7 13 17 5 33 27 36 40 53 15 100 23 153 30 200 45 300 Experiment 1: number of nymphs per unit (0.15 liter) Experiment 2: number of nymphs per unit (0.75 liter) Nymphs per liter 3 4 1 7 13 17 5 33 27 36 40 53 15 100 23 153 30 200 45 300 Open in new tab Experiment 1 (0.15-Liter Container) In Experiment 1, rearing units were 0.15-liter cylindrical plastic container, with the top slightly wider than the bottom (top ⌀ 5.5 × height 7 × bottom ⌀ 4.5 cm). The experiment was arranged as 6 × 3 full factorial treatment design, including six rearing densities (1, 5, 15, 23, 30 vs 45 individuals per container; see Table 1 for densities scaled to 1 liter) and three rearing environments (‘net’, ‘spikes’, and ‘oat’). There were eight replicates for each combination of rearing density and rearing environment, resulting in a total of 144 containers, and 2,856 individual nymphs. The containers were randomly assigned into five rearing chambers, to take into account the potential microclimatic variation in the rearing rooms. The rearing environment ‘net’ consisted of 39 cm long × 5 cm wide metal net (holes 1.2 × 1.2 cm), with a green plastic covering, curled in the experimental container. In total, the net created 311 cm of vertical and horizontal hiding space for the nymphs. The rearing environment ‘spikes’ included a structure with 25 pieces of 4 cm long green plastic spikes vertically embedded in a foamy plastic base, creating 75 cm of vertical hiding space. In both treatments, the containers were covered with a plastic film with ventilation holes. The rearing environment ‘oat’ consisted of germinating live oats growing in the experimental container. These containers were covered with cotton mesh and plastic film with holes. For images of each rearing environment, see Supp Fig. S1 (online only). Insects were fed with oatmeal and reindeer pellet (Poro-Elo 1, Suomen Rehu, Finland). They had ad libitum access to feed throughout the experiment and access to water that was absorbed in cotton wool. Once a week, individuals were moved to clean containers. In the ‘oat’ treatments, the nymphs were moved to a new container with fresh oat. The number of surviving nymphs was recorded once a week over a 4-wk period (0, 7, 14, 21, and 28 d after the start of experiment). Experiment 2 (0.75-Liter Container) In Experiment 2, rearing units were 0.75-liter plastic containers (12 × 10 × 9 cm) with netted holes on the top for air circulation. The experiment was arranged as a 4 × 2 full factorial treatment design, including four rearing densities (3, 13, 27 vs 40 individuals per container, see Table 1 for densities scaled to 1 liter) and two rearing environments (‘net’ and ‘spikes’). There were five replicates for each combination of rearing density and environment. The replicates were randomly arranged in five rearing chambers, resulting in a total of 40 containers with 810 individuals. The rearing environment ‘net’ consisted of 62 cm long × 9 cm wide metal net (holes 1.2 × 1.2 cm), with a green plastic covering, folded in the rearing container. In total, the net created 893 cm of vertical and horizontal hiding space for the nymphs. The rearing environment ‘spikes’ consisted of 24 pieces of 8 cm long green plastic spikes embedded in foamy plastic, creating 168 cm of vertical hiding space. Insects were fed with oatmeal, reindeer pellet (Poro-Elo 1, Suomen Rehu, Hyvinkää), crushed linseeds, and piece of carrot. Insects had ad libitum access to feed throughout the experiment, and both the carrot and dry food were replenished every third or fourth day. In each container, insects had access to water that was absorbed in cotton wool. Insects were moved to a clean container every second week. The number of surviving nymphs was counted every third or fourth day, when the food was replenished. Statistical Analyses To analyze if the proportions of R. differens alive decreased during rearing at different rates in different rearing densities or environments, we fitted repeated measures generalized linear models (binary logistic models for events/trials data). The models were fitted separately to data from Experiments 1 and 2. In each model, the rearing container was the replicate and the number of individuals surviving (events) out of those included in the beginning of the experiment (trials) was the response variable. As fixed factors, we included the rearing density, rearing environment, recording day (days since the start of the experiment, see details below), the two-way interactions between these variables, and rearing chamber (due to the low number of rearing chambers, this could not be included as random factor). We took into account the correlation of repeated measurements from each container by including the recording day of each container (subject) as a random factor (repeated covariance type specified as first-order autoregressive, AR1). The recording days were 0, 7, 14, 21, and 28 for Experiment 1, and 0, 7, 14, 21, 28, 55, and 71 d for Experiment 2 (since the start of the experiment). In Experiment 2, the number of surviving nymphs was counted every third or fourth day and therefore, for each container, we estimated the number of individuals alive on day 0, 7, 14, 21, 28, 55, and 71 using LOESS, i.e., a nonparametric regression model, fitted to the data of actual observation date. In Experiment 1, four containers had to be excluded from analyses due to human error with their handling; in total, 140 containers were included in the survival analysis. The LOESS was run with software R (R Development Core Team 2014) version 3.5.0 and all the other analyses with IBM SPSS statistics version 25. Results Survival Experiment 1 The proportion of R. differens nymphs alive decreased over the course of the experiment but with a significantly different rate in the six density treatments (Table 2). Overall, the lower the rearing densities were, the higher the survival remained over the course of the experiment (Fig. 1). There was also a significant interaction between the rearing environment and the recording day (Table 2), suggesting that the proportion of nymphs alive decreased at different rates in the three studied environments. In the environments ‘net’ and ‘spikes’, and in the lowest rearing density, the proportion of individuals alive dropped slightly during the first rearing week, and after this it leveled off (>80% of individuals staying alive). While in environment ‘oat’, there was virtually no mortality during the first 21 d, after which the proportion of individuals alive approximately halved (Fig. 1). In higher rearing densities, the means of survival by the end of the experiment in the environment ‘oat’ ranged only between 7 and 25%, while in the environments ‘net’ and ‘spikes’ the means ranged between 16 and 63%. Table 2. The repeated measures generalized linear model explaining the proportion of R. differens alive in Experiment 1 df1 df2 F P Environment 2 670 0.667 0.514 Density 5 670 10.46 <0.001 Recording day 1 670 289.984 <0.001 Environment × density 10 670 1.157 0.317 Environment × recording day 2 670 8.535 <0.001 Density × recording day 5 670 3.181 0.008 Rearing chamber 4 670 6.659 <0.001 df1 df2 F P Environment 2 670 0.667 0.514 Density 5 670 10.46 <0.001 Recording day 1 670 289.984 <0.001 Environment × density 10 670 1.157 0.317 Environment × recording day 2 670 8.535 <0.001 Density × recording day 5 670 3.181 0.008 Rearing chamber 4 670 6.659 <0.001 In the fitted model, the rearing container (a total of 140) was the replicate and the number of individuals surviving (events) out of those included in the beginning of the experiment (trials) was the response variable. Open in new tab Table 2. The repeated measures generalized linear model explaining the proportion of R. differens alive in Experiment 1 df1 df2 F P Environment 2 670 0.667 0.514 Density 5 670 10.46 <0.001 Recording day 1 670 289.984 <0.001 Environment × density 10 670 1.157 0.317 Environment × recording day 2 670 8.535 <0.001 Density × recording day 5 670 3.181 0.008 Rearing chamber 4 670 6.659 <0.001 df1 df2 F P Environment 2 670 0.667 0.514 Density 5 670 10.46 <0.001 Recording day 1 670 289.984 <0.001 Environment × density 10 670 1.157 0.317 Environment × recording day 2 670 8.535 <0.001 Density × recording day 5 670 3.181 0.008 Rearing chamber 4 670 6.659 <0.001 In the fitted model, the rearing container (a total of 140) was the replicate and the number of individuals surviving (events) out of those included in the beginning of the experiment (trials) was the response variable. Open in new tab Fig. 1. Open in new tabDownload slide Mean survival in Experiment 1, where neonate R. differens were reared in 0.75-liter containers at six densities for 28 d. The three panels show the three rearing environments: ‘net’ (A), ‘spikes’ (B), and ‘oat’ (C). Variation is shown in Supp Table S1 (online only). Fig. 1. Open in new tabDownload slide Mean survival in Experiment 1, where neonate R. differens were reared in 0.75-liter containers at six densities for 28 d. The three panels show the three rearing environments: ‘net’ (A), ‘spikes’ (B), and ‘oat’ (C). Variation is shown in Supp Table S1 (online only). Experiment 2 The proportion of R. differens nymphs alive decreased over the course of the experiment, but at a different rate in the four density treatments (Table 3). Overall, the lower the rearing density was, the higher the survival remained over the course of the experiment (Fig. 2). There was also a significant interaction between the rearing environment and the density and between the rearing environment and the recording day (Table 3), suggesting that the overall effect of density on survival was not similar in the two environments and that the proportion of nymphs alive decreased at different rates in the two studied environments. Survival in the environment ‘net’ was higher (means of survival across densities ranged 39–73%) compared to ‘spikes’ (33–67%) 28 d after the start of the experiment. After this, survival started to decline more rapidly in both environments. Toward the end of the experiment, the survival in the lower densities in environment ‘spikes’ was slightly higher (means of survival across density treatments 7–17/liters at day 71 ± 1 ranged between 28 and 40%) than in the environment ‘net’ (27–28%). Table 3. The repeated measures generalized linear model explaining the proportion of R. differens alive in Experiment 2 df1 df2 F P Environment 1 263 0.473 0.492 Density 3 263 1.956 0.121 Recording day 1 263 201.0 <0.001 Environment × density 3 263 2.862 0.037 Environment × recording day 3 263 3.204 0.024 Density × recording day 1 263 4.357 0.038 Rearing chamber 4 263 0.473 0.755 df1 df2 F P Environment 1 263 0.473 0.492 Density 3 263 1.956 0.121 Recording day 1 263 201.0 <0.001 Environment × density 3 263 2.862 0.037 Environment × recording day 3 263 3.204 0.024 Density × recording day 1 263 4.357 0.038 Rearing chamber 4 263 0.473 0.755 In the fitted model, the rearing container (a total of 40) was the replicate and the number of individuals surviving (events) out of those included in the beginning of the experiment (trials) was the response variable. Open in new tab Table 3. The repeated measures generalized linear model explaining the proportion of R. differens alive in Experiment 2 df1 df2 F P Environment 1 263 0.473 0.492 Density 3 263 1.956 0.121 Recording day 1 263 201.0 <0.001 Environment × density 3 263 2.862 0.037 Environment × recording day 3 263 3.204 0.024 Density × recording day 1 263 4.357 0.038 Rearing chamber 4 263 0.473 0.755 df1 df2 F P Environment 1 263 0.473 0.492 Density 3 263 1.956 0.121 Recording day 1 263 201.0 <0.001 Environment × density 3 263 2.862 0.037 Environment × recording day 3 263 3.204 0.024 Density × recording day 1 263 4.357 0.038 Rearing chamber 4 263 0.473 0.755 In the fitted model, the rearing container (a total of 40) was the replicate and the number of individuals surviving (events) out of those included in the beginning of the experiment (trials) was the response variable. Open in new tab Fig. 2. Open in new tabDownload slide Mean survival in Experiment 2, where neonate R. differens were reared in 0.75-liter containers at four densities for 71 ± 1 d. The two rearing environments: ‘net’ (A) and ‘spikes’ (B). Variation is shown in Supp Table S2 (online only). Fig. 2. Open in new tabDownload slide Mean survival in Experiment 2, where neonate R. differens were reared in 0.75-liter containers at four densities for 71 ± 1 d. The two rearing environments: ‘net’ (A) and ‘spikes’ (B). Variation is shown in Supp Table S2 (online only). When inspecting the raw counts (not proportions, as above) of the individuals alive over the course of the two experiments (Fig. 3), we found a gradient, where rearing densities ≤36 nymphs per liter produced a relatively flat curve up to 28 d of rearing. In higher rearing densities, the first 28 d of rearing lead to increasingly high losses on the individual level. With densities ≤36 nymphs per liter, a survival rate of 60% (mean for survival across all density treatments ≤36 nymphs per liter and environments at day 28) can be expected up to 28 d (Supp Tables S1 and S2 [online only]). Fig. 3. Open in new tabDownload slide Mean number of R. differens nymphs alive at each recording day from the start of the experiment. The values represent data from both Experiments 1 and 2. Data across the different rearing environments are pooled together. Experiment 1 (black lines) lasted for 28 d, whereas Experiment 2 (gray lines) for 71 ± 1 d. Fig. 3. Open in new tabDownload slide Mean number of R. differens nymphs alive at each recording day from the start of the experiment. The values represent data from both Experiments 1 and 2. Data across the different rearing environments are pooled together. Experiment 1 (black lines) lasted for 28 d, whereas Experiment 2 (gray lines) for 71 ± 1 d. Discussion In this study, we found that a suitable rearing density for R. differens young nymphs should be ≤36 nymphs per liter. In densities higher than this, the mortality of nymphs increases rapidly over the course of time. With rearing densities below 36 nymphs per liter, a 60% survival can be expected up to 28 d of rearing. The density-dependent mortality of R. differens nymphs in increasingly high densities in small crowded containers is likely explained by cannibalism and competition (Fox 1975). Cannibalism has been frequently reported from laboratory-reared R. differens (e.g., Hartley 1967). Hartley (1967) suggested that in R. differens, cannibalism primarily takes place to acquire fluid rather than nutritional benefit. In other insect species cannibalism has been shown to increase due to starvation, but it can also just be a response to the presence of vulnerable individuals in crowded environments and does not require starvation as initiator (Fox 1975). In this study, cannibalism was inferred to take place, as in all environments and densities, there were missing individuals between successive survival counts. Higher densities may also increase the competition for food, leading to increased starvation and thereby mortality. In this study, mortality was frequently observed, as dead individuals were found during the counts (but unfortunately, we did not record the ratio between missing and dead individuals). The impact of crowding may differ at different life stages. In early instars, crowding may increase competition, which actually stimulates diet consumption and weight gain in some insect species (Weaver and McFarlane 1990, Schowalter 2006), but as the insects grow, the contact between individuals increases leading to competitive disadvantages. In our study, food was offered in excess and therefore a lower survival in higher densities is more likely due to interference competition rather than depletion competition (Rockwood 2015). In the experimental containers, food was placed in one location and therefore the nymphs may not have had the opportunity to feed ad libitum due to fear of cannibalism. Comparison of our data to previous works indicates that the development of R. differens, in all rearing densities studied here, was markedly delayed compared to the situation when they are reared individually (Brits and Thornton 1981, Lehtovaara et al. 2018). In Experiment 2, none of the individuals reached the adult stage on the 55th day and only 9% reached adult stage by the end of the experiment on the 71 ± 1st day (Supp Table S3 [online only]). When R. differens have been reared individually and in comparable temperatures, the development from neonate nymph to adult takes only 46–56 d (Brits and Thornton 1981, Lehtovaara et al. 2018). Similarly, increased densities have been shown to delay development and reduce pupal mass in tenebrionids (Cortes Ortiz 2016), delay imaginal molt in Gryllus bimaculatus (Otrhoptera: Gryllidae) (Iba et al. 1995), and reduce efficiency of food conversion in T. molitor (Morales-Ramos and Rojas 2015). We assume that, in this study, the developmental delay was caused by the increasing space requirement of growing nymphs. This emphasizes the need of increased space requirement as the nymphs grow in size. However, for mass-rearing to be economically feasible, high densities are required, even though it might lead to a higher mortality or slower development (Cortes Ortiz et al. 2016). Optimization can be done by modifying the rearing density or container size as the insects develop and grow. To establish the needed space requirement of large R. differens nymphs and adults, experiments in larger containers are needed. This would also benefit from analyses of yields, which would give a better idea of the profitability of rearing using this species. Even though the scalability of the results of this study for larger nymphs and adults needs further research, our results create the basis for the efforts to upscale the rearing of R. differens in the future. Yet, our results already indicate that R. differens cannot be generally reared in as high densities as other comparable edible insects (Table 4). Mass-production of edible insects is defined as the production of 1,000 kg/d (Vantomme et al. 2012) which would require 1.7 million R. differens individuals produced per day, assuming their fresh weight on average is 0.6 g (Bailey and McCrae 1978, Brits and Thornton 1981). Using the lowest rearing density in this study, 4 individuals per liter, which produces a very low mortality for nymphs and is probably closer to a suitable rearing density for the adults, mass-production would require space of at least 420 m3 (more than 210 m2). This estimate does not take into account mortality nor breeding stock. House cricket A. domesticus can be reared to harvestable size in densities of at least 133 insects per liter (Parajulee et al. 1993) with a yield of 400–1,000 kg/mo in room size of 50–100 m2 (Mott 2017). Rearing densities of short-horned grasshoppers are more similar to the densities found in this study but with higher survival. Two short-horned grasshopper species, Indian short horned grasshopper, Oxya fuscovittata Marschall (Orthoptera: Acrididae) and Spathosternum prasiniferum prasiniferum Walker (Orthoptera: Acrididae) have been reported to be reared in densities of 10 and 14 insects per liter, respectively, with 12 and 15% mortality (Das et al. 2009). Yet, it should be noted that densities are not directly comparable as the individuals of different species vary in size. Table 4. Rearing densities reported in mass-rearing or high-density rearing of common edible insect species with available data Edible insect species Rearing density per litera Space requirement Number of individuals per experimental unit Author Ruspolia differens 5 nymphsb 60 cm2 per adulta 30 per 21 × ⌀ 19 cm Brits and Thornton (1981) Ruspolia differens 0.3 adults 50–60 per 76 × 46 × 46 cm Hartley (1967) Ruspolia differens 0.6 adults 30 per 0.05 m3 Robinson and Hartley (1978) Acheta domesticus 133 nymphsc 6,000 per 50 × 44 × 20.5 cm Parajulee et al. (1993) Acheta domesticus 400–1,000 kg/mo in 50–100 m2 Mott (2017) Acheta domesticus 2.5 cm2 per cricket Patton (1978) Acheta domesticus 4–8 adults 300–600 per 76 × 31 × 31 cm Clifford and Woodring (1990) Oxya fuscovittata 10 adults 25 per 10 × 10 × 25 cm Das et al. (2009) Spathosternum prasiniferum 14 adults 35 per 10 × 10 × 25 cm Das et al. (2009) Edible insect species Rearing density per litera Space requirement Number of individuals per experimental unit Author Ruspolia differens 5 nymphsb 60 cm2 per adulta 30 per 21 × ⌀ 19 cm Brits and Thornton (1981) Ruspolia differens 0.3 adults 50–60 per 76 × 46 × 46 cm Hartley (1967) Ruspolia differens 0.6 adults 30 per 0.05 m3 Robinson and Hartley (1978) Acheta domesticus 133 nymphsc 6,000 per 50 × 44 × 20.5 cm Parajulee et al. (1993) Acheta domesticus 400–1,000 kg/mo in 50–100 m2 Mott (2017) Acheta domesticus 2.5 cm2 per cricket Patton (1978) Acheta domesticus 4–8 adults 300–600 per 76 × 31 × 31 cm Clifford and Woodring (1990) Oxya fuscovittata 10 adults 25 per 10 × 10 × 25 cm Das et al. (2009) Spathosternum prasiniferum 14 adults 35 per 10 × 10 × 25 cm Das et al. (2009) aDensities are calculated by authors from values given in the article. bNymphs until third instar. cUp to harvestable size. Open in new tab Table 4. Rearing densities reported in mass-rearing or high-density rearing of common edible insect species with available data Edible insect species Rearing density per litera Space requirement Number of individuals per experimental unit Author Ruspolia differens 5 nymphsb 60 cm2 per adulta 30 per 21 × ⌀ 19 cm Brits and Thornton (1981) Ruspolia differens 0.3 adults 50–60 per 76 × 46 × 46 cm Hartley (1967) Ruspolia differens 0.6 adults 30 per 0.05 m3 Robinson and Hartley (1978) Acheta domesticus 133 nymphsc 6,000 per 50 × 44 × 20.5 cm Parajulee et al. (1993) Acheta domesticus 400–1,000 kg/mo in 50–100 m2 Mott (2017) Acheta domesticus 2.5 cm2 per cricket Patton (1978) Acheta domesticus 4–8 adults 300–600 per 76 × 31 × 31 cm Clifford and Woodring (1990) Oxya fuscovittata 10 adults 25 per 10 × 10 × 25 cm Das et al. (2009) Spathosternum prasiniferum 14 adults 35 per 10 × 10 × 25 cm Das et al. (2009) Edible insect species Rearing density per litera Space requirement Number of individuals per experimental unit Author Ruspolia differens 5 nymphsb 60 cm2 per adulta 30 per 21 × ⌀ 19 cm Brits and Thornton (1981) Ruspolia differens 0.3 adults 50–60 per 76 × 46 × 46 cm Hartley (1967) Ruspolia differens 0.6 adults 30 per 0.05 m3 Robinson and Hartley (1978) Acheta domesticus 133 nymphsc 6,000 per 50 × 44 × 20.5 cm Parajulee et al. (1993) Acheta domesticus 400–1,000 kg/mo in 50–100 m2 Mott (2017) Acheta domesticus 2.5 cm2 per cricket Patton (1978) Acheta domesticus 4–8 adults 300–600 per 76 × 31 × 31 cm Clifford and Woodring (1990) Oxya fuscovittata 10 adults 25 per 10 × 10 × 25 cm Das et al. (2009) Spathosternum prasiniferum 14 adults 35 per 10 × 10 × 25 cm Das et al. (2009) aDensities are calculated by authors from values given in the article. bNymphs until third instar. cUp to harvestable size. Open in new tab The environments used in this experiment had a relatively minor impact on survival. Unlike expected, there were no large differences between the environments ‘net’ and ‘spikes’. Furthermore, environment ‘oat’ was possibly beneficial for the youngest nymphs when reared in the lowest rearing densities, but this advantage disappeared, or turned negative later on over the course of rearing. For the youngest nymphs, oat sprouts provided both refuge and extra feed. We assume that the oat sprouts were proportionally too large for the rearing container, and possibly limited the space available for more larger nymphs. Oat sprouts also likely caused a higher humidity than in other environments, which possibly increased the growth of molds in the container. Inclusion of live feeds for the first instars might be more feasible when rearing takes place in a larger scale, when there would be larger empty spaces between the oat sprouts and possibility for enhanced ventilation. This type of mass-rearing environment for R. differens, where the rearing cage is placed on top of a suitable grassland patch in the field, might be functional as a seminatural rearing technique in East Africa. In conclusion, the suitable rearing density of R. differens up to the first 28 d should be below 36 nymphs per liter. In these densities, we can expect survival rate of 60% for young nymphs. However, comparison of our data to previous works indicates that the development of R. differens was markedly delayed when compared to situation when they are reared individually. Yet, for mass-rearing to be economically feasible, higher densities are required, even though it might lead to slower development or higher mortality but with greater yield. Our results are essential for future efforts to develop mass-rearing methods for R. differens, which is needed for improvement of food security and protection of the wild populations in East Africa. Acknowledgments The authors would like to thank Elina Koivisto for her help in the laboratory. Furthermore, we are grateful to the Ministry of Agriculture, Animal Industry and Fisheries and the Department of Livestock Health and Entomology (Uganda) for permission to export the insects. Funding was provided by the Academy of Finland (Project no. 14956 to H.R.) and Joensuu University foundation (to V.J.L.). The experiment was conceived and designed by V.J.L., H.R., and A.V. J.T. and V.J.L. performed the laboratory study. V.J.L. and A.V. conducted statistical analysis. V.J.L. drafted the manuscript. All authors contributed to the interpretation of the data, writing and editing of the manuscript. References Cited Agea , J. G. , D. Biryomumaisho , M. Buyinza , and G. N. Nabanoga . 2008 . Commercialization of Ruspolia nitidula (nsenene grasshoppers) in central Uganda . Afr. J. Food Agric. Nutr. Dev . 8 : 319 – 332 . WorldCat Bailey , W. J. , and A. W. R. McCrae . 1978 . The general biology and phenology of swarming in the East African tettigoniid Ruspolia differens (Serville) (Orthoptera) . J. Nat. Hist . 12 : 259 – 288 . Google Scholar Crossref Search ADS WorldCat Brits , J. A. , and C. H. Thornton . 1981 . On the biology of Ruspolia differens (Serville) (Orthoptera: Tettigoniidae) in South Africa . Phytophylactica . 13 : 169 – 173 . WorldCat Clifford , C. W. , and J. P. Woodring . 1990 . Methods for rearing the house cricket, Acheta domesticus (L.), along with baseline values for feeding rates, growth rates, development times, and blood composition . J. Appl. Entomol . 109 : 1 – 14 . Google Scholar Crossref Search ADS WorldCat Cortes Ortiz , J. A. , A. T. Ruiz , J. A. Morales-Ramos , M. Thomas , M. G. Rojas , J. K. Tomberlin , L. Yi , R. Han , L. Giroud , and R. L. Jullien . 2016 . Insect mass production technologies, pp. 153 – 201 . In A. T. Dossey , J. A. Morales-Ramos , and M. G. Rojas (eds.), Insects as sustainable food ingredients: production, processing and food applications . Academic Press , London, United Kingdom . Google Preview WorldCat COPAC Das , M. , A. Ganguly , and P. Haldar . 2009 . Space requirements for mass rearing of two common Indian Acridid adults (Orthoptera: Acrididae) in laboratory condition . Am. Eur. J. Agric. Environ. Sci . 6 : 313 – 316 . WorldCat Fox , L. R . 1975 . Cannibalism in natural populations . Annu. Rev. Ecol. Syst . 6 : 87 – 106 . Google Scholar Crossref Search ADS WorldCat Hartley , J. H . 1967 . Laboratory culture of a tettigoniid, Homorocoryphus nitidulus vicinus (VLK.) (Orthoptera) . Bull. Entomol. Res . 57 : 203 – 205 . Google Scholar Crossref Search ADS WorldCat van Huis , A. , J. van Itterbeeck , H. Klunder , E. Mertens , A. Halloran , G. Muir , and P. Vantomme . 2013 . Edible insects: future prospects for food and feed security . UNFAO, FAO Forestry Paper , Rome, Italy . Google Preview WorldCat COPAC Iba , M. , N. Takashi , and A. Urano . 1995 . Effects of population density on growth, behavior and levels of biogenic amines in the cricket, Gryllus bimaculatus . Zool. Sci . 12 : 695 – 702 . Google Scholar Crossref Search ADS WorldCat Kinyuru , J. N. , G. M. Kenji , S. N. Muhoho , and M. Ayieko . 2010 . Nutritional potential of longhorn grasshopper (Ruspolia differens) consumed in Siaya district, Kenya . J. Agric. Sci. Technol . 12 : 32 – 46 . WorldCat Kok , R . 2017 . Insect production and facility design, pp. 142 – 173 . In A. van Huis and J. K. Tomberlin (eds.), Insects as food and feed: from production to consumption . Wageningen Academic Publishers , Wageningen, The Netherlands . Google Preview WorldCat COPAC Lehtovaara , V. J. , A. Valtonen , J. Sorjonen , M. Hiltunen , K. Rutaro , G. M. Malinga , P. Nyeko , and H. Roininen . 2017 . The fatty acid contents of the edible grasshopper Ruspolia differens can be manipulated using artificial diets . J. Insects Food Feed . 3 : 253 – 262 . Google Scholar Crossref Search ADS WorldCat Lehtovaara , V. J. , H. Roininen , and A. Valtonen . 2018 . Optimal temperature for rearing the edible Ruspolia differens (Orthoptera: Tettigoniidae) . J. Econ. Entomol . 111 : 2652 – 2659 . Google Scholar PubMed WorldCat Malinga , G. M. , A. Valtonen , V. J. Lehtovaara , K. Rutaro , R. Opoke , P. Nyeko , and H. Roininen . 2018a . Diet acceptance and preference of the edible grasshopper Ruspolia differens (Orthoptera: Tettigoniidae) . Appl. Entomol. Zool . 53 : 229 – 236 . Google Scholar Crossref Search ADS WorldCat Malinga , G. M. , A. Valtonen , V. J. Lehtovaara , K. Rutaro , R. Opoke , P. Nyeko , and H. Roininen . 2018b . Mixed artificial diets enhance the developmental and reproductive performance of the edible grasshopper, Ruspolia differens (Orthoptera: Tettigoniidae) . Appl. Entomol. Zool . 53 : 237 – 242 . Google Scholar Crossref Search ADS WorldCat Matojo , N. D. , and M. A. Njau . 2010 . Plasticity and biosystematics of swarming of the conehead Ruspolia differens Serville (Orthoptera: Conocephalidae) . Int. J. Integr. Biol . 9 : 97 – 103 . WorldCat Morales-Ramos , J. A. , and M. G. Rojas . 2015 . Effect of larval density on food utilization efficiency of Tenebrio molitor (Coleoptera: Tenebrionidae) . J. Econ. Entomol . 108 : 2259 – 2267 . Google Scholar Crossref Search ADS PubMed WorldCat Mmari , M. W. , J. N. Kinyuru , H. S. Laswai , and J. K. Okoth . 2017 . Traditions, beliefs and indigenous technologies in connection with the edible longhorn grasshopper Ruspolia differens (Serville 1838) in Tanzania . J. Ethnobiol. Ethnomed . 13 : 60 . Google Scholar Crossref Search ADS PubMed WorldCat Mott , G . 2017 . Cricket rearing, pp. 270 – 287 . In A. van Huis and J. K. Tomberlin (eds.), Insects as food and feed: from production to consumption . Wageningen Academic Publishers , Wageningen, The Netherlands . Google Preview WorldCat COPAC Okia , C. A. , W. Odongo , P. Nzabamwita , J. Ndimubandi , N. Nalika , and P. Nyeko . 2017 . Local knowledge and practices on use and management of edible insects in Lake Victoria basin, East Africa . J. Insects Food Feed . 3 : 83 – 93 . Google Scholar Crossref Search ADS WorldCat Patton, R.L. 1978. Growth and development parameters for Acheta domesticus. Ann. Entomol. Soc. Am. 71: 40–42. Parajulee , M. N. , G. R. DeFoliart , and D. B. Hogg . 1993 . Model for use in mass-production of Acheta domesticus (Orthoptera: Gryllidae) as food . J. Econ. Entomol . 86 : 1424 – 1428 . Google Scholar Crossref Search ADS WorldCat Peters , T. M. , and P. Barbosa . 1977 . Influence of population density on size, fecundity, and developmental rate of insects in culture . Ann. Rev. Entomol . 22 : 431 – 450 . Google Scholar Crossref Search ADS WorldCat R Development Core Team . 2014 . R: a language and environment for statistical computing . R Foundation for Statistical Computing , Vienna, Austria . http://www.r-project.org/ (accessed 26 June 2018 ). Google Preview WorldCat COPAC Robinson , D. J. , and J. C. Hartley . 1978 . Laboratory studies of a tettigoniid (Insecta: Orthoptera) Ruspolia differens (Serville): colour polymorphism . J. Nat. Hist . 12 : 81 – 86 . Google Scholar Crossref Search ADS WorldCat Rockwood , L. L . 2015 . Introduction to population ecology , 2nd ed. John Wiley and Sons , Chichester, United Kingdom . Google Preview WorldCat COPAC Rutaro , K. , G. M. Malinga , R. O. Opoke , V. J. Lehtovaara , F. Omujal , P. Nyeko , H. Roininen , and A. Valtonen . 2018 . Artificial diets determine fatty acid composition in edible Ruspolia differens (Orthoptera: Tettigoniidae) . J. Asia Pac. Entomol . 21 : 1342 – 1349 . Google Scholar Crossref Search ADS WorldCat Schowalter , T. D . 2006 . Insect Ecology: an ecosystem approach , 2nd ed. Academic Press , San Diego, CA . Google Preview WorldCat COPAC Siulapwa , N. , A. Mwambungu , E. Lungu , and W. Sichilima . 2014 . Nutritional value of four common edible insects in Zambia . Int. J. Sci. Res . 3 : 876 – 884 . WorldCat Ssepuuya , G. , C. M. Tanga , I. Yekko , F. Sengendo , C. T. Ndagire , K. K. M. Fiaboe , J. Karungi , and D. Nakimbugwe . 2018 . Suitability of egg hatching conditions and commonly available food plants for rearing the long-horned grasshopper Ruspolia differens Serville (Orthoptera: Tettigoniidae) . J. Insects Food Feed . J. Insects Food Feed 4: 253–261. doi:. WorldCat Crossref Valtonen , A. , G. M. Malinga , P. Junes , R. Opoke , V. J. Lehtovaara , P. Nyeko , and H. Roininen . 2018 . The edible katydid Ruspolia differens is a selective feeder on the inflorescences and leaves of grass species . Entomol. Exp. Appl . 166 : 592 – 602 . Google Scholar Crossref Search ADS WorldCat Vantomme , P. , E. Mertens , A. Van Huis , and H. C. Klunder . 2012 . Assessing the potential of insects as food and feed in assuring food security . In FAO (ed.), Summary report of the technical consultation meeting . Rome, Italy . http://www.fao.org/3/an233e/an233e00.pdf Google Preview WorldCat COPAC Weaver , D. K. , and J. E. McFarlane . 1990 . The effect of larval density on growth and development of Tenebrio molitor . J. Insect Physiol . 36 : 531 – 536 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Space and Shelter Requirement of Nymphs in the Mass-Rearing of the Edible Ruspolia differens (Orthoptera: Tettigoniidae)
JF - Journal of Economic Entomology
DO - 10.1093/jee/toz065
DA - 2019-08-03
UR - https://www.deepdyve.com/lp/oxford-university-press/space-and-shelter-requirement-of-nymphs-in-the-mass-rearing-of-the-tbJlDigc1c
SP - 1651
VL - 112
IS - 4
DP - DeepDyve
ER -