2 College of Life Science, Tarim University, Alar 843300, China;
3 School of Life Science and Ecology, Hainan Tropical Ocean University, Sanya 572022, China;
4 Department of Zoology, Weber State University, 1415 Edvalson Street, Dept 2505, Ogden, UT 84408-2505, USA
Resistance characterized by actively overcoming environmental stress either at an individual or group level and tolerance characterized by minimizing vital functions or/and forming resistant stages are two main strategies used by organisms to adapt to environmental heterogeneity (Lopes et al., 2004; Alekseev et al., 2006). A wide range of invertebrates exhibits tolerance by undergoing dormancy in some fixed life stages (Alekseev et al., 2007; Koštál and Denlinger, 2011). There are two main types of dormancy, quiescence and diapause (Koštál, 2006). Quiescence refers to a hypo-metabolic state imposed directly by environmental (exogenous) conditions; whereas diapause is a dormancy state where the arrest of development is initiated by internal factors (Lavens and Sorgeloos, 1987; Brendonck, 1996). In insects, diapause-inducing signals (or token stimuli) are always perceived during a fixed and specific sensitive period, which is named sensitive stage or critical stage for diapause induction. The stage is genetically determined and ranged from various periods within the parental generation to different stages of embryonal, larval, pupal development and adult individual (Koštál, 2006).
The brine shrimp Artemia (Crustacea: Anostraca) is an embryonic diapausing crustacean that has two reproductive modes, oviparity producing diapause embryos / diapause cysts / resting eggs and ovoviviparity producing free-swimming nauplii (Alekseev and Starobogatov, 1996; Nambu et al., 2004). Many studies have shown that physicochemical factors like photoperiod, temperature, and salinity (especially photoperiod) are the token stimuli inducing oviparity in Artemia (e.g. Berthélémy-Okazaki and Hedgecock, 1987; Nambu et al., 2004; Dai et al., 2011; Wang et al., 2017).
Previous studies have documented some evidence for the existence of a critical life-stage sensitive to environmental cues (Provasoli and Pintner, 1980; Berthélémy-Okazaki and Hedgecock, 1987), but the stage has not been precisely determined (Dai et al., 2011). The identification of the critical stage receiving the stimulus to switch reproduction mode is an important step to better understand the complex Artemia life cycle. To determine the critical stage for inducing oviparity in Artemia, shift-culture experiments were carried out in this study. Following the recommendation of Abatzopoulos et al. (2003), a clonal parthenogenetic Artemia, sensitive to photoperiod and temperature (Wang et al., 2017), was selected as the experimental animal for two reasons. Firstly, bisexual species might be affected by maternal and paternal heterozygosity (Gajardo and Beardmore, 1989). Secondly, wild parthenogenetic populations have sometimes different genetic structures and/or with different ploidy levels (Zhang and Lefcort, 1991).2 MATERIAL AND METHOD 2.1 Artemia and stock cultures
The previously established diploid parthenogenetic Artemia clone (BRK53) obtained from the population at Barkol Lake (43°40′N, 92°47′E), Xinjiang, China, was used in this study. The reproduction mode of this clone is sensitive to photoperiod and temperature. Under longer daylight and higher temperature (18 h L:6 h D and 27℃), more than 90% broods were nauplii and almost 100% females produced nauplii in their first brood; while under shorter daylight and lower temperature (6 h L:18 h D and 19℃), the production of diapaused cysts increased drastically (Wang et al., 2017).
The stock population was cultured in 3-L glass aquaria at the condition of producing nauplii (temperature of 27℃, photoperiod of 18L׃6D and salinity of 70). Culture media were prepared by adding natural sea salt (Qingdao Salt Distribution Office, Qingdao, China) to seawater. Temperatures and photoperiods were controlled by illumination incubators (GZX-300, Ningbo Jiangnan Instrument Factory, Ningbo, China). Artemia was fed a homogenized mixture of Dunaliella sp. powder (Tianjian Biology Technology Co. Ltd., China) and LANSY-Shrimp ZM powder (INVE Asia Services Ltd., Thailand) (1:1) following the protocols of Triantaphyllidis et al. (1995). During the experiment period, the evaporative loss of water was replenished with distilled water. Every time an offspring brood was deposited, the culture medium was exchanged, and the resting eggs or nauplii were removed.2.2 Shift-culture experiments
Two shift-culture experiments were conducted following the method previously used in insects (Wagner et al., 1999; Kurban et al., 2005). They are mentioned below as "main experiment" and "supplementary experiment", respectively.
In the main experiment, Artemia were shifted by two different ways, from the conditions of producing nauplii (temperature 27℃, photoperiod 18L׃6D) to the conditions of producing diapaused cysts (temperature 19℃, photoperiod 6L׃18D), and from the conditions of producing diapaused cysts to the conditions of producing nauplii, respectively. Referring to the results of previous studies (Jackson and Clegg, 1996; Dai et al., 2011) and our pilot experiment, post-larva Ⅱ, post-larva Ⅲ, post-larva Ⅳ, post-larva Ⅴ (oocytes appearing as opaque dots in ovaries), adult Ⅰ (oocytes in oviducts) and adult Ⅱ (oocytes in uterus), were selected as the stages for shifting Artemia. Morphological characters, observed and photographed under a Nikon SMZ800 stereomicroscope adapted with a Nikon DS-5M digital camera, were used to distinguish each life stage following Cohen et al. (1999). When transferring from nauplius to cyst production conditions, experimental Artemia were selected from the stock culture, and each individual was moved to a 50-mL Falcon tube containing 30 mL culture medium. The brine shrimp were then cultured under the cyst production conditions. Meanwhile, another set of individuals were prepared by the same method but were cultured under the conditions of the stock cultures (nauplius production conditions). These cultures were designated as control.
In the reverse direction experiments, newly-born nauplii (Nauplius Ⅰ; see Cohen et al., 1999) in the stock cultures were selected and moved to a different aquarium for the reverse experimental stock cultures. They were reared by the same methods as for the control stock cultures but under the conditions of producing diapaused cysts. When enough (>200) individuals developed to each of the previously mentioned stages, individuals were selected and moved to the Falcon tubes as in the last experiment, but here the control brine shrimps were reared successively under cyst production conditions, while the experimental brine shrimps were shifted to the conditions of producing nauplii.
Each control/experimental consisted of 30 individuals and the experiments were replicated thrice. All individuals were acclimated at 23℃ for ~6 h before being transferred from 19℃ to 27℃, and vice versa. Other experimental protocols were the same as the stock culture. The stock, control and experimental cultures are summarized in Table 1.
During the experimental period, the reproductive modes of the first two broods were recorded for each brine shrimp. If an individual produced a brood of cysts, an "oviparous brood" was recorded; if the shrimp produced nauplii, an "ovoviviparous brood" was recorded. In some cases, many nauplii and a small number of "irregular" cysts were observed together; these broods were defined as ovoviviparous broods (for more information see Berthélémy-Okazaki and Hedgecock, 1987; Wang et al., 2017). The incidence of oviparous broods (the proportion of oviparous females in total reproducing females) was taken as the criterion for evaluating the oviparity-inducing effect.
After the critical stage (which should be placed between two adjacent stages that could and could not convert the reproductive mode) was roughly determined by the main experiment, a supplementary experiment was conducted to precisely determine the critical stage for oviparity induction. In this experiment, individuals at these two stages (reproductive mode converted, and reproductive mode not converted in the main experiment) and a stage between them (see below) were used as shifting for the brine shrimp experiments. As in the main experiment, Artemia were also shifted by two different ways. The experimental procedure was the same as that in the main experiment.
The data were statistically analyzed by SPSS 16. Differences among means of the controls or treatments were analyzed by one-way ANOVA coupled with a Tukey's multiple-comparison test (P < 0.05). The t-test was used to determine significant differences (P < 0.05) between controls and treatments at the same stage.2.3 Histological studies
During the supplementary experiment, specimens (n=3) at each stage were sampled to determine the histological characteristics of the female reproductive organs. They were fixed in 4% paraformaldehyde in 0.1 mol/L phosphate buffered saline (PBS, pH 7.4) for about 24 h. After dehydration in gradient alcohol series, the tissues were passed through xylene and then embedded in paraffin. After the blocks were cut at 6 μm transversely, the sections were mounted on slides and then stained with hematoxylin and eosin (HE). The ovarian development stages were photographed with a Nikon E600 microscope adapted with an Olympus DP72 camera.3 RESULT 3.1 Reproduction modes of females shifted at different development stages
Results of the main experiment are shown in Tables 2 and 3. When shifting Artemia from nauplius production condition to cyst production condition (Table 2), low percentages of oviparity (0.0%–4.5%) in both the first and second broods of the controls showed no significant differences among the different stages (post-larva Ⅱ, post-larva Ⅲ, post-larva Ⅳ, post-larva Ⅴ, adult Ⅰ and adult Ⅱ). For the first brood of treatments, almost all females produced diapause cysts when individuals were shifted at post-larva Ⅱ stage (98.9%±1.9% of the reproducing females (same below)) and at post-larva Ⅲ stage (96.7%±3.3%) which both were significantly different (P < 0.01) from the controls. Females shifted at post-larva Ⅳ, postlarva Ⅴ, adult Ⅰ and adult Ⅱ stages produced diapause cysts (1.2%–4.5%) which were not significantly different from the controls. Similar results were obtained in the second brood; however, significantly more (P=0.01) females shifted at post-larva Ⅳ stage produced diapause cysts (30.7%±9.8%) compared to the controls (3.4%±3.3%). For both broods, percentages of oviparity (96.1%–100.0%) in the females shifted at post-larva Ⅱ/Ⅲ stage were significantly higher (P < 0.05) compared to those shifted at the later stages (1.1%–30.7%). For either brood, no significant difference was detected between treatments of post-larval Ⅱ and post-larval Ⅲ stages. Among the other treatments, only the second brood of the post-larval Ⅳ treatment had a higher oviparity (30.7%) compared to treatments of the post-larval Ⅴ, adult Ⅰ and adult Ⅱ (1.1%–3.7%).
When shifting females from cyst to nauplius production condition (Table 3), the percentages of oviparous broods in all controls (97.7%–100.0%) showed no significant differences among them. For the first brood of treatment, the oviparity (1.3%–1.6%) of the post-larva Ⅱ and Ⅲ stages showed significant differences (P < 0.01) compared to the controls (97.7%–98.7%); however, there were no significant differences between the control and the treatment when the females were shifted after the post-larva Ⅲ stage. For the second brood, oviparity of the treatments post-larva Ⅱ/Ⅲ/Ⅳ stages (1.3%–7.1%) and the corresponding controls (97.7%–100.0%) showed significant differences (P < 0.01); however, no significant differences were found when females were shifted after the post-larva Ⅳ stage.
Based on the results of the main experiment, females at post-larva Ⅲ and post-larva Ⅳ stages, and a stage between them were shifted in the supplementary experiment. The morphological definition of the "between post-larva Ⅲ and Ⅳ" stage and its comparison with the two other stages are shown in Fig. 1. At post-larva Ⅲ stage, the ovisac was a rudimentary hump that reached the beginning of the third abdominal segment (Fig. 1a). At the stage postlarva Ⅳ, the ovisac concealed two-thirds of the third abdominal segment and a pair of sharp ventral spines were visible (Fig. 1c). At the "between post-larva Ⅲ and Ⅳ" stage, a developing ovisac was observed and reached the middle of the third abdominal segment, and its blunt ventral spines were visible (Fig. 1b). These characters are closer to post-larva Ⅳ than postlarva Ⅲ, and thus the stage is called an "early phase of post-larva Ⅳ". In all the three stages, the ovaries were transparent. Opaque dots, which was a character of the post-larva Ⅴ (Criel, 1989) were not observed microscopically on the ovaries. (Fig. 1).
In the supplementary experiment when females were shifted from the nauplius to cyst production conditions, the results were similar to those of the main experiment. Low oviparity (3.6%–5.8%) was observed in females from the control groups. Significantly (P < 0.01) more females shifted at postlarva Ⅲ and "early phase of post-larva Ⅳ" stages produced cysts (96.5%±3.4% and 67.7%±4.7%, respectively), compared to the controls (Table 4). In contrast, low oviparity (6.1%±2.2%) was observed in individuals shifted at the post-larva Ⅳ stage in the first brood, a non-significant result compared to that of the controls. All three treatments were significantly different from each other (P < 0.01). For second brood, females shifted at the post-larva Ⅲ and "early phase of post-larva Ⅳ" stages showed high oviparity (97.7%±4.0% and 96.4%±3.6%, respectively). Additionally, higher oviparity (31.9%±7.6%) was found in females shifted at the post-larva Ⅳ stage compared to that of the first brood. Oviparity values of the post-larva Ⅲ and "early phase of post-larva Ⅳ" stages were significantly different (P < 0.01) compared to those of the post-larva Ⅳ stage. The oviparity of the treatments between the post-larva Ⅲ and "early phase of post-larva Ⅳ" stages showed no significant differences. All oviparity values of the three treatments were significantly different (P < 0.01) compared to those of the controls.
In the experiment of shifting females from cyst to nauplius production condition (Table 5), most control females (97.6%–100%) produced diapause cysts in both the first and second broods. In the first brood of treatment groups, low oviparity (1.1%±1.9%) was observed in females shifted at post-larva Ⅲ stage; while higher oviparity (33.9%±9.8%) was found in "early phase of post-larva Ⅳ" stage; both values were significantly different (P < 0.01) compared to those of the controls. Although observed oviparity between females shifted at post-larva Ⅳ stage (89.9%±3.2%) and the control (97.6%±2.1%) was relatively similar, the difference was significant (P=0.03). Oviparity among all the three treatments was significantly different (P < 0.01). For the second broods, low oviparity (< 5%) was observed; oviparity among the females shifted at the three different stages were nonsignificant.3.2 Histology
Histological observations showed that the ovary expanded markedly from the stage post-larva Ⅲ to the stage post-larva Ⅳ, with its area (single ovary) in the section increasing from less than 2% to more than 4% of the whole section area (Fig. 2a, b, c). Ovaries at post-larva Ⅲ stage had similar-sized cells (5–7 μm in diameter) (Fig. 2a′), while ovaries of Artemia at the "early phase of post-larva Ⅳ" and post-larva Ⅳ stages contained cells of different diameter sizes, with the smaller ones being 5–10 μm, the median ones 12–15 μm (observed in "early phase of post-larva Ⅳ" and post-larva Ⅳ stages; Fig. 2b′, c′), and the larger ones 25–40 μm (seen only at post-larva Ⅳ stage; Fig. 2c′).4 DISCUSSION
Abiotic factors exerting on some developmental stages have been proposed to affect changes in the reproductive modes of Artemia. Provasoli and Pintner (1980) showed that a change of photoperiod from shorter to longer daylight could shift the reproductive mode of Artemia to ovoviviparity, when the shift was applied 10–12 days before the production of first brood. Berthélémy-Okazaki and Hedgecock (1987) suggested that the critical time for the oviparity determination of Artemia franciscana Kellogg should happen in the late juvenile stage (6–7 mm long), just before the ovaries become visible microscopically. According to the studies on morphology and expression of p26 gene, Liang and MacRae (1999) considered that the reproductive mode of A. franciscana was determined at early oocyte stage when shell glands were already differentiated. Dai et al. (2011) recently showed that the differential gene expression determined the reproductive mode of Artemia at oocyte stage.
In the main experiment of this study, the reproductive mode of the first brood could not be converted when Artemia were shifted at post-larva Ⅳ stage, but it was effectively converted when females were shifted at the post-larva Ⅲ stage. This finding confirms the existence of a sensitive stage for oviparity/diapause induction in the life history of Artemia.
The results of supplementary experiment further indicated that the exact sensitive stage was placed at the so-called "early phase of post-larva Ⅳ" stage, which is characterized by the ovisac reaching posteriorly to the middle of the third abdominal segment, ventral blunt spines visible, and ovary with dimorphic cells. Before and at this stage, either reproductive modes (i.e. oviparity or ovoviviparity) could be switched to another by changing the culture conditions; whereas after this stage the reproductive mode of the first brood would be fixed. However, the token stimuli exerted on this critical stage could not alternate the reproductive mode of the second brood, suggesting that reproductive cells, rather than the ovary or whole animal, could be the final effectors responding to the token stimuli.
Oocytes of Artemia are known to mature in batches (Versichele and Sorgeloos, 1980). The larger cells observed in the ovary of Artemia at post-larva Ⅳ stage, as well as the early phase of post-larva Ⅳ might represent oocytes destined to become offspring of first brood. The reproductive mode of these larger oocytes may be first influenced by the token stimuli (at the early phase of post-larva Ⅳ), while the reproduction mode of the smaller ones could be determined in the later stages to become the offspring of later broods (e.g. second brood). Since the size and cytoplasmic volume of reproductive cells have begun to increase at the early phase of post-larva Ⅳ, the development stage of oocytes should be at the previtellogenesis phase (Criel, 1989). During this stage, numerous free ribosomes, mitochondria and single cisterns of the rough endoplasmic reticulum are in the cytoplasm (Poprawa, 2005), indicating the beginning of active gene expression. Dai et al. (2011) reported when oocytes enter the oviduct, many genes, including several known diapause-specific genes (e.g. ArHsp22), are differentially expressed between diapause- and nauplii-destined oocytes. The previtellogenesis is likely the stage that oocytes receive the token stimuli signals, which trigger the differential gene expression of the two destined oocytes.
Regarding the second brood, this study showed that its reproductive mode could be converted when Artemia were shifted at post-larva Ⅳ or earlier stages (Tables 2–5) but could not be converted when shifting performed on post-larva Ⅴ or later stages (Tables 2, 3). When shifting at post-larva Ⅳ, the conversion seemed not complete, especially when shifted from ovoviviparity condition to oviparity condition (oviparity ratio ׃3.4%±3.3% in control and 30.7%±9.8% in treatment of the main experiment; 4.0%±4.0% in control and 31.9%±7.6% in treatment of the supplementary experiment (Tables 2, 4). These results suggest that the maternal stage critical to the second brood could be post-larva Ⅳ.5 CONCLUSION
The critical period inducing oviparity (embryo diapause) might be at the previtellogenesis stage of oocytes during the maternal "early phase of post-larva Ⅳ" for the first brood, and post-larva Ⅳ for the second brood. During the critical stage, different environmental conditions may induce different patterns of gene expression and thereby the oocytes to develop differently to become either swimming nauplii or diapaused embryos.6 DATA AVAILABILITY STATEMENT
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.7 ACKNOWLEDGEMENT
We are grateful to LIAO Qiuping (Tarim University) for her technical assistance in the histological protocols.
Abatzopoulos T J, El-Bermawi N, Vasdekis C, Baxevanis A D, Sorgeloos P. 2003. Effects of salinity and temperature on reproductive and life span characteristics of clonal Artemia. (International Study on Artemia. LXVI). Hydrobiologia, 492(1-3): 191-199. DOI:10.1023/A:1024826702830
Alekseev V R, Hwang J S, Tseng M H. 2006. Diapause in aquatic invertebrates:what's known and what's next in research and medical application. Journal of Marine Science and Technology, 14(4): 269-286.
Alekseev V R, Ravera O, de Stasio B T. 2007. Introduction to diapause. In: Alekseev V R, De Stasio B, Gilbert J J eds.Diapause in Aquatic Invertebrates Theory and Human Use. Springer, Dordrecht. p.3-10.
Alekseev V R, Starobogatov Y I. 1996. Types of diapause in Crustacea:definitions, distribution, evolution. Hydrobiologia, 320(1-3): 15-26. DOI:10.1007/BF00016801
Berthélémy-Okazaki N J, Hedgecock D. 1987. Effect of environmental factors on cyst formation in the brine shrimp Artemia. In: Sorgeloos P, Bengtson D A, Decleir W, Jaspers E eds. Artemia Research and Its Applications, Vol. 3, Ecology, Culturing, Use in Aquaculture. Universa Press, Wetteren. p.167-182.
Brendonck L. 1996. Diapause, quiescence, hatching requirements:what we can learn from large freshwater branchiopods (Crustacea:Branchiopoda:Anostraca, Notostraca, Conchostraca). Hydrobiologia, 320(1-3): 85-97. DOI:10.1007/BF00016809
Cohen R G, Rodríguez Gil S G, Vélez C G. 1999. The postembryonic development of Artemia persimilis Piccinelli & Prosdocimi. Hydrobiologia, 391(1-3): 63-80.
Criel G R J. 1989. Morphological study of the ovary of Artemia. In: Warner A H, MacRae T H, Bagshaw J C eds.Cell and Molecular Biology of Artemia Development.Springer, Boston, MA. p.99-129.
Dai Z M, Li R, Dai L, Yang J S, Chen S, Zeng Q G, Yang F, Yang W J. 2011. Determination in oocytes of the reproductive modes for the brine shrimp Artemia parthenogenetica. Bioscience Reports, 31(1): 17-30. DOI:10.1042/BSR20090141
Gajardo G M, Beardmore J A. 1989. Ability to switch reproductive mode in Artemia is related to maternal heterozygosity. Marine Ecology Progress Series, 55: 191-195. DOI:10.3354/meps055191
Jackson S A, Clegg J S. 1996. Ontogeny of low molecular weight stress protein p26 during early development of the brine shrimp, Artemia franciscana. Development, Growth & Differentiation, 38(2): 153-160.
Koštál V. 2006. Eco-physiological phases of insect diapause. Journal of Insect Physiology, 52(2): 113-127. DOI:10.1016/j.jinsphys.2005.09.008
Koštál V, Denlinger D L. 2011. Dormancy and developmental arrest in invertebrates. Journal of Insect Physiology, 57(5): 537. DOI:10.1016/j.jinsphys.2011.04.001
Kurban A, Yoshida H, Izumi Y, Sonoda S, Tsumuki H. 2005. Pupal diapause of Helicoverpa armigera:sensitive stage for photoperiodic induction. Applied Entomology and Zoology, 40(3): 457-460. DOI:10.1303/aez.2005.457
Lavens P, Sorgeloos P. 1987. The cryptobiotic state of Artemia cysts, its diapause deactivation and hatching: a review. In: Sorgeloos P, Bengtson D A, Decleir W, Jaspers E eds.Artemia Research and Its Applications, Vol. 3, Ecology, Culturing, Use in Aquaculture. Universa Press, Wetteren.p.27-63.
Liang P, MacRae T H. 1999. The synthesis of a small heat shock/α-crystallin protein in Artemia and its relationship to stress tolerance during development. Developmental Biology, 207(2): 445-456. DOI:10.1006/dbio.1998.9138
Lopes F L, Desmarais J A, Murphy B D. 2004. Embryonic diapause and its regulation. Reproduction, 128(6): 669-678. DOI:10.1530/rep.1.00444
Nambu Z, Tanaka S, Nambu F. 2004. Influence of photoperiod and temperature on reproductive mode in the brine shrimp, Artemia franciscana. Journal of Experimental Zoology Part A:Comparative Experimental Biology, 301A(6): 542-546. DOI:10.1002/jez.a.80
Poprawa I. 2005. The ovary structure, previtellogenic and vitellogenic stages in parthenogenetic species Dactylobiotus dispar (Murray, 1907) (Tardigrada:Eutardigrada). Tissue and Cell, 37(5): 385-392. DOI:10.1016/j.tice.2005.06.003
Provasoli L, Pintner I J. 1980. Biphasic particulate media for the parthenogenetic Artemia of Séte. In: Persoone G, Sorgeloos P, Roels O, Jaspers E eds. The Brine Shrimp Artemia, Vol. 2, Physiology, Biochemistry, Molecular Biology. Universa Press, Wetteren. p.231-238.
Triantaphyllidis G V, Poulopoulou K, Abatzopoulos T J, Pinto Pérez C A, Sorgeloos P. 1995. International study on Artemia XLIX. Salinity effects on survival, maturity, growth, biometrics, reproductive and lifespan characteristics of a bisexual and a parthenogenetic population of Artemia. Hydrobiologia, 302(3): 215-227.
Versichele D, Sorgeloos P. 1980. Controlled production of Artemia cysts in batch cultures. In: Persoone G, Sorgeloos P, Roels O, Jaspers E eds. The Brine Shrimp Artemia, Vol. 3, Ecology, Culturing, Use in Aquaculture. Universa Press, Wetteren. p.231-246.
Wagner T L, Villavaso E J, Willers J L. 1999. Diapause in the boll weevil (Coleoptera:Curculionidae):life-stage sensitivity to environmental cues. Annals of the Entomological Society of America, 92(3): 396-402. DOI:10.1093/aesa/92.3.396
Wang Z C, Asem A, Sun S C. 2017. Coupled effects of photoperiod, temperature and salinity on diapause induction of the parthenogenetic Artemia (Crustacea:Anostraca) from Barkol Lake, China. North-Western Journal of Zoology, 13(1): 12-17.
Zhang L, Lefcort H. 1991. The effects of ploidy level on the thermal distributions of brine shrimp Artemia parthenogenetica and its ecological implications. Heredity, 66(3): 445-452. DOI:10.1038/hdy.1991.54