This is not an up-to-date or comprehensive description of fairy shrimp. It is based mostly on physical publications I read and copied decades ago. Digital journal articles now are mostly behind pay or registration walls.
This page is best treated as an introduction to a group of animals with biological features unlike the mammals, other vertebrates, and insects familiar to most of us. It is intended to provide the fairy-shrimp searcher with enough information to search intelligently and to appreciate the remarkable ecological adaptations of these animals.
For citations, go to References.
Anatomy
Food
Predators
Habitats
Life Cycle
Resting Eggs
Other Crustaceans You May Find With Fairy Shrimp
Anatomy
Because fairy shrimp are small (mostly less than 2.5 cm, or 1″, long) and constantly in motion, their anatomy is not obvious to a casual observer. What to call the body parts you do see may also not be obvious. Linder (1941, p. 107), Moore (1969, p. R177), and Kaestner (1970, p. 86) have good, and Thorp and Covich (2001, p. 892 and 900) less good, diagrams of complete bodies with labeled parts but I won’t bother to ask for permission to reproduce them here. Lynch (1960) and Lynch (1964) have good diagrams but lack labels.
This section applies only to adult fairy shrimp. Pre-adult (naupliar) stages are morphologically very different. They are also mostly too small to see.
The fairy shrimp body has a head, a thorax, and an abdomen. Because fairy shrimp swim head first, there is rarely any doubt as to where the head is. The details are less apparent. The head is the short front end of the body that lacks legs. Its most prominent visual feature is a pair of compound eyes (black in most of the species I have seen) on short stalks that stick out laterally from the head. On pale fairy shrimp, you may be able to see a tiny black spot on the front of the head midway between the eyes. This is the ocellus (Dexter, 1959, p. 559) or naupliar (pre-adult) eye (Belk, 1982). It is sensitive to light. A third eye indicative of wisdom figures prominently in Hindu symbolism. Fairy shrimp actually have a third eye but I’m not so sure about the wisdom. The head also has a maxillary gland, mandibles, and maxillae, which all figure in eating (Kaestner, 1970, p. 86), but you would probably have to dissect one to see them.
A male fairy shrimp (Branchinecta coloradensis) with body parts annotated.
The head has 2 pairs of antennae, here referred to here as antennae I and antennae II. Antennae I are smaller and more toward the front of the head than antennae II. They have only one segment, or joint. They are straight and extend forward so they look like antennae. Antennae II are sexually dimorphic. In females, they are small and inconspicuous although fatter than antennae I. In males, antennae II are quite large and have developed as a means to grasp females before and during copulation. There are such great interspecies variations in the male antennae II that they are often the most important feature used for species identification (Belk, 1975). They have 2 segments (except Polyartemiella – Belk, 1975). The basal, or proximal, segment is fatter and mostly longer than the distal, or apical, segment and is generally more or less cylindrical. It may have a mound, pad, pulvillus, ridge, or bulge with or without a spiny roughness, or a peg-like or finger-like process or outgrowth, or some combination of these (Linder, 1941, p. 127-144; Belk, 1975). A microscope is required to see such features. The distal segment may be cylindrical, flattened, bladed, or broadened and may be curved, kinked, or twisted. It may taper to a point, terminate bluntly, or end with a feature that is shaped like a human foot, or something else. In normal swimming, antennae II project from the head back over the legs (fairy shrimp swim on their backs). In some species, the pair of antennae II may cross above the legs 1/2-3/4 of the way down the thorax. For species with large antennae II, you may be able to see them and get a rough idea of their form. If you take a photograph and enlarge it later, you will get a better idea.
But that’s not all, males of some species have antennal appendages. These grow from the basal segments of the male antennae II. They may be longer or shorter than antennae II. Thicknesses and cross-sections vary between species and the appendages have spines, outgrowths, fingers, branches, processes, or projections which defy a short description (Linder, 1941, p. 127-144; Belk, 1975; Pennak, 1978, p. 342-343). Belk (1975) distinguished horn-like, lamelliform, and triangular appendages. Some species have a single frontal appendage, which is interpreted as the fusion during early development of the antennal appendages of each antenna II (Belk, 1982). It emerges from the front of the male’s head. This suggests some species of fairy shrimp could be considered real life unicorns although it is unlikely that the original conception of unicorns came from fairy shrimp. A frontal appendage can be as complex as the antennal appendages but I haven’t found any good descriptions or diagrams.
Something as bizarre and complex as antennal appendages reeks of sexual selection but this has apparently attracted little research interest. Belk (1984) demonstrated that males that had had the antennal appendages snipped off were not as successful at mating as males with intact antennal appendages. He concluded the antennal appendages play a “role in mate recognition”. Sexual selection of antennal appendages suggests that male antennae II are also subject to sexual selection in species without antennal appendages as variations of male antennae II are unlikely to serve feeding or other survival purposes. Yet, I have not come across studies of how females and males use visual, chemical, or tactile senses to recognize potential mates or of why the vast majority of mating attempts fail (e.g., 79% in Belk’s 1984 study).
The thorax begins with the first segment behind the one with the second maxillae (Linder, 1941, p. 112). Visibly, it begins with the first pair of legs. In all but 2 genera, the thorax consists of 11 segments, each with a pair of legs. The Polyartemiella thorax has 17 segments and the Polyartemia thorax has 19 (Linder, 1941, p. 107). The legs are always in motion so the thorax generally looks fuzzy or blurred. When photographing fairy shrimp, a shutter speed of 1/125 second will just about freeze the leg motion. For the microscopist, thoracic segments may have spines, warts, or ridges, depending on species (Linder, 1941, p. 114). The thoracic segments and legs are generally all alike but Linder (1941, p. 112) found that Parartemia females lack legs on the 11th segment and Pennak (1978, p. 328) stated that the rearmost pairs of legs are smaller.
The legs are complex structures with multiple lobes, flaps, and irregularities along their margins, which have names such as endites, endopodites, epipodites, exopodites, exites, and pre-epipodites (Dexter, 1959, p. 550; Kaestner, 1970, p. 86-87). They are not jointed like the legs of familiar crustaceans, such as crabs and lobsters. Parts of the legs are fringed with arrays of bristles that are important for feeding. Gas exchange occurs principally through the legs so legs serve the same purpose as gills. The legs are stiffened by internal fluid pressure rather than by a stiff integument (Kaestner, 1970, p. 85), or outer covering. Fairy shrimp legs lack the chitin or calcification found in other arthropods.
Male fairy shrimp on its side in clayey water that makes details of legs visible. They are not stiff or jointed like crab legs.
The legs have a simple motion, beating back and forth at 140-400 beats/minute (Kaestner, 1970, p. 91). The legs move in wave-like pulses progressing from the head to the abdomen. Most of the motion is for collecting food particles and pushing them into the ventral food groove and up toward the mandibles (Kaestner, 1970, p. 90-92). Forward body motion, in contrast, is produced primarily by propeller-like motions of the terminal lobes of the legs (exopodites), which can be inclined at various angles (Pennak, 1978, p. 329). Moore (1969, p. 178) referred to experiments that showed that anostracans can regenerate their legs, but with a very low success rate.
The relatively slow, continuous swimming motion of fairy shrimp is helpful in identification. I haven’t seen anything else swim like fairy shrimp do. Clam shrimp and tadpole shrimp swim at about the same speed but they stop from time to time (within a minute or 2?). Fairy shrimp never stop unless they are scraping the bottom of the pond with their legs. Insects have much shorter swimming spurts, probably less than 20 seconds or so. Insects can be relatively slow, like dytiscid larvae (order Coleoptera, family Dytiscidae), but most seem to be fast, like water boatmen (sub-order Heteroptera, family Corixidae). Amphipods are also fast swimmers and the ones I have seen don’t swim for more than about 30 seconds without stopping, or disappearing from my view. Ostracods are continuous swimmers (at least for as long as I can track them, which may be only 30 seconds or so) but they are fast and look like tiny clams. A nearly spherical animal with a continuous swimming pattern that is a little faster than fairy shrimp is not likely to be confused with them because they are less than 3 mm in diameter. I think such spherical animals may be water mites.
The abdomen is long and thin and has no legs. It looks like a tail to a casual observer, such as me. It consists of 9 segments (8 for Polyartemiella and Polyartemia). There are 2 fused segments at the front and the last segment is called a telson, or furca, and has the anus (Linder, 1941, p. 107; Kaestner, 1970, p. 85). The 2 fused segments comprise the genital region. In males, they have 2 penes (left and right) and in females, they have 2 oviducts which lead to a single ovisac (Belk, 1982). The telson has 2 short cercopods, which are also referred to as uropods, caudal appendages, or rami (Belk, 1975; Belk, 1982). They are stick out from the end of the abdomen to give it the appearance of a forked tail in some species. The cercopods may be rod-like and pointy or flattened and flap-like (Belk, 1975). The pointy ones may be straight or curved. They are generally fringed with spines which are usually too small to see.
The abdomen may be longer or shorter than the thorax (Linder, 1941, p. 116) but the lengths do not differ by much. The intestine extends the full length of the abdomen as a thin tube. Because fairy shrimp bodies are usually translucent, the food matter in the intestine can make the fairy shrimp more visible if it is darker than pale water or paler than dark water. The ovisac of females attaches to the forward end of the abdomen and floats over the abdomen during swimming (fairy shrimp usually swim on their backs). Like everything else, the ovisac is translucent and hard to see unless it contains eggs. Eggs of a different color than the water help make fairy shrimp more visible. In some species, the ovisac is a long pouch which extends for more than half the length of the abdomen. In others, it is a short, globular pouch at the upper end of the abdomen. Although not commonly used for species identification, different types of ovisacs can alert you to the presence of different species.
A female fairy shrimp (Streptocephalus seali) with body parts annotated and abundant eggs.
A female fairy shrimp (probably genus Artemia) with a bulbous or globular ovisac and indistinct eggs.
Quiz: Identify which of these 2 fairy shrimp is female and which is male. Can you see the ocellus on each?
Fairy shrimp have a heart! But it’s a tube (Pennak, 1978, p. 331), like most crustaceans I guess. It lies between the intestine and the back and extends through most or all of the body segments (Pennak, 1978, p. 331). Pennak (1978, p. 331) wrote that the heart is much shorter in Notostraca and Conchostraca so it may be safe to say that fairy shrimp are the most big-hearted of branchiopods. Fairy shrimp are pretty small so they don’t need blood vessels. The blood circulated by the heart, then, isn’t really blood, it’s just the intercellular fluid that fills the spaces between the cells of the body. It is also called hemolymph or extracellular fluid.
One of the functions of the intercellular fluid is to distribute oxygen to the cells, like a vertebrate circulatory system. Respiration occurs through the branchia, also known as exites, of the legs, which are small finger-like lobes (Fig. 4-2 in Kaestner, 1970, p. 87). Pennak (1978, p. 332) added that gas exchange “probably takes place through all exposed surfaces of the body”. Oxygen diffuses from the water through the branchia and into the intercellular fluid. A figure in Kaestner (1970, p. 86) shows “blood” flowing within the legs, from the forward legs toward the rearward legs, and from each segment into the centrally located heart. There are openings in the septa between the body segments for the heart. Pennak (1978, p. 332) also mentioned “slitlike ostia”, which are the openings within the segments into the heart. Peristaltic contractions push the intercellular fluid forward through the heart into the head (Pennak, 1978, p. 332), where it is just kind of dumped. Neither Kaestner (1970) nor Pennak (1978) mentioned heart rate.
Fairy shrimp (and other branchiopods) are not exactly red-blooded but they do have hemoglobin. They certainly do not have blue intercellular fluid like crabs and many other crustaceans. The blue bloods have hemocyanin rather than hemoglobin for carrying oxygen. In high-TDS water or water with a low oxygen concentration, fairy shrimp produce more hemoglobin and become pink or red (Kaestner, 1970, p. 93). Most of the studies have been done on fairy shrimp of the genus Artemia, which live in high-TDS water and are commonly a shade of red. I have seen both red and nearly colorless Artemia. Sanchez and others (2016) found red and colorless Artemia in the same high-TDS waters (140,000-200,000 mg/L) and determined that the red ones were infected by cestode parasites but the colorless ones were not. The parasites stimulated the production of both hemoglobin and orange carotenoids in the Artemia. Rarely, fairy shrimp I have seen in low-TDS waters are a shade of orange or pink, such as those in Dead Ant Rock Pool and those in Small Rare Plant Habitat Pond. What does that say about the very much more common colorless fairy shrimp? Have they not bothered to synthesize much hemoglobin?
Red fairy shrimp of the genus Artemia with pale food in their intestines. That the water has high TDS can be inferred from the presence of white mineral crusts along the shore.
Essentially colorless fairy shrimp (probably of the genus Artemia) in high-TDS water. Dark bulbous ovisacs or pale food in the intestines make some fairy shrimp easier to see. In others, the intestine along the back is dark. A female at upper left has a white ovisac. Note the bright reflections from recently precipitated mineral crystals floating on the surface of the water. TDS must be very high for minerals to precipitate.
Fairy shrimp do not have a brain but they have 2 parallel nerve cords that go all the way from the eyes to the end of the abdomen. The diagram in Pennak (1978, p. 333) shows a pair of fat nerves connecting the left and right nerve cords within each thoracic segment, pairs of small nerves extending laterally from the nerve cords outward into each thoracic segment, occasional wispy nerves extending from the nerve cords into the abdominal segments, and a thick mass of nerves connecting the eyes and ocellus to the nerve cords. Pennak (1978, p. 333) also mentioned a modified patch of “cephalic epithelium” with a “probably sensory” function known as the “dorsal organ”. Maybe fairy shrimp have something of a brain after all.
The eyes and ocellus both sense light and are used to orient the body with the back side down. Fairy shrimp will orient back side up if light comes from below. In darkness, fairy shrimp orient back side down (Pennak, 1978, p. 332). The sense of taste is localized near the mouth parts and the sense of touch is distributed throughout the body (Pennak, 1978, p. 332). What they taste and what they feel remains speculative. They have sensory bristles (“setae”) and “hairs” for touch, or sensing water motion, and chemical sensing (Kaestner, 1970, p. 87). No word on what chemicals are sensed.
I have not read articles concerning the role of vision in fairy shrimp living in water that is opaque due to suspended clay. Is their vision more sensitive because it is harder to see or have the vision neurons been co-opted by another sense that is more useful?
When the water is this opaque, what can fairy shrimp see? Are they using other senses to navigate, to keep from swimming into each other, and to find mates? Look for the dark abdomens in this photograph.
Fairy shrimp can see you. Rogers and Fugate (2001) reported that Branchinecta hiberna in one pond avoided humans so strongly that they had to be “stalked” in order to be captured. In my experience, they may or may not move away when you approach and the water is clear enough for you to see them. If you stay still and watch, they will carry on as if you are not there in most cases. Does my ability to take photographs within 0.5 m (20″) of fairy shrimp indicate that they have poor vision or that they do not see me as a threat? In some populations, such as the one in Bivouac Lake (Wind River Mountains), the fairy shrimp swim away from an approaching finger. In opaque water, you might be able to touch them but you won’t know because they offer too little resistance to feel.
Food
Fairy shrimp eat bacteria, algae, protozoans, rotifers, diatoms, and small organic particles (Moore, 1969, p. R178; Kaestner, 1970, p. 90; Pennak, 1978, p. 330; Belk, 1982). Although branchiopod specialists haven’t mentioned fungal spores as possible food, mycologists have (e.g., Strullu-Derrien and others, 2016). There are two predatory species of fairy shrimp. They eat fairy shrimp of another, smaller species (Belk, 1982). Copepods, fairy shrimp eggs, and organic detritus have also been found in the guts of predatory fairy shrimp (Kaestner, 1970, p. 92; before the second one was identified). As filter feeders, fairy shrimp obtain food by swimming through the water and straining small food items out of the water with the bristles on their legs. Under certain conditions, fairy shrimp turn ventral side down and scrape pond bottom surfaces with their legs (Belk, 1982). This has been interpreted as feeding but how do we know? Maybe they are sharpening their bristles as cats “sharpen” their claws. Maybe they need resistance for their exercising.
The digestive system is simple in the sense that there are not a lot of organs and glands involved but feeding is not mechanically simple. Food particles are strained non-selectively by the bristles on the legs. Food collects in a central groove running the length of the thorax and open to the water. Dexter (1959, p. 561) called the material in the groove a “mucilaginous string”. The food is passed forward up the groove by lobes or bumps on the lower parts of the legs according to Pennak (1978, p, 330) but he doesn’t explain how. Food passes over 2 pairs of maxillae and a pair of mandibles before entering the mouth (Dexter, 1959, p. 559). The mandibles are large, relative to body size, and vertical but the grooved chewing surfaces are bent horizontally over the ventral groove (Belk, 1982). A finger-shaped labrum serves as a lip over the ventral groove next to the mouth (Linder, 1941). Mastication is accomplished by grinding between the mandibles outside of the mouth. Larger particles or excess food is somehow sloughed off (Kaestner, 1970, p. 92; Pennak, 1978, p. 330). I have seen some cases of coarse particles being cast off by leg motion before reaching the head. Mucus is added to the food under the labrum before entering the mouth (Kaestner, 1970, p. 92). From the mouth, food enters a short fore-gut, or esophagus, and proceeds through a globular stomach before leaving the head and entering the intestine. The intestine is a narrow tube that extends all the way through the thorax and abdomen to the anus (Dexter, 1959, 559; Belk, 1982). You may occasionally see a string of food material that has left the anus but hasn’t yet dispersed in the water. Feeding is continuous (Pennak, 1978, p. 330) and a normal consequence of swimming but it is not always the same. Fairy shrimp in opaque, clay-rich waters in Nevada sometimes lie motionless while vigorously churning their legs to collect more food (see Macari East Fairy Shrimp Video 2022-01-31a-r under Clay Dancers of “Carson Lake” Playa, Fairy Shrimp Videos). No word on digestive enzymes.
How fairy shrimp use their legs to strain and push food forward toward the mouth while pushing water backwards to swim would seem to be a remarkable accomplishment. Kaestner (1970, p. 91-92) provided a diagram and described the process in some detail but it is nonetheless hard to grasp. It involves suction. Walossek (1993, p. 60-61) also diagrammed and described the process which he inferred to occur in fossil Rehbachiella. His Figure 38 offers an understandable explanation. Looking down, the legs are convex toward the front of the animal and bristles extend from the inner ends of the legs over the food groove (principally rearward in the figure). When the outer edge of the leg moves toward the rear to push water back, the bristles on the inner edge move forward in a levered or rotational motion. If bristles extended into the food groove, they could physically push food forward but they apparently don’t. Instead, just as water is pulled into the space behind a paddle being pushed backward, water and food in the food groove could be pulled forward by the bristles. Walossek (1993) explained it differently though. In his view, water is pulled from the inner edges of the legs through the spaces between the legs and is expelled between the outer edges of the legs during a “stroke” and this action pulls the food forward. How food motion is related to the continuous propeller-like motion of the exopodites involved in swimming (see Anatomy section), or not, is not addressed.
Food motion between the legs can be seen in Macari East Stop Sign Pond Fairy Shrimp Video 2022-01-31a-r. In slow motion, the video shows clay between the legs being pushed toward the head of the fairy shrimp as the legs move forward and then being pushed upward and outward toward the rear of the fairy shrimp as the ends of the legs sweep backward. It’s not hard to imagine food moving forward as the clay between the legs moves forward due to a sort of horizontal twisting leg motion, like the rotational motion suggested above. The forceful expulsion of clay from between the legs on the back stroke could create suction that would also pull the food forward, as in Walossek’s (1993) explanation. If food moves forward on both strokes of the legs, that would be quite ingenious. The Beauty Peak East Pond videos for 2021-04-30 (Bodie Hills) show similar behavior.
Does clay have any nutritional value? In the clay-rich waters that some fairy shrimp inhabit, the food stream likely contains a considerable amount of clay as the clay is too fine to be ejected from the food groove. Considerable water and clay, along with food, enters the foregut. I haven’t found any discussion of that. I have seen intestines with very pale material that is suggestive of a significant clay content.
Two habitats where fairy shrimp commonly live are considered to have low productivity and have been given distinctive names. The habitats have low concentrations of phytoplankton (used here as a synonym for photosynthetic algae and bacteria) and, hence, less food for grazers like fairy shrimp. One such habitat is alpine lakes. They have clear waters with low concentrations of phytoplankton. Such waters are referred to as oligotrophic (oligo=little, trophic=nourishment – New Webster’s Dictionary). Playa lakes and ponds on clay flats also have low concentrations of phytoplankton. This is because suspended clay makes the water nearly opaque so sunlight does not penetrate far. That leaves only a small volume of water where photosynthesizers can survive. Such lakes are referred to as argillotrophic (argos=white, which clay commonly is). Fairy shrimp do fine in these habitats.
In fact, fairy shrimp are more common in low productivity lakes than in high productivity eutrophic lakes. Eutrophic lakes have abundant phytoplankton. In some cases, they have so much phytoplankton and plants that their decomposition reduces oxygen concentrations to levels most animals cannot tolerate. Eutrophic lakes may also have abundant rooted or floating multicellular plants that impede filter feeding and prevent colonization by fairy shrimp. The North American Lake Management Society, which has a name that suggests it knows what it writes about, made a long argument about why this old lake classification scheme is unworkable and misleading (“Defining Trophic State” at www.nalms.org). Nonetheless, if you read of oligotrophic or argillotrophic lakes in the literature, recognize that they indicate potential fairy shrimp habitat. Keep in mind though that oligotrophic lakes commonly have fish. This digression begs 2 questions: 1) how much bacteria, algae, and protozoans are needed to support a fairy shrimp population? and 2) how do you recognize lakes with such low phytoplankton concentrations that fairy shrimp populations cannot survive?
Kaestner (1970, p. 90) pointed out that fairy shrimp often occur in ponds with no other macroinvertebrates and consequently no food competition. However, fairy shrimp do occur with other herbivores or omnivores in many cases. Oftentimes, these are other branchiopods, such as tadpole shrimp, clam shrimp, or cladocerans. Maeda-Martinez and others (1997) found that of 134 ponds in northern Mexico and 166 in Arizona, 45% had more than 1 species of fairy shrimp, tadpole shrimp, or clam shrimp. Copepods, another common herbivore, also occur with fairy shrimp. To what extent does food competition limit fairy shrimp population sizes or even colonization?
Abundant fairy shrimp and cladocerans caught in one sweep of the net through an argillotrophic lake. The cladocerans are almost transparent, lentil-size blobs with black spots. They are particularly abundant in the left part of the net.
Food abundance no doubt plays some role in controlling fairy shrimp populations. However, because the food is microscopic, its abundance is not obvious to the observer. Sometimes fairy shrimp are so abundant that it is hard to imagine how they get enough food (e.g., the abundant red fairy shrimp in the photograph Steamboat Lake 2nd East Pond 1989-06, #0205, Granite Mountains, shown in the Anatomy section above).
Boom-and-bust cladoceran population dynamics may apply to some fairy shrimp populations. Populations of the cladoceran genus Daphnia typically increase rapidly during a spring algal bloom in northern temperate climates, up to 100-fold, but then diminish during a clear-water phase after most of the phytoplankton have been eaten. The reproductive rate also drops (Thorp and Covich, 2001, p. 867). The population collapse is due to diminished food supply as well as predation, which affects different cladoceran species differently depending on size, among other factors (Thorp and Covich, 2001, p. 867-868). This scenario would require a relatively long-lived pond.
High density of fairy shrimp in a small, alpine pond that has experienced a drop in water level over the summer. The water is clear and an abundance of food is not obvious. The fairy shrimp are also competing with probable copepods and possible ostracods.
Predators
The list of predators which eat fairy shrimp is long. Fish are at the top. Fairy shrimp are easily gobbled up by fast-swimming fish. Unfortunately for fairy shrimp, the natural distribution of fish has been greatly expanded by humans. Humans have stocked and continue to stock fish in just about every body of water that could possibly support fish and some that can’t. No doubt, many fairy shrimp populations have been wiped out by fish-stocking.
Beyond fish, Moore (1969, p. R178) listed tadpoles, salamanders, ostracods, and insect larvae. Pennak (1978, p. 337) further specified insect larvae of the order Coleoptera, family Dytiscidae (predacious diving beetles or dytiscids), those of the order Trichoptera (caddis flies), and “perhaps a few others”. The other Moore, Moore (1963), identified amphipods, Odonata nymphs (dragonflies, damselflies), and Notonectidae (backswimmers) and Coleoptera (families not specified) larvae as predators. Kaestner (1970, p. 90) stated that “caddis flies” (not larvae?) eat fairy shrimp.
A big salamander at the bottom of Bull Canyon Pond (Antelope Hills). Salamanders eat fairy shrimp. Bull Canyon Pond had fairy shrimp at the time.
A salamander larva with gills swimming in yellowish water of Bull Canyon Pond. Salamander larvae may be more likely to eat fairy shrimp than adult salamanders because they are always swimming around. Bull Canyon Pond had fairy shrimp at the time.
3 frogs (probably Pseudacris regilla) that still have tails are now motionless on the pond bottom hoping I can’t see them. The head of the paler one in the sunlight is masked by a grass stem above the water. Other frogs at this pond (Virginia Pale Green Pond, East-Central Sierra Nevada) have lost their tails and are hopping around in the grass. Tadpoles and sub-adult frogs eat fairy shrimp. This pond doesn’t have fairy shrimp.
A northern leopard frog (Rana pipiens) in “Coyote Lake” Antelope Hills which has fairy shrimp.
Caddisfly larvae eating fairy shrimp seems unlikely to me. Thorp and Covich (2001, p. 754) described caddis fly larvae as bottom-feeding “grazers”. The 2 caddisfly families that are predacious live only in streams (Thorp and Covich, 2001, p. 680). The caddis fly larvae I have seen are significantly smaller than fairy shrimp and crawl along the pond bottom with their heads barely poking out of a cumbersome, tubular case of sand grains or bits of vegetation that they have glued together to enclose their bodies. They can’t swim. How could they catch even slow-swimming fairy shrimp?
This is a caddisfly larva in a case made of bits of twig. Its head is at the left end of the case, where one protruding leg can be seen. The case looks too cumbersome for these larvae to be a threat to swimming fairy shrimp.
The 4 pale, tubular objects are cases of caddisfly larvae (order Trichoptera) which have been constructed of sand grains. The larvae are actively climbing up the rocks and occasionally rolling down so they can climb up again. Do they look like they could eat fairy shrimp?
Ostracods also seem like unlikely predators as adults are usually less than 3 mm long. Perhaps early hatching ostracods eat fairy shrimp nauplii before they grow to be more than a millimeter or two long. According to Thorp and Covich (2001, p. 827), ostracods are “predominantly herbivores and detritivores”. A few species are carnivorous but I don’t know if the carnivorous species listed by Thorp and Covich (2001, p. 827) occur in fairy shrimp habitats.
Amphipods are fast-swimming, shrimp-like crustaceans with hard exoskeletons, 2 pairs of antennae, and 10 leg-like pereopods and pleopods. They are commonly known as scuds or sideswimmers (Thorp and Covich, 2001, p. 788-789). They belong to the superorder Peracarida and while juvenile peracaridians eat algae and bacteria, adults are more opportunistic (Thorp and Covich, 2001, p. 792). In ponds without fish, amphipods consume “phytoplankton and zooplankton”. “[I]ntraguild predation and cannibalism are also thought to be relatively common” in some species of the genus Gammarus (Thorp and Covich, 2001, p. 792). These characteristics suggest amphipods could be significant predators of fairy shrimp. In Wyoming ponds, I more commonly saw amphipods without fairy shrimp than amphipods with fairy shrimp. Denton Belk did not support my hypothesis that amphipod predation could eliminate or prevent fairy shrimp colonization.
Amphipods are usually swimming so fast they can’t be photographed but this one has stopped to try to eat a large ant (round, reddish-black head below the amphipod’s head). Such opportunistic predation suggests amphipods would eat fairy shrimp when available.
In their descriptions of aquatic insects by order and family, Thorp and Covich (2001) did not identify food sources in sufficient detail to know for sure whether they eat fairy shrimp but they did identify predators. By order:
- Odonata (dragonflies, damselflies) larvae eat “mostly other insects” (Thorp and Covich, 2001, p. 669).
- Plecoptera (stoneflies) includes a few species with predacious larvae but they live in streams (Thorp and Covich, 2001, p. 672).
- Megaloptera (fishflies, alderflies) larvae are predators and “often attain a very large size” (Thorp and Covich, 2001, p. 680-681).
Of the order Hemiptera, sub-order Heteroptera families:
- Belostomatidae (giant water bugs) adults are “powerful predators” that can take fish and frogs (Thorp and Covich, 2001, p. 684).
- Notonectidae (backswimmers) larvae and adults pursue or ambush “other invertebrates and small vertebrates” (Thorp and Covich, 2001, p. 685).
- Pleidae (pygmy backswimmers) adults and nymphs are <3 mm (0.1″) long and feed on “small invertebrates” (Thorp and Covich, 2001, p. 685).
Of the order Coleoptera families:
- For the Dytiscidae, the common name – predacious diving beetle – apparently says it all because Thorp and Covich (2001, p. 688) did not mention what they eat.
- Gyrinidae (whirligig beetles) larvae feed on other invertebrates (Thorp and Covich, 2001, p. 689).
- Hydrophilidae (water scavenger beetles) larvae are all predators (Thorp and Covich, 2001, p. 692).
Dytiscid larvae look particularly nasty as they have a broad abdomen that tapers and curves up over their bodies like scorpion tails and they have large, curved pincers (mandibles) on each side of the head. They can be larger or smaller than fairy shrimp. Notonectid adults (backswimmers) are fast swimmers and although they have smaller bodies than most adult fairy shrimp, they have piercing mouth parts that can kill one. These larvae are the most common predators I have seen.
Small, young dytiscid larva on a leaf. In this view, the head with antennae and pincers can be clearly seen but they are not fearsome yet. The pond had fairy shrimp at the time.
A dytiscid larva at center has a broad, pale abdomen, with forked tail-like features at the end that is curved up toward the water surface. Its head, on the pond bottom, is obscured by a small plant stem. The fairy shrimp do not appear to be avoiding the dytiscid larva but that may be partly due to the high density of fairy shrimp and partly to the fact that the larva is not moving.
A backswimmer at center is swimming with its back characteristically down and its pair of long, oar-like legs angled forward. The body is a little less than 10 mm (0.4″) long. There are also several brown copepods in the photograph. Backswimmers eat fairy shrimp. The copepods may be too small to be of interest to the backswimmers or there may be so many copepods that there aren’t enough backswimmers to eat them all. Upper South Fork Pine Creek Pond Toquima Range does not have fairy shrimp.
This beetle has long legs for a rowing-like motion like backswimmers but it is a water boatman (Family Corixidae, sub-order Heteroptera). It is darker colored than backswimmers and swims with its back up. Unfortunately, when it is hard to tell whether the legs are above or below the body and what color the animal is, whether due to waves or turbidity or molting, an accurate identification is iffy. The oar-like legs are not enough. Water boatmen feed primarily on algae and protozoans but may occasionally eat a mosquito larva (Thorp and Covich, 2001, p. 684). The family is not a problem for fairy shrimp.
Dexter (1959, p. 561) maintained that fairy shrimp “can usually withstand predation from amphibians and carnivorous insects”. However, of 24 ponds sampled by Dodson (1970), 15 contained salamanders and no fairy shrimp and 8 contained fairy shrimp and no salamanders. No ponds had both.
Birds are important predators of fairy shrimp even though Kaestner (1970) and Pennak (1978) did not mention them. Stilts and avocets “can be significant consumers” of fairy shrimp of the family Artemiidae in Australia (Timms, 2012). Alaska Ecology Cards (available online), produced by the Alaska Division of Wildlife Conservation, listed ducks and phalaropes in addition to water shrews and “diving beetles” as predators of fairy shrimp. Proctor (1964) hatched fairy shrimp from eggs found in duck feces (cited by Thorp and Covich, 2001, p. 896). Van Stappen (1996) reported that waterfowl, “especially flamingos” “are the most important natural dispersion vectors for Artemia fairy shrimp” and that eggs survive in the digestive tracts of birds “for at least a couple of days”. Artemia are recognized as an important food source for waterfowl at Great Salt Lake (Utah Department of Environmental Quality, 2011).
Analysis of the stomach contents of birds at “Mono Lake” found that 92% was fairy shrimp in 13 California gull chicks, 7% was fairy shrimp in 5 Wilson’s phalaropes, and 4% was fairy shrimp in 21 northern phalaropes (Winkler, 1977, Fig. 5-3, p. 96). Jones & Stokes Associates (1993c) reported fairy shrimp account for 90% of the diet of eared grebes at “Mono Lake” in late summer (p. 3F-23) and an unknown percentage of the diet of Wilson’s phalaropes (p. 3F-31). Fairy shrimp were 57% of the diet of California gull chicks in another study reviewed by Jones & Stokes Associates (1993a, p. C-13).
If you are not interested in math, skip the following digression on conditional probability (Skip).
Simple field observations of birds in fairy shrimp ponds strongly suggest that birds eat fairy shrimp. Wading avocets and some ducks drag their bills back and forth through the water. Phalaropes peck at the water while swimming or, less commonly, while wading along the shore. Although correlation is not cause, I saw a suggestive association at Wyoming ponds between avocets, phalaropes, and fairy shrimp.
Antelope Hills
- avocets observed at 13 ponds, 8 of those had fairy shrimp
- phalaropes observed at 9 ponds, 4 of those had fairy shrimp
Granite Mountains
- avocets observed at 6 ponds, 5 of those had fairy shrimp
- phalaropes observed at 6 ponds, 5 of those had fairy shrimp
Great Divide Basin
- avocets observed at 8 ponds, 3 of those had fairy shrimp
- phalaropes observed at 3 ponds, 1 of those had fairy shrimp
Looking at the conditional probabilities for a single pond visit rather than for ponds (as above), the probability of finding fairy shrimp was 47% for visits to ponds with water in the Antelope Hills, Granite Mountains, and Great Divide Basin (51 out of 108 pond visits). That is not counting rock pools because I never saw birds at rock pools. Avocets improve your chances. The probability of finding fairy shrimp given that an avocet is present during the visit is 63%. Conversely, the probability of finding fairy shrimp if no avocet is present is 41%. But look anyway.
[19 visits had fairy shrimp and avocets (true positives) and 11 had avocets but no fairy shrimp (false positives). 46 visits had no fairy shrimp and no avocets (true negatives) but 32 had fairy shrimp and no avocets (false negatives).
probability of fairy shrimp, given avocets = probability of fairy shrimp with avocets / probability of avocets = 19/30, where the observed frequency is the estimate of probability]
These data are biased by my occasional failures to record data for ponds when I did not find fairy shrimp but there are more serious deficiencies. I could argue that the true conditional probabilities for avocets is better than what I calculated because I did not spend the whole day at a pond waiting for avocets and did not take into consideration avocets close by, such as avocets at “Scotty Lake” but not at North “Scotty Lake” West Pond when it had fairy shrimp or avocets at Eastern “Soda Lake” but not at “Soda Lakes” Far Eastern Pond when it had fairy shrimp or avocets at “Piaya Lake” but not at Little “Piaya Lake” when both had fairy shrimp. Alternatively, I could argue that the true conditional probabilities for avocets are worse than what I calculated because the avocets at Northeastern “Lewiston Lakes”, Cuesta Pond, and Chinook Pond were really there for the clam shrimp and not for the fairy shrimp. Or, maybe it is more about nesting preferences than food sources.
For phalaropes, it is close to a coin toss. The probability of finding fairy shrimp given that a phalarope is present during the visit is 53%. Conversely, the probability of finding fairy shrimp if no phalarope is present is 46%. The more important point is that it is worth keeping track of the clues.
[10 visits had fairy shrimp and phalaropes (true positives) and 9 had phalaropes but no fairy shrimp (false positives). 48 visits had no fairy shrimp and no phalaropes (true negatives) but 41 had fairy shrimp and no phalaropes (false negatives).
probability of fairy shrimp, given phalaropes = probability of fairy shrimp with phalaropes / probability of phalaropes = 10/19, where the observed frequency is the estimate of probability]
One phalarope is on the water to the right of center and 2 are along the shore (only the rear end of the one at left is visible). “Coyote Lake” Antelope Hills has fairy shrimp.
Several avocets wading and a few more are coming in for a landing (at right). Although avocets feed on fairy shrimp where available, they also help disperse the eggs. Separation Rim “Soda Lake” Great Divide Basin has fairy shrimp.
The ducks behind the fence post at center are standing in Labou North Playa Pond Fairview Valley which currently has fairy shrimp and apparently nothing else. These and other ducks have been dragging their bills back and forth through water less than 5 cm (2″) deep in what appears to be feeding behavior.
The birds floating on and standing in the water include ducks and maybe coots and the 2 birds with white markings at left are probably gulls. Rhodes Big Lake Rhodes Salt Marsh has fairy shrimp. Gulls at “Mono Lake”, 90 km (54 miles) to the west, eat lots of fairy shrimp (see above).
One problem with evolution is that all niches are eventually occupied. Fairy shrimp adapted to clay-rich water had probably lived peacefully for millions of years in ephemeral, opaque ponds that fish, most insects, and amphibians never adapted to. Except for dytiscid larvae (e.g., Beauty Peak East Pond, June 12, 2019, Bodie Hills) and maybe a few others, the niche for an aquatic predator that could live in the same water was largely vacant. So fairy shrimp did it to themselves. They evolved a giant fairy shrimp that can live in clay-rich water and primarily eats other fairy shrimp (also cladocerans, Daborn, 1977). 2 predator species are now known (e.g., Rogers and Hill, 2013). They may have evolved independently as they are not genetically close (Rogers and Aquilar, 2020). As yet, the fairy shrimp most tolerant of high-TDS water (i.e., the genera Artemia and Parartemia) are still free of predators which also live in the same water. Maybe the energy cost of osmoregulation has prevented the evolution of a giant Artemia species that can eat smaller Artemia.
Humans are also fairy shrimp predators. Dried fairy shrimp from the “Great Salt Lake” were used as food by local Native Americans (Pennak, 1978, p. 337). Now, humans collect fairy shrimp or eggs of the genus Artemia from salty lakes and salt works throughout the world. Dried and crushed fairy shrimp are used for fish and shellfish food. Eggs (marketed as cysts) are sold to those who feed the nauplii (hatchlings) to juvenile fish or shellfish larvae. In the larviculture of marine fish and shellfish, Artemia nauplii “constitute the most widely used food item” (Van Stappen, 1996). Artemia eggs have thus become a hot commodity with over 2,000* metric tons (2,200 U.S. tons) sold annually worldwide as of 1996 (Van Stappen, 1996). For a quick look at the breadth of the industry, scroll through search results for “artemia cysts”. Originally (1950s), the eggs came from “Great Salt Lake” but now they are also collected from salt works in San Francisco Bay, China, Siberia, and elsewhere. Artemia have been introduced into salt works in South and Central America, Australia, and Southeast Asia (Van Stappen, 1996). It seems likely that they have been introduced into natural lakes, too, but Van Stappen (1996) did not make that clear. Fairy shrimp evidently do not interfere with salt production.
*A variety of vendors claim a range of 200,000-300,000 Artemia nauplii per gram of cysts at high hatching rates. Hatching rates are less than 90% (as low as 70% for some products) so each gram actually has more eggs than the number of nauplii that are expected to emerge. Assuming 200,000 eggs per gram for a conservative estimate, 2,000 metric tons (2 x 10^9 grams) would have about 4 x 10^14, or 400 trillion, eggs. Just a kilogram (2.2 pounds) would have about 200,000,000 eggs. Do humans kill more than 400 trillion individuals of any other animal genus every year?
High variability in the nutritional characteristics of Artemia nauplii have caused problems for the larviculture industry. Bio-encapsulation methods have been developed to add unsaturated fatty acids, vitamins, chemotherapeutics, and vaccines (Van Stappen, 1996). This is apparently accomplished by adding these products to the water and capitalizing on the non-selective filter feeding and maybe respiration of the nauplii. Earth now has gazillions more Artemia fairy shrimp than ever before but most are fed to fish or shellfish larvae in human-controlled environments before they reach adulthood. Something to ponder the next time you come upon a high-TDS pond with a bunch of fairy shrimp or eat farmed “seafood”.
Humans also kill fairy shrimp for toxicology studies. The user guide for Thamnotoxkit F, “Crustacean Toxicity Screening Test for Freshwater” by Microbiotests, a Belgian company, can be found online. Thamnotoxkit F is an implementation of the ISO 14380:2011 standard for “Water quality – Determination of the acute toxicity to Thamnocephalus platyurus (Crustacea, Anostraca)”. I saw other test kits online that used Artemia.
The user guide, or “Standard Operating Procedure”, for Thamnotoxkit F is informative and well written. The procedure is pretty simple. Prepare a solution of “Standard Freshwater” by adding the contents of 5 vials of solutions with high concentrations of ions such as calcium and sulfate to deionized water. Add “Standard Freshwater” and more deionized water to a test tube with eggs, shake it, and pour it into a petri dish. Incubate the petri dish under constant illumination for 20-22 hours. Add various concentrations of the possible toxin, effluent, wastewater, etc., to small depressions on a test plate. Under a microscope, add 10 fairy shrimp nauplii to each test well using a micropipette. After 24 hours, count how many nauplii are moving and how many are not moving. The goal is to determine the concentration that kills 50% of the animals, or LC50. LC50 is a standard measure of toxicity that is used by regulatory agencies to regulate chemicals, compounds, and effluents. Results are more reliable if the test wells exhibit a range of mortality from 100% to 0%.
It may sound cruel to kill off 50% or more of newly hatched nauplii to achieve a measure of toxicity that is not easily translated into human risk. However, those that aren’t killed by the tested chemical are probably just flushed down the toilet or left to dry up on the test plate. No testing lab is in the business of raising fairy shrimp. It’s safe to say that none of the eggs in any Thamnotoxkit F or similar test kits ever grow to be mature fairy shrimp. So add the eggs for toxicology tests to the 400 trillion killed as fish or shellfish food each year.
A more benign use of Artemia eggs has been for sale as Sea Monkeys. They come with a book-size plastic aquarium and a salt packet. I tried this. The fairy shrimp can live for a few weeks until the food packet runs out. I didn’t know what else fairy shrimp could eat at the time. I have since learned that yeast can be used as food. The tiny aquarium looks awfully cramped. Better to get out of the house and search for fairy shrimp in the wild. It’s a lot more fun for you and for them.
Habitats
Distribution
pH
Temperature
Total Dissolved Solids
Turbidity
Osmoregulation and Thermoregulation
Colonization and Extirpation
Other Thoughts
Distribution
Fairy shrimp occur in ponds in a wide variety of habitats on all continents, including Antarctica (Belk, 1982). Individual species may be distributed on several continents or known in only a few ponds. Fairy shrimp inhabit deserts (e.g., Naceur and others, 2012) and tundra on the northernmost tips of land circling the North Pole (e.g., Lindholm and others, 2016). They range from coastal salt works in Australia (Timms, 2012) to alpine ponds at 3,800 m (12,500′) in the Rocky Mountains of Colorado (Shantz, 1905). The species in that Colorado lake also occurs in a rock pool 1.2 m (4 feet) across and 7.6 cm (3″) deep at 550 m (1,800′) elevation in Texas (Moore, 1950). In California alone, fairy shrimp are found in Central Valley grasslands at elevations less than 100 m (330′), in the Mojave Desert at 400-1,200 m (1,300-4,000′), and in the alpine terrain of the Sierra Nevada up to 3,500 m (11,500′) (Eng, Belk, and Eriksen, 1990). They occur in saline lakes with huge volumes, such as “Great Salt Lake”, and in rock pools of only a few liters (gallons) (Pennak, 1978, p. 337). They swim about in clear water as well as in opaque clay-rich water and in water colored yellow, brown, or red by organic acids (my observations and inference from Moore, 1963). Fairy shrimp are not confined to natural water bodies. They also occur in anthropogenic water bodies such as the salt works in San Francisco Bay and roadside ditches (Eng, Belk, and Eriksen, 1990). Vanschoenwinkel and others (2013) observed that “military domains in Eastern Europe” have proven to be “particularly suitable” as refuges for fairy shrimp and other large branchiopods due to “natural habitat that was historically set apart, unsealed roads with puddles and wheel tracks and regular disturbance by vehicles”.
Fairy shrimp do not occur in most of the larger ponds and lakes in the regions they inhabit because they cannot coexist with fish. Consequently, they are typically found in ponds not connected to streams and that are too small or too shallow (i.e., freeze solid in winter) to support fish, dry up in most years, or have TDS too high for freshwater fish. They typically do not even occupy all the fish-free ponds in an area. Their presence is related to a variety of factors that haven’t been fully elucidated. Some of these are discussed below.
In the ponds where fairy shrimp do occur, they don’t necessarily hatch every year. In the case of ephemeral ponds, fairy shrimp obviously can’t hatch when there is no water and some occupied ponds remain dry for a few years. However, even when ephemeral ponds do fill with water, fairy shrimp don’t always hatch. When Moore (1963) monitored a pond in Louisiana with populations of Streptocephalus seali and Eubranchipus holmani, he observed that S. seali but not E. holmani was present in May 1960 after the pond had dried out and refilled and that E. holmani but not S. seali was present from January through April of 1961. Both species occurred together in January through March of 1960. Moore (1963) suggested that low concentrations of dissolved oxygen prevented S. seali from hatching in 1961 when the pond warmed up to a suitable hatching temperature. Similarly, Donald (1983) monitored an ephemeral pond with 5 fairy shrimp species in Alberta for 14 years and observed that no species hatched in every year when water was present. Branchinecta lindahli failed to hatch in 6 years. Donald (1983) collected more than 100 specimens for identification in all but 1 year and was able to estimate species population sizes as small as 1% of all fairy shrimp.
My less frequent observations also indicate that ponds with fairy shrimp don’t have fairy shrimp every year. For example, I found fairy shrimp in “Coyote Lake” (Antelope Hills) in 1987 and 1993 but not in 1988 or 1989. Bull Canyon Pond, also in the Antelope Hills, had fairy shrimp in 1987 and 1995 but not in 1988. For ephemeral ponds like these, timing of filling or warm up could have been an issue but that is less likely the case for the alpine Little Sandy Overlook Pond (Wind River Mountains). In August 1987, it had fairy shrimp. In August 1989 and 2006, it did not. Similarly for Cinquefoil Pond, which had fairy shrimp in August 1993 but not in August 2006.
pH
There are physiological limitations on the physical and chemical conditions of ponds that fairy shrimp can live in and these vary between species. They play an obvious role in determining which ponds fairy shrimp live in but the large number of variables and the different requirements of different species seems to have prevented development of a comprehensive model.
Acidity, as measured by pH, has not been considered an important control on fairy shrimp habitat (e.g., Horne, 1967; Gonzalez and others, 1996). Nonetheless, the range of pH possibilities is worth noting and there may be important differences between species. For context, the U.S. Environmental Protection Agency has set a drinking water secondary standard of 6.5-8.5 for pH due to the possibility of bitter metallic taste and corrosion at less than 6.5 and of soda taste and mineral deposits at greater than 8.5. Orange juices have pH’s of 3.3-4.2 (wikipedia). Here are some examples of water pH associated with particular species (for multiple ponds, the range is given).
- 3-7.4….Parartemia acidiphilia in Western Australia and South Australia (Timms, 2012),
- 4.9-5.4…Streptocephalus seali in colored forest pools in southeastern Louisiana (Moore, 1963),
- 4.9-5.4…Eubranchipus holmani in colored forest pools in southeastern Louisiana (Moore, 1963),
- 5-7.7……Streptocephalus seali in California (Eng, Belk, and Eriksen, 1990),
- 5.5-6.5…Branchinecta dissimilis in California (Eng, Belk, and Eriksen, 1990),
- 6.1-6.5…Streptocephalus dorothae in southern Utah (Maynard and Romney, 1975),
- 6.2-8.1…Linderiella occidentalis in California (Eng, Belk, and Eriksen, 1990),
- 6.3-8.1…Branchinecta lynchi in California (table 2, Eng, Belk, and Eriksen, 1990),
- 5.5-7.0…Branchinecta lynchi in California (text, Eng, Belk, and Eriksen, 1990),
- 6.4-7.1…Streptocephalus woottoni in California (Eng, Belk, and Eriksen, 1990),
- 7.0………Streptocephalus woottoni in San Diego County, California (Gonzalez and others, 1996),
- 6.6………Branchinecta coloradensis in a rock pool on Enchanted Rock in Texas (Moore, 1950),
- 6.8-8.1…Branchinecta coloradensis in mountains and on Laramie plains of southeastern Wyoming (Horne, 1967),
- 7.5-8.5…Branchinecta coloradensis in California (Eng, Belk, and Eriksen, 1990),
- 6.8-7.5…Branchinecta conservatio in California (Eng, Belk, and Eriksen, 1990),
- 6.8-7.6…Branchinecta longiantenna in California (Eng, Belk, and Eriksen, 1990),
- 6.8-9.3…Branchinecta lindahli in California (table 2, Eng, Belk, and Eriksen, 1990),
- 6.4-9.8…Branchinecta lindahli in California (text, Eng, Belk, and Eriksen, 1990),
- 7.1-8.7…Branchinecta lindahli on Laramie plains of southeastern Wyoming (Horne, 1967),
- 7.8………Branchinecta lindahli in San Diego County, California (Gonzalez and others, 1996),
- 8.8………Branchinecta lindahli in Mojave Desert, California (Gonzalez and others, 1996),
- 6.9………Chirocephalopsis bundyi in mountains of southeastern Wyoming (Horne, 1967),
- 7.0………Branchinecta paludosa in mountains of southeastern Wyoming (Horne, 1967),
- 7.1-7.2…Branchinecta sandiegonensis in San Diego County, California (Gonzalez and others, 1996),
- 7.1-8.6…Branchinecta packardi on Laramie plains of southeastern Wyoming (Horne, 1967),
- 8.7………Branchinecta packardi in a rock pool in Canyonlands National Park, Utah (Maynard and Romney, 1975),
- 7.4-7.9…Branchinecta campestris in Salt Lake County, Utah (Maynard and Romney, 1975),
- 7.4-9.2…Parartemia purpurea in Western Australia (Timms, 2012),
- 7.4-9.1…Artemia franciscana in California (Eng, Belk, and Eriksen, 1990), and
- 7.9-9.2…Artemia franciscana (previously A. salina) on Laramie plains of southeastern Wyoming (Horne, 1967),
- 7.5-8.5…Streptochephalus texanus on Laramie plains of southeastern Wyoming (Horne, 1967),
- 7.6-8.5…Thamnocephalus platyurus on Laramie plains of southeastern Wyoming (Horne, 1967),
- 7.5-8.9…Thamnocephalus platyurus in California (Eng, Belk, and Eriksen, 1990),
- 7.5-10….Parartemia zietziana in South Australia, Victoria, Tasmania (Timms, 2012),
- 8.1-8.9…Branchinecta lutulenta in Grant County, Washington (Rogers and Hill, 2013),
- 8-9………Parartemia longicaudata in Western Australia (Timms, 2012),
- 8.1-9……Parartemia extracta in Western Australia (Timms, 2012),
- 8.5-9.5…Parartemia informis in Western Australia (Timms, 2012),
- 8.8-9.2…Branchinecta mackini in Mojave Desert, California (Gonzalez and others, 1996),
- 8.8-9.2…Branchinecta mackini in Mojave Desert, California (Brown and Carpelan, 1971),
- 7-9.8……Branchinecta mackini in California (Eng, Belk, and Eriksen, 1990),
- 8.9………Branchinecta mackini in Salt Lake County, Utah (Maynard and Romney, 1975),
- 8.8-9.2…Branchinecta gigas in Mojave Desert, California (Brown and Carpelan, 1971),
- 7.7-9.7…Branchinecta gigas in California (Eng, Belk, and Eriksen, 1990),
- greater than 10.0….Branchinecta gigas in various Oregon-Washington-California locations (Rogers and Hill, 2013),
- 9.5-10.0..Artemia monica in California (Eng, Belk, and Eriksen, 1990).
3 species of California fairy shrimp were experimentally subjected to water with pH’s of 8, 9, 10, and 10.5 for 8 hours. All individuals of the species Branchinecta sandiegonensis, Branchinecta mackini, and Streptocephalus woottoni survived (Gonzalez, 1996).
Temperature (degrees Celsius)
Fairy shrimp inhabit the tropics and the poles but different species seem to prefer cooler or warmer environments (e.g., Belk, 1977). Temperature has been considered a control on egg hatching (e.g., Horne, 1967). Here are some examples of water temperatures at which particular species have been found, including laboratory experiments on adults and eggs.
(For reference, 0 C=32 F, 5 C=41 F, 10 C=50 F, 15 C=59 F, 20 C=68 F, 30 C=86 F, 35 C=95 F, 40 C=104 F)
- 0-4.7….Branchinecta coloradensis was collected when the lake was still partly covered by ice and water temperatures differed at different locations (Shantz, 1905),
- 1-26…..Branchinecta coloradensis in California (summarized by Eng, Belk, and Eriksen, 1990),
- 1.5-22..Branchinecta lindahli in California (summarized by Eng, Belk, and Eriksen, 1990),
- 5……….Branchinecta lindahli eggs (about 90%) hatched in laboratory (Belk, 1977),
- 9……….Branchinecta lindahli eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 10……..Branchinecta lindahli eggs (about 75%) hatched in laboratory (Belk, 1977),
- 12……..Branchinecta lindahli eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 15……..Branchinecta lindahli eggs (about 85%) hatched in laboratory (Belk, 1977),
- 15……..Branchinecta lindahli eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 18……..Branchinecta lindahli eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 20……..Branchinecta lindahli eggs (about 85%) hatched in laboratory (Belk, 1977),
- 34.5…..Branchinecta lindahli found in turbid pond (turbidity lowers deeper temperature) in northern Arizona (Belk, 1977),
- 36……..Branchinecta lindahli died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 5……….Branchinecta packardi eggs (about 75%) hatched in laboratory (Belk, 1977),
- 9……….Branchinecta packardi eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 10……..Branchinecta packardi eggs (about 70%) hatched in laboratory (Belk, 1977),
- 12……..Branchinecta packardi eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 15……..Branchinecta packardi eggs (about 60%) hatched in laboratory (Belk, 1977),
- 15……..Branchinecta packardi eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 18……..Branchinecta packardi eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 20……..Branchinecta packardi eggs (about 55%) hatched in laboratory (Belk, 1977),
- 37……..Branchinecta packardi died within 1 hour in laboratory, male at 36, female at 38 (“LD50” of Belk, 1977),
- 5-12…..Streptocephalus seali eggs hatched during period with this temperature range in Louisiana (Moore, 1963),
- 10.6-23.4..Streptocephalus seali in California (summarized by Eng, Belk, and Eriksen, 1990),
- 38……..Streptocephalus seali died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 42……..Streptocephalus seali found in pond in Louisiana (Moore, 1963),
- 43-44…Streptocephalus seali died within 18 hours of 30-minute exposure to this temperature in laboratory (Moore, 1963),
- 1-14…..Branchinecta mackini occurred at this temperature in Mojave Desert in winter (with diurnal temperature variation of 3-12) (Brown and Carpelan, 1971),
- 12-32…Branchinecta mackini occurred at this temperature in Mojave Desert in August (with diurnal temperature variation of 12-19) (Brown and Carpelan, 1971),
- 1-16….Eubranchipus oregonus in California (summarized by Eng, Belk, and Eriksen, 1990),
- 1-17….Eubranchipus serratus in California (summarized by Eng, Belk, and Eriksen, 1990),
- 5………Eubranchipus serratus eggs (about 85%) hatched in laboratory (Belk, 1977),
- 10…….Eubranchipus serratus eggs (about 75%) hatched in laboratory (Belk, 1977),
- 15…….Eubranchipus serratus rarely found above this temperature (Pennak, 1978, p. 336),
- 31…….Eubranchipus serratus died within 1 hour in laboratory, male at 30, female at 32 (“LD50” of Belk, 1977),
- 4……..Branchinecta campestris hatched in Penley Lake, Washington (Broch, 1969),
- 5……..Branchinecta paludosa eggs (about 80%) hatched in laboratory (Belk, 1977),
- 32……Branchinecta paludosa (female) died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 5-12…Eubranchipus holmani eggs hatched during period with this temperature range in Louisiana (Moore, 1963),
- 26-28..Eubranchipus holmani died upon 2-4-hour exposure to this temperature in laboratory (Moore, 1963),
- 30……Eubranchipus holmani died upon 30-minute exposure to this temperature in laboratory (Moore, 1963),
- 5……..Eubranchipus bundyi eggs (about 75%) hatched in laboratory (Belk, 1977),
- 15……Eubranchipus bundyi rarely found above this temperature (Pennak, 1978, p. 336),
- 31……Eubranchipus bundyi died within 1 hour in laboratory, male at 29, female at 32 (“LD50” of Belk, 1977),
- 6-20…Branchinecta lynchi in California (summarized by Eng, Belk, and Eriksen, 1990),
- 6……..Artemia franciscana (previously A. salina) disappeared from “Great Salt Lake” below this temperature in autumn (Pennak, 1978, p. 336),
- 10-13..Artemia franciscana (previously A. salina) hatched in “Penley Lake”, Washington (Broch, 1969),
- 8-21…Branchinecta gigas in California (summarized by Eng, Belk, and Eriksen, 1990),
- 15……Thamnocephalus platyurus eggs (about 35%) hatched in laboratory (Belk, 1977),
- 20……Thamnocephalus platyurus eggs (about 60%) hatched in laboratory (Belk, 1977),
- 25…..Thamnocephalus platyurus eggs (about 45%) hatched in laboratory (Belk, 1977),
- 30…..Thamnocephalus platyurus eggs (about 10%) hatched in laboratory (Belk, 1977),
- 41….Thamnocephalus platyurus died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 15….Streptocephalus dorothae eggs (about 30%) hatched in laboratory (Belk, 1977),
- 20….Streptocephalus dorothae eggs (about 35%) hatched in laboratory (Belk, 1977),
- 25….Streptocephalus dorothae eggs (about 25%) hatched in laboratory (Belk, 1977),
- 30….Streptocephalus dorothae eggs (about 20%) hatched in laboratory (Belk, 1977),
- 40….Streptocephalus dorothae died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 15….Streptocephalus mackini eggs (about 25%) hatched in laboratory (Belk, 1977),
- 20….Streptocephalus mackini eggs (about 45%) hatched in laboratory (Belk, 1977),
- 25….Streptocephalus mackini eggs (about 60%) hatched in laboratory (Belk, 1977),
- 30….Streptocephalus mackini eggs (about 50%) hatched in laboratory (Belk, 1977),
- 35….Streptocephalus mackini eggs (about 40%) hatched in laboratory (Belk, 1977),
- 41….Streptocephalus mackini died within 1 hour in laboratory (“LD50” of Belk, 1977),
- 15….Streptocephalus texanus eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 18….Streptocephalus texanus eggs hatched at this mean temperature in laboratory (Horne, 1967),
- 34….Streptocephalus texanus found in pool in mostly dry stream bed in Texas (Moore, 1950),
- 17-24..Branchinecta dissimilis collected at high-elevation sites in California (summarized by Eng, Belk, and Eriksen, 1990),
- 20….Thamnocephalus mexicanus eggs (about 70%) hatched in laboratory (Belk, 1977),
- 25…Thamnocephalus mexicanus eggs (about 80%) hatched in laboratory (Belk, 1977), and
- 30….Thamnocephalus mexicanus eggs (about 40%) hatched in laboratory (Belk, 1977).
Belk (1977) also tested the changes of temperature that could be tolerated. Female Streptocephalus mackini were acclimated to 11.5 C and 25 C water for 24 hours. When transferred to 39 C water, all of the low temperature group died but only 3% of the high temperature group died.
Total Dissolved Solids (mg/L)
Total dissolved solids (TDS, see the Acronyms page) seems to be an important control on which ponds particular species inhabit but the ranges of tolerable TDS by species are quite large and overlapping. Because some populations of the genera Artemia and Parartemia inhabit waters with TDS more than double the TDS of seawater, the TDS range for the order Anostraca is mind-boggling. The secondary drinking water standard for TDS in the United States is 500 mg/L due to possible unpleasant aesthetic effects above that concentration. Seawater has TDS of about 35,000 mg/L. Units below are milligrams per liter.
- 10……….(my sum of 7 major ions) Chirocephalopsis bundyi in mountains, southeastern Wyoming (Horne, 1967),
- 15……….(my sum of 7 major ions) Branchinecta paludosa in mountains, southeastern Wyoming (Horne, 1967),
- 20-22……Streptochephalus seali in California (summarized by Eng, Belk, and Eriksen, 1990),
- 20-33……Branchinecta conservatio in California (summarized by Eng, Belk, and Eriksen, 1990),
- 33-273….Linderiella occidentalis in California (summarized by Eng, Belk, and Eriksen, 1990),
- 35-135….Streptochephalus woottoni in California (summarized by Eng, Belk, and Eriksen, 1990),
- 35-3,060..Branchinecta lindahli in California (summarized by Eng, Belk, and Eriksen, 1990),
- 3,000…….Branchinecta lindahli eggs hatched in laboratory (Horne, 1967),
- 16,465-21,572..Branchinecta lindahli on Laramie plains, southeastern Wyoming (Horne, 1967),
- 48-410…..Branchinecta lynchi in California (summarized by Eng, Belk, and Eriksen, 1990),
- 130-590….Branchinecta longiantenna in California (summarized by Eng, Belk, and Eriksen, 1990),
- 574………..Streptocephalus dorothae in southern Utah (Maynard and Romney, 1975),
- 486-4,800..Branchinecta mackini in California (summarized by Eng, Belk, and Eriksen, 1990),
- 2,861……..Branchinecta mackini in Salt Lake County, Utah (Maynard and Romney, 1975),
- 5,954……..(my sum of 7 major ions) Branchinecta mackini in Mojave Desert, California (Brown and Carpelan, 1971),
- 807-1,622..Branchinecta gigas in California (summarized by Eng, Belk, and Eriksen, 1990),
- 5,954………(my sum of 7 major ions) Branchinecta gigas in Mojave Desert, California (Brown and Carpelan, 1971),
- 1,042………Thamnocephalus platyurus on Laramie plains, southeastern Wyoming (Horne, 1967),
- 1,350-11,122..Branchinecta campestris in Salt Lake County, Utah (Maynard and Romney, 1975),
- 2,000………Branchinecta packardi eggs hatched in laboratory (Horne, 1967),
- 3,000………Branchinecta packardi eggs did not hatch in laboratory (Horne, 1967),
- 2,000………Streptocephalus texanus eggs hatched in laboratory (Horne, 1967),
- 3,000………Streptocephalus texanus eggs did not hatch in laboratory (Horne, 1967),
- 16,465…….Streptocephalus texanus on Laramie plains, southeastern Wyoming (Horne, 1967),
- 34,914-172,000..Artemia franciscana in California (summarized by Eng, Belk, and Eriksen, 1990),
- 75,000-96,400..Artemia monica in California (summarized by Eng, Belk, and Eriksen, 1990),
- 2,000-255,000…Parartemia minuta in Australia (summarized by Timms, 2012),
- 8,000-141,000…Parartemia laticaudata in Australia (summarized by Timms, 2012),
- 15,000-262,000..Parartemia serventyi in Australia (summarized by Timms, 2012),
- 20,000-95,000….Parartemia mouritzi in Australia (summarized by Timms, 2012),
- 22,000-353,000..Parartemia zietziana in Australia (summarized by Timms, 2012),
- 35,000-210,000..Parartemia acidiphila in Australia (summarized by Timms, 2012),
- 50,000-120,000..Parartemia boomeranga in Australia (summarized by Timms, 2012),
- 80,000-240,000..Parartemia contracta in Australia (summarized by Timms, 2012).
In addition to populating habitats with a huge range of TDS, individual fairy shrimp can survive large changes in TDS, as they must do if their pond is drying up. Branchinecta lindahli, Branchinecta packardi, Streptocephalus texanus, and Thamnocephalus platyurus survived increases of TDS from 1,100 mg/L to 4,600 mg/L over a 7-day period in the laboratory (Horne, 1967). B. lindahli and S. texanus also survived increases of TDS from 9,500 mg/L to 16,500 mg/L over a 3-day period in the pond (Horne, 1967). In a winter pond in the Mojave Desert with Branchinecta mackini and Branchinecta gigas, TDS increased by about 1,930 mg/L, or more than 6 times the starting TDS (assuming the increase in TDS is the same as the reported increase in salinity), over a period of about 72 days (Brown and Carpelan, 1971). Much more rapid increases in TDS in 2 summer ponds which lasted less than 6 days apparently inhibited egg hatching as less than 100 individuals were found in water siphoned from the last remaining puddles (Brown and Carpelan, 1971).
Turbidity
In a very helpful table, Eng, Belk, and Eriksen (1990) reported turbidities by species for several of the anostracan species in California. They listed means, standard deviations, and ranges for each species. The range from the least turbid to the most turbid measurement obtained regardless of species is 3 to 15,800 (NTU or JTU). In practical terms, that means from very clear to very opaque. The most turbid waters were found in the playa lakes inhabited by Branchinecta mackini, Branchinecta gigas, and Branchinecta lindahli. Suspended clay from the floors of the playas causes the turbidity. Fairy shrimp have struck it rich by adapting to turbid waters as they are likely to have fewer predators and fewer competitors than in clearer waters. Although predatory fairy shrimp inhabit such waters, they can’t extirpate the fairy shrimp prey species without extirpating themselves. The downside of turbidity is that the negligible penetration of sunlight inhibits phytoplankton. The fact that population densities in opaque water can be high (e.g., videos of Bouncing Acrobats of the Bodie Hills on the Fairy Shrimp Videos page) indicate that fairy shrimp have found some food(s) to compensate for the lack of phytoplankton.
Osmoregulation and Thermoregulation
Fairy shrimp live in waters with wide ranges of TDS and individual fairy shrimp may experience wide changes in TDS over their lifetimes. To maintain the integrity of the intracellular and intercellular fluids, the animals regulate the exchange of ions and organic molecules with the water they live in. This process of osmoregulation is as important for survival in fairy shrimp as it is in other species which experience large changes in TDS, such as crustaceans that live in estuaries and salmon which live in both marine and fresh water. Consequently, it has attracted some research interest. The only question answered in the papers I have read is whether fairy shrimp maintain the TDS (in some cases only sodium and chloride ions have been measured in experiments) of their intercellular fluids (also referred to as hemolymph) above or below the TDS of the water they live in. The ramifications of different osmoregulation strategies has not been discussed but the fact that fairy shrimp manage it over a huge range of TDS is worth noting.
Fairy shrimp of the genera Parartemia and Artemia live in water with TDS above that of seawater. Living cells have lower TDS. When waters with different TDS come into contact, they mix until they both have the same intermediate TDS. If intracellular fluid is allowed to mix with seawater, the cell would dehydrate as water flows out of the cell and into the seawater. To help prevent that, the external surface of an Artemia fairy shrimp has low permeability except for the parts of the legs used for breathing, the epipodites (Kaestner, 1970). The intercellular fluid is nonetheless exposed to high TDS water through the gut. Artemia ingest water with food and even if they were to stop feeding. To maintain TDS concentrations below that of the surrounding water and at a level the cells can tolerate, fairy shrimp excrete sodium and chloride through the epipodites (Kaestner, 1970, p. 92). In experiments at 2 water concentrations which differed by 60%, the TDS concentration of the intercellular fluid remained the same. At very high TDS concentrations, the intercellular fluid TDS concentration increased but by a much smaller percentage than that of the outside water. When the concentration of the surrounding water was reduced to about 9,000 mg/L (“1/4 seawater”), the intercellular fluid concentration switched to being greater than that of the outside water (Kaestner, 1970, p. 93). This is a problem for Artemia and they do not survive long in water with low TDS.
Recognizing that Branchinecta campestris sometimes occurs with Artemia and also survives in high-TDS water, Broch (1969) measured their intercellular fluid compared to water at various concentrations. At moderate outside water concentrations, intercellular fluid concentrations were higher. At high outside water concentrations, intercellular fluid concentrations were nearly the same as water concentrations. At the highest water concentration tested, all the animals died within 24 hours.
Species which maintain intercellular fluid TDS above the TDS of the water they live in were subjected to experiments by Gonzalez and others (1996) but they only measured sodium concentrations rather than concentrations for all the major ions. Although sodium can make up as little as 10% of TDS (e.g. Horne, 1967), it may range from 23% in moderate-TDS water of 2,861 mg/L to 36% in high-TDS water of 11,122 mg/L (Maynard and Romney, 1975). Branchinecta sandiegonensis and Streptocephalus woottoni maintain relatively constant internal sodium concentrations of approximately 1,886 mg/L and 1,587 mg/L, respectively, over a range of water sodium concentrations of 12-1,380 mg/L. However, at a water concentration of 1,840 mg/L sodium, which is not too different from that of the intercellular fluid, both species increased the sodium concentration of the intercellular fluid to about 2,415 mg/L (Gonzalez, 1996). This suggests a need to maintain an ionic gradient with respect to the external water to support some bodily function(s). In water with 2,300 mg/L sodium, over half of Branchinecta sandiegonensis and all of the Streptocephalus woottoni individuals died.
Branchinecta mackini and Branchinecta lindahli were somewhat hardier in the experiments by Gonzalez and others (1996). Instead of maintaining a near-constant sodium concentration in the intercellular fluid, they both allowed it to increase as the concentration in the water increased, but at a slower rate. At water sodium concentrations of 12 mg/L, intercellular fluid concentrations were about 1,240 mg/L higher. At water sodium concentrations of 1,840 mg/L, intercellular fluid concentrations were only about 350 mg/L higher and in water with 2,300 mg/L sodium they were almost the same. None of these test animals died at 2,300 mg/L. Combined with the Artemia data, these experiments demonstrate that fairy shrimp use a variety of strategies to deal with varying concentrations of TDS in pond water. Nonetheless, whether species typically maintain intercellular fluid TDS higher or lower than water TDS, there is an energy cost for ionic regulation. The generally small sizes of Artemia fairy shrimp may reflect this. Pennak (1978, p. 340) reported an average length of 10 mm for Artemia franciscana (previously A. salina). Parartemia species are not so dimunitive though as P. informis males average 26.7 mm in length (Timms, 2012).
Like TDS, the pH of the intercellular fluid must be maintained within certain limits for cells to function. However, research interest seems to be lacking. This may be because individual fairy shrimp do not experience pH changes over their lifetimes as large as those of TDS. The data summarized above do suggest though that while most species do fine in waters with pH 6.5-8.0, only a few (e.g. Branchinecta lindahli, Branchinecta mackini, Branchinecta gigas, Artemia franciscana, various Parartemia species) can handle waters with pH 8.5-10.0.
Fairy shrimp also experience startling changes in temperature when they live in shallow water. Fairy shrimp do not regulate their body temperature, as is true of all crustaceans. Nonetheless, their cells somehow work at 5 C as well as at 30 C, depending on species. Individual fairy shrimp in the ponds in Rabbit Dry Lake in the Mojave Desert experienced changes of up to 19 C in a single day (Brown and Carpelan, 1971). This range is the same as going from 90 F to 124 F. A human whose internal body temperature increases more than 6 C above normal usually dies (en.wikipedia.org/wiki/Human_body_temperature). Many fairy shrimp likely experience similar temperature changes over their lifetimes, if not in a single day. Notwithstanding all those other hardy animals living on the seafloor or elsewhere at very low temperatures or in the tropics at temperatures above 25 C, this is a remarkable accomplishment. It makes you wonder what’s wrong with warm-blooded animals.
Colonization and Extirpation
The basics of colonization are seemingly straightforward.
- A sufficient number of resting eggs must arrive at a new pond location.
- The pond must at some time hold water that persists long enough for hatched individuals to grow to sexual maturity.
- The physical and chemical characteristics of the pond water must be within the ranges suitable for the newly arrived species to hatch and to grow to maturity.
- Hatched individuals must survive predation and other causes of death long enough to reproduce.
- Mature individuals must meet and produce enough viable eggs for the next generation to reproduce.
Fairy shrimp resting eggs can be transported to a pond site by various dispersal agents. Machines like wheeled vehicles or tanks that pick up mud by driving through ponds or tools like shovels are possible but probably of negligible importance in the grand scheme of things. Water, mud flows, and debris flows are theoretically possible but also unlikely. Mud flows and debris flows do reach playas but their source areas are commonly steep slopes that are not likely to have ponds. Water in streams could easily transport floating resting eggs from one pond to another but ponds on streams are unlikely to have fairy shrimp because of fish. Moreover, high spring flows or storm-related floods could wash most eggs downstream and inhibit the establishment of a persistent fairy shrimp population.
Wind is a possible dispersal agent. In Nevada, the National Weather Service uses the term “blowing dust” in forecasts for particularly windy days. Strong winds can lift dust on a playa hundreds of meters (hundreds to thousands of feet) into the air and blow it laterally for 10s of km (miles) [my observations]. The dust clouds may be so dense that it is difficult to see the mountains on the far side of the playa a few km (miles) away [my observations]. Some playas do have fairy shrimp populations (e.g., Monte Cristo Valley) but sources may not be available in other windy areas. Wyoming has stronger and more persistent winds, on average, than Nevada but it also has more vegetation, on average, which means that less dust and fewer fairy shrimp eggs likely get into the air.
Animals are the most efficient and the most probable dispersal agents. Wind is not efficient because it only works with a dry pond mostly lacking in vegetation and the chances an egg lands in a pond depends on the percentage of pond area in the down wind location. Nonetheless, dust storms that routinely occur on Nevada playas could be effective in transferring eggs to pond locations even though the probability of success for an individual egg is low. In contrast, animals that visit ponds for water or food spend more time at ponds than their areal percentage would suggest. Animals can acquire fairy shrimp eggs by various means. Eggs could stick directly to fur, feathers, or other body parts or be contained in mud stuck to animals. Eggs could be swallowed in pond water or eaten along with mature female fairy shrimp. Fairy shrimp eggs survive the digestive tracts of birds (Van Stappen, 1996) and probably of some other animals. Eggs could be deposited at a new pond by rubbing off, washing off, or defecation. While only a few eggs might be rubbed or washed off at a time, the feces of a bird that has eaten several mature female fairy shrimp are likely to have hundreds of eggs. This suggests that birds which eat fairy shrimp are awesome dispersal agents.
Waiting for water is not a problem for fairy shrimp eggs. That is what they have evolved to do. Resting eggs remain viable for at least a few years (see Resting Eggs).
As summarized above, fairy shrimp species are adapted to ranges of physical and chemical water characteristics that limit the possibilities of new ponds that could be colonized. However, those ranges have evolved to be quite wide for many species. Branchinecta coloradensis lives in alpine lakes in the Wind River Mountains (Bivouac Lake) and also in a roadside ditch in the Great Divide Basin (Sweetwater Mill Road Pond). Branchinecta paludosa inhabits an alpine pond (Jons Snowy Range Pond, Snowy Range), rock pools (Lankin Dome Summit Rock Pools, Granite Mountains), a small sagebrush steppe pond (Section Marker Pond), and a large sagebrush steppe pond (“Coyote Lake”). In 1993, it was found in opaque water of “Coyote Lake” at the same time as the clay-loving Branchinecta mackini. Although I don’t have physical and chemical data for these occurrences the different environments suggest considerable variation.
Fairy shrimp survive in the presence of many predators, other than fish. Part of the survival strategy is to hatch in large numbers. A few newly introduced eggs can’t do that. In some cases, the eggs may hatch long enough before the most dangerous predators hatch that the fairy shrimp can reproduce before they get eaten. It is easy to imagine other cases where they can’t. Predation may be an important obstacle to colonization. This may make colonization by egg-rich feces of bird predators even more advantageous. Another way to beat the predators would be to colonize the pond before the predators get there. I don’t know the chances of that.
Fairy shrimp inhabit ponds that are more than 500 m (1,640′) across. If a few eggs are introduced into such a pond, how would the mature fairy shrimp find each other in order to reproduce? I haven’t read much of anything on fairy shrimp mating.
Accounts of fairy shrimp colonization that address the above factors seem to be absent from the online literature. Some indications of the frequency of colonization can nonetheless be gleaned from a few studies.
Anderson (1971) surveyed 146 fish-free alpine and subalpine ponds (elevations less than 1,550 m, 5,100 feet) in the mountains of British Columbia over a 5 year period. Most have TDS less than 150 mg/L, which is low. He mostly documented copepods and cladocerans but 14 ponds had fairy shrimp. All of the fairy shrimp observed by Anderson (1971) were in “ponds”, not “lakes”. This make sense if many of the “lakes” were previously stocked with fish but Anderson (1971) makes no mention of that possibility. Anderson (1971) identified the northern phalarope as “possibly the most important dispersal vector in the lakes of this study” and concluded that “it is unlikely that many waters in this area escape potential colonization by the most commonly occurring” zooplankton. If that is true, the scarcity of fairy shrimp implies that colonization by fairy shrimp is infrequent even in the presence of abundant apparently suitable water bodies and effective dispersal agents.
Donald (1983) surveyed a pond near Calgary, Alberta, over a period of 14 years and collected plankton samples at various times, often more than once per year. The pond is situated in an area with many other temporary ponds and a few permanent ponds. The maximum depth of the pond was 58 cm (23″). In some years, the pond remained dry. Donald (1983) identified 5 species of fairy shrimp in the pond and found up to 4 species in the pond during the same year. In most years, at least 2 of the species did not appear. Donald (1983) considered colonization-extinction and egg-viability hypotheses to explain changes in the relative abundances of species over the 14 years. The colonization hypothesis is plausible because waterfowl were commonly observed in the study area and it is “probable that in most years anostracan eggs are carried to and defecated into the pond”.
Was the reappearance of Branchinecta paludosa after a 7-year absence (Donald, 1983) consistent with characteristics of colonization? Colonization would be expected to result in initially low relative abundances of the new species (Donald, 1983) simply due to the small number of eggs involved. The relative abundances of B. paludosa were 1% in spring 1980, 3% in summer 1980 (after the pond had dried up and refilled), and 38% in summer 1981 (pond dry during spring). This is consistent with colonization but the fact that B. lindahli was absent for a longer period of time and its eggs had remained viable renders the B. paludosa colonization case inconclusive.
A better case for colonization can be made for Eubranchipus ornatus (Donald, 1983). It was not observed in 1969 through 1977 (including 2 dry years), appeared in spring 1978 with a 5% relative abundance, was absent in 1979 like all other fairy shrimp, had a 34% relative abundance in spring 1980, was absent in summer 1980 and summer 1981, and then came back with a 15% relative abundance in 1982. Because E. ornatus was not observed at all before 1978, no case can be made for an egg repository that wasn’t triggered by appropriate environmental conditions. The key questions are what change(s) enabled E. ornatus to colonize the pond and how long before colonization did that change(s) occur?
Whereas the observations of Anderson (1971) suggest fairy shrimp colonization is rare (i.e., 14 ponds out of 146) on the decadal or centennial time frame for the existence of the ponds he investigated, those of Donald (1983) suggest it is not (i.e., 1 (E. ornatus) or possibly 2 (+ B. paludosa) new species in 1 pond within 14 years).
Artemia parthenogenetica is cheating in the colonization game. This species reproduces without fertilization of the eggs so, in theory, just one egg could be enough for a new population. Genetic analysis of diploid (normal 2 sets of chromosomes, other populations have more sets) samples of Artemia parthenogenetica and sexually reproducing species in Asia showed that the species probably evolved from an as yet undescribed sexual species found in Kazakhstan and possibly also from one in Iran within the last 1 million years or so (Munoz and others, 2010). The recent origin is suggested by the lack of genetic diversity within the species. The fact that A. parthenogenetica is now widely distributed from the Canary Islands to China whereas many sexual species have smaller regional distributions (Munoz and others, 2010) does suggest some evolutionary advantage. If this is such a great idea, why now? Maybe the glacial cycles in Europe, northern Asia, and Tibet provided favorable conditions for the spread of a parthenogenetic species by freezing out existing populations and leaving unpopulated habitats for when things warmed up.
Continuing the digression on parthenogenetic fairy shrimp, such populations consistently have males but at a rate of less than 0.1% of the total population (Maccari and others, 2013). These males have the surprisingly jargon-free label of “rare males”. Rare males are morphologically normal and fertile but cannot fertilize eggs from females of their own population because parthenogenetic females produce diploid (2 sets of chromosomes) rather than haploid (1 set of chromosomes, as in normal gametes) eggs. However, the rare males produce viable eggs when mated with females of the Asian species Artemia urmiana, A. sinica, A. tibetiana, and an undescribed Artemia species from Kazakhstan (Maccari and others, 2013). The percentages of fertilized eggs were very close to those of intraspecies controls and in one case was actually greater. The hybrid offspring were normal (Maccari and others, 2013). This “suggests an evolutionary role” for rare males (Maccari and others, 2013) in a parthenogenetic species but I will leave it to you to ponder how this came about and what the ramifications could be.
Once a pond has been colonized by fairy shrimp, the population must avoid extirpation in order to persist. A number of processes could eliminate the population, such as a change in TDS due to a drier or wetter climate, an increase in a predator population, an increase in competition from other herbivores, or the destruction of the pond topography. Ponds near cities or industrial sites could be subject to chemical poisoning.
Rogers (2015a) listed reports of anostracan extinctions but did not mention causes.
There has been a tremendous amount of work on climate change but not on changes to the TDS of ponds as a result, as far as I know. Extirpation by TDS change, in particular, may have escaped notice if it only occurs on a decadal or centennial time scale.
Stocking a pond with fish is an easy way to extirpate a fairy shrimp population. There are thousands of stocked ponds just in the western United States but I haven’t seen definitive accounts of the pre-stocking specie s.
I haven’t come across any research on competition by other herbivores affecting fairy shrimp populations.
Elimination of fairy shrimp by anthropogenic pond destruction necessarily occurs on a planet with 8 billion humans. It may be so common that no one bothers to write about it.
Ironically, fairy shrimp are used for water quality testing (see Predators) but there is apparently no interest in threats to fairy shrimp ponds.
In spite of the lack of information on the likelihood of extirpation of fairy shrimp and on the reasons, fairy shrimp remain widely distributed and common. It can’t be all bad.
The extinction hypothesis is hard to evaluate in the study of Donald (1983) due to the short length of the record. Branchinecta paludosa was present in 1969 through 1972 and then absent until 1980 (the pond was dry during 2 of those years and not sampled during 2 others). The low abundance of B. paludosa in 1980 is consistent with re-colonization and supports the case for extinction after 1972. However, it is possible that the eggs had remained viable but didn’t hatch in the interim due to undetermined factors. Branchinecta lindahli was present in 1969 through 1970 and then absent until 1980. This would be a good case for extinction except that Donald (1983) was able to hatch eggs collected from the dirt in 1972. Did these eggs lose their viability over the next 7 years or did they remain viable and eventually hatch in 1980 due to some change in environmental conditions? The 61% relative abundance of B. lindahli in 1980 suggests either the initial re-colonization by a necessarily small population was not observed by Donald (1983) or there was a large, viable egg repository that finally hatched. In the end, even if the observations of Donald (1983) document extirpation, the causes were not determined.
Other Thoughts
What does the summary of the physical characteristics of fairy shrimp ponds above mean for the fairy shrimp searcher? It means you don’t have to carry a pH or conductivity meter or a nephelometer. There are likely local seasonal controls on fairy shrimp occurrences but water chemistry should not be a concern unless some sort of acid pollution or weathering of sulfide minerals has lowered the pH below 5. Find a pond. There is likely a species of fairy shrimp adapted to that pond’s chemical and physical although not necessarily biological conditions.
A means to predict whether a pond has fairy shrimp has not been developed. The physical factors summarized above are involved to some extent and there are other physical and biological factors to consider. There are also unpredictable (aka random, stochastic) factors, such as whether an individual duck carrying fairy shrimp eggs lands in pond A or pond B. A human may be able to determine whether pond A or pond B has more duck food but the duck may not know that, may consider unrelated factors like wind direction, and may have personal biases for particular pond shapes or colors, for example. Or, there may be a human waiting there with a shotgun and the duck doesn’t fall in the water. Probably the best that could be hoped for would be a probability model that would say what percentage of ponds with a specified range of characteristics would be expected to have fairy shrimp after a certain period of time given the availability and itinerancy of possible dispersal agents. This would be something like a weather forecast giving a 30% chance of rain. The day may come and go without any rain but out of 1,000 days forecasted to have a 30% chance of rain, some number close to 300 would have rain.
Experiments toward developing a probability model are easily designed but unlikely to be implemented. In the simplest, most unrealistic terms: dig lots of ponds in lots of different areas with lots of different physical, chemical, and biological characteristics and visit them a few times during fairy shrimp season every year (also remeasure all the physical, chemical, and biological characteristics, of course) for several decades. Also, track the number of pond visits by possible dispersal agents. Good luck trying to find a corps of graduate students willing to spend 30 years working on their Master’s degrees.
An alternative epidemiological approach of looking at the frequency of disease, or fairy shrimp populations, in the current population of people, or ponds, and then analyzing the characteristics of people, or ponds, to see which are most likely to be associated with infection, or fairy shrimp presence, would require orders of magnitude more ponds to arrive at a statistically significant result because current measurable characteristics may differ in unknown, but hopefully random, ways from characteristics in previous years. Good luck trying to find an army of graduate students willing to share authorship with an army of graduate students on a project with an uncertain, but probably low, chance of success.
If you want to find fairy shrimp, just get out there and have a look.
Life Cycle
What hatches from a fairy shrimp egg isn’t a fairy shrimp; it is a nauplius barely bigger than the egg (Daborn, 1977). The first nauplius stage of the genus Artemia is 0.35 mm long (Walossek, 1993, p. 86). In some species, the hatchling is a more developed metanauplius with antennae I, antennae II, and a pair of mandibles for appendages although Pennak (1978, p. 335) may have been referring to branchiopods other than Anostraca. Like the young of other arthropods, fairy shrimp nauplii have bodies unlike adults. The huge increase in body size to adulthood is accomplished by moulting. The nauplius grows, moults, and becomes the next instar with more and bigger appendages. Antennae II are used for swimming in the early stages. The earliest stage(s) may not feed at all (e.g., until 3rd instar, Daborn, 1977). Different species have 14-18 instars. Within a single species, the number of instars may change with environmental conditions (Pennak, 1978, p. 335-336).
Growth rates have been measured in a few cases. Daborn (1977) found that Branchinecta mackini grew at an average rate of about 1 mm (0.04″) per day in Alberta while Brown and Carpelan (1971) observed growth rates of 0.6-2 mm (0.02-0.08″) per day for the same species in southern California. Eubranchipus bundyi has growth rates within that range (Daborn, 1976). In a Louisiana pond, an early winter generation of Eubranchipus holmani had growth rates of 0.87 mm/day (0.034″) for the first 9 days and then 0.32 mm/day (0.013″) while a later generation had growth rates of only 0.20 mm/day (0.008″) for the first 14 days (Moore, 1963). Brown and Carpelan (1971, p. 43) inferred that the fairy shrimp they observed reached maturity at lengths of about 10 mm, at which point growth stopped while the gonads developed. Growth subsequently resumed at a slower rate but was not measured. Daborn (1976) found that a pause in the growth of Eubranchipus bundyi in Alberta 16-19 days after hatching coincided with a drop in temperature. Sampling of Branchinecta mackini in Alberta every 3-7 days showed that the fairy shrimp took at least 8 days to reach average lengths of 2 mm and then another 31 days, or so, to reach 20 mm (Daborn, 1977), by which length the growth rate had begun to slow and some females carried eggs.
Fairy shrimp reach sexual maturity in a few days to a few weeks. Some fairy shrimp of the genus Chirocephalus mature in 8 days under favorable conditions (Kaestner, 1970, p. 89). Branchinecta packardi matured in 15 days in Utah (Maynard and Romney, 1975). Artemia franciscana (previously A. salina) has taken 18 days in some studies (Pennak, 1978, p. 336). Eubranchipus holmani took 22-46 days in Louisiana, based on weekly counts by Moore (1969). Rogers (2015b) reported the number of days from pond inundation to first observation of fertilized eggs for several species in California. The fastest was Branchinecta sandiegonensis, which took only 3 days in November 2021. The slowest was the large predator Branchinecta gigas, which took 31 days in March 2006. Others species took 27, 21, 19, 17, 16, 12, and 6 days (Rogers, 2015b). In laboratory experiments designed to mimic natural conditions, “Mono Lake” fairy shrimp (Artemia monica) reached sexual maturity in 57 or 62 days under cool spring temperatures, with high or low food supply respectively, and 24 days under warmer summer temperatures and low food supply (reported in Jones & Stokes Associates, 1993b, p. J-2).
Some fairy shrimp populations have only females. For these, the eggs are not fertilized but apparently go through the same processes as fertilized eggs.
In populations with males, eggs are fertilized after copulation. To copulate, a male swims under a female, grabs the female at the end of the thorax (i.e., headward of the genital segments to avoid all those constantly fluttering legs) with its antennae II, and curls its body around to deposit sperm in the opening of the ovisac using one of the 2 penes (Pennak, 1978, p. 333) on its first abdominal segment. The ovisacs of Artemiopsis stefanssoni have no opening (Belk, 1982) so fertilization is a bit of a mystery (Thorp and Covich, 2001, p. 893).
Copulation appears to be hit or miss. In my experience, misses are close to 100%. Belk (1984) observed 21% successes in the lab using only biologically receptive females with eggs. Males commonly swim under females but this seems more like a compulsive behavior than attempts to copulate. Males sometimes swim under other males, too. Males of one species may grab females of a different species (Belk and Serpa, 1992). When males do swim under a female and grab it, the female almost always makes a jerking motion and swims away. I haven’t read how males recognize females, how females recognize males, or whether female or male choices are possible. Because fairy shrimp routinely swim away from other fairy shrimp (and moving human fingers), males must use some sense to override that tendency.
As discussed in the Anatomy section, Belk (1984) found evidence for “mate recognition”. Belk and Serpa (1992) mentioned an experiment that found that pairings were more likely between males and females from the same pond rather than between those from different ponds. This suggests not only mate recognition but also mate preference. So fairy shrimp somehow find a potential mate, decide whether to mate, and mate as a matter of routine. Like pretty much every other animal species on Earth. A high male miss rate is no deterrence to species success.
With the genus Artemia, copulation is a different story. Although I have not seen an Artemia male in the act of successfully grabbing a female, I have seen numerous males latched on to and swimming with females. They may remain paired for hours (Kaestner, 1970, p. 88), maybe days. I have even seen 2 males hanging on to the same female. I also saw a case of a male attached backwards so that it and the female were swimming in opposite directions. The 2 sort of drifted slowly in the direction the female was going. The male should have figured out that something was wrong on its first try at copulation. Maybe this indicates that pairing for long periods of time is not to facilitate multiple copulations with the same female as suggested by Kaestner (1970, p. 88) but to ensure that the male is the first to mate with the paired female. The male strategy could be to grab a female without eggs and wait for the eggs to appear. Prolonged pairing could also be used to prevent subsequent mating by other males, like chastity belts for humans.
Eggs are retained in the ovisac for one to several days before being expelled into the water (Daborn, 1977, reported up to 14 days) or remain in the ovisac until the female dies (Pennak, 1978, p. 334). They are presumably fertilized by sperm swimming around in the fluid in the ovisac during this time. It’s possible some eggs don’t get fertilized, particularly in large ovisacs with abundant eggs. The eggs of Branchinecta mackini are more dense than water and sink (Brown and Carpelan, 1971) as do those of Artemia monica in the high-TDS “Mono Lake” (as reported by Jones & Stokes Associates, 1993b, p. J-2).
The common presence of eggs in mature populations of fairy shrimp suggests that fairy shrimp remain sexually active until they die. Daborn (1977) observed that once about 40% of the females carried eggs, that percentage did not drop below 40% as long as the population lasted. Female fairy shrimp are often carrying eggs when they die. I have observed dying or dead females with eggs on pond bottoms several times.
Life spans of fairy shrimp are clearly dependent on how long the pond they grow up in lasts. Some fairy shrimp hang on until the last few millimeters of water. Examples occurred in Smith Creek Ranch Road Long Ditch Ponds in March 2013, Burnt Cabin Summit Playa Lake in June 2019, Edwards Creek Playa Lake in May 2023, and Beauty Peak East Pond in April 2021. Others die at the edges of ponds that still have ample water like those in the videos at Digging in the Shallows of Soda Spring Valley (Fairy Shrimp Videos) and at Stinking Springs Well Pond in February 2022.
Life spans not cut short by evaporation have been reported as 70-101 days for Eubranchipus holmani and 151 days for Streptocephalus seali in Louisiana by Moore (1969) and 43 days for Eubranchipus bundyi in Alberta by Daborn (1976). Artemia franciscana (previously A. salina) can survive in a laboratory for 270 days (Pennak, 1978, p. 336). For some species, life span is inversely correlated with temperature. Chirocephalus grubei lived about 90 days at 5.2 C, 78 days at 7.9 C, and 50 days at 13.5 C (Kaestner, 1970, p. 89).
There are conflicting reports of sexual differences in life spans. In “Mono Lake”, Artemia monica males outnumber females by a 2:1 ratio (as reported by Jones & Stokes Associates, 1993b). One possible explanation is that females have shorter life spans due to the higher energy requirements of producing eggs. In Alberta ponds however, males died off first in a population of Eubranchipus bundyi (Daborn, 1976) and “slightly earlier than females” in a population of Branchinecta mackini (Daborn, 1977).
Likely causes of death are few. Being eaten or dismembered by a predator is certainly probable but the great abundance of fairy shrimp in some ponds and the absence of predators from some ponds suggest those deaths may not be the most common. I haven’t read any accounts of starvation. Experiments described in the Habitats section above indicate high temperatures or high TDS can cause death. Temperatures low enough to freeze all the water in a pond would also be deadly. Large numbers of dead fairy shrimp observed in green water of Smith Creek US 50 Well Pond (Smith Creek Valley) suggest the presence of fatal algal or cyanobacterial toxins. There are likely other possible chemical causes of death, such as low oxygen concentrations or possibly high concentrations of specific elements or ions, or pollutants. One might suppose that fairy shrimp would suffocate or become exhausted in opaque, clay-rich water but those living on “Carson Lake” Playa and in Candelaria Playa Ponds seem to prefer it (see Cruising the Clay-Water Boundary Layer on the Fairy Shrimp Videos page). The clay soup of West Northumberland Road Pond #7 on August 30, 2023 (Big Smoky Valley) may have crossed a threshold though as I didn’t see fairy shrimp there. There is the possibility of disease but I haven’t read of any. Fairy shrimp are afflicted by parasites (e.g., Sanchez and others, 2016) but parasitism in the cited case was beneficial, not deadly. Short life spans lessen the chances for cancer. The lack of arteries eliminates the risk of heart attack. Being underwater, fairy shrimp don’t have falling accidents. They seem to do a pretty good job of not crashing into rocks where I have seen them in rocky ponds. In Squaw Flat Playa Lake Fairy Shrimp Video 2023-04-16-cr (“Little Fish Lake” Valley), fairy shrimp occasionally collided with the walls of a white container on a sunny day but they didn’t appear to be injured.
Fairy shrimp can take steps to reduce some life-threatening risks. They can survive by swimming at the surface if the water has low oxygen levels (Belk and Cole, 1975) or at depth if the near-surface water is too hot or too cold. Fairy shrimp are often difficult to see due to colors similar to the waters they are found in. This may save some individuals from getting picked off by birds even though the population is an important food source. Fairy shrimp don’t try to outswim predators although they perform a sudden jerking motion that may help in some cases. Fairy shrimp populations are simply absent from ponds with voracious, fast-swimming, visual predators like fish. In ponds with slower predators, a population may survive by using a cicada-like you-can’t-kill-us-all strategy. Fairy shrimp often hatch in large numbers.
As in other animals, aging may be a problem. Moore (1969) observed that Eubranchipus holmani “displays increasing evidence of senility with an accompanying high mortality rate following the initial phase of maximum egg production”. Such evidence included slow swimming and damaged appendages. Maybe senility can explain the deaths in the videos at Digging in the Shallows of Soda Spring Valley (Fairy Shrimp Videos) while smaller individuals seemed to be doing fine.
Life itself is straightforward for fairy shrimp. They swim. That’s good for respiration, feeding, moving toward more food, moving away from slow predators, moving toward thermally attractive locations, and just exploring the pond. Swimming comes naturally, right out of the shell. If they get hungry, they just keep swimming. If it gets hot, they just keep swimming. If the water gets a bit salty, well, they just keep swimming. There’s not much they can do about it.
I haven’t read any reports of fairy shrimp sleeping. Of course, I haven’t read any reports of other crustaceans sleeping either, but octopuses do. Dolphins can keep swimming and breathing while they sleep because only one of the brain hemispheres sleeps at a time. If only brains need to sleep, fairy shrimp probably don’t have to bother.
Although fairy shrimp sometimes occur in clusters with other fairy shrimp (Small Rare Plant Habitat Pond, April 24, 2019, Pine Grove Hills), I haven’t seen any explanations of their social behavior.
Winkler (1977) found an inhomogeneous distribution of Artemia monica in “Mono Lake”. At 2 sampling sites in normal lake water, Winkler (1977, p. 61) measured summer average densities of about 1,300 fairy shrimp per cubic meter (35 cubic feet). Animal densities are much greater in “sublacustrine springs” with 38,000 fairy shrimp per cubic meter and “plumes” with average densities of about 200,000 fairy shrimp per cubic meter. Restated in terms of cubic centimeters, the average for plumes is 0.2 per cubic centimeter or 1 fairy shrimp per 5 cubic centimeters (which is 0.3 cubic inches; or about 15 fairy shrimp per cubic inch). The fairy shrimp were believed to concentrate over “sublacustrine springs” due to a higher density of algae in the upwelling zones above freshwater springs on the lake floor or in methane seeps arising from “bacterial decomposition in the bottom sediments” and were to some degree “held” by the upwelling water currents (Winkler, 1977, p. 59). One hypothesis for the fairy shrimp “plumes” is that they are due to “the warming of littoral waters by solar radiation in combination with a photo- and rheotaxic-response by the brine shrimp”. “Heated bottom water rises, often along an easy upward path such as tufa and rock” and the fairy shrimp swim downward (Winkler, 1977, p. 59). Even if the fairy shrimp were guided simply by food availability and water temperature preferences, there may well have been some coordination between individuals as there is in fish schools and bird flocks. Otherwise, they might collide with each other. Neither of the explanations for the “Mono Lake” plumes and springs could account for the clusters in Small Rare Plant Habitat Pond.
Conversely, fairy shrimp may communicate in order to not cluster. This may be the case in “Alkali Lake” Fairy Shrimp Video 2023-07-06c-r (Alkali Valley). The wide spacing and varied swimming directions of individual fairy shrimp seem too consistent to be random. This could be an evolved strategy to confuse predators and make it difficult for them to choose which individual to eat. It would also make it more difficult for large predators like birds to catch more than 1 fairy shrimp at a time.
Fairy shrimp maintain their self-reliance and self-sufficiency simply by swimming, which they do every day. They don’t have much brains so they probably don’t realize they should be bored. They aren’t likely to worry about the future or remember stupid things they did in the past. If you need guidance for living in the present, look at fairy shrimp.
Resting Eggs
Branchiopod resting eggs are commonly referred to as cysts. I don’t use the term here. As eggs in an ovisac are not resting eggs and eggs found in soil are necessarily resting eggs, I will routinely omit the term “resting” when referring to resting eggs.
Fairy shrimp produce lots of eggs but the average number of eggs per female varies between individuals, between generations in the same pond, and between species. Female fairy shrimp are said to produce clutches of 10-250 eggs every 2-6 days (Kaestner, 1970, p. 88; Pennak, 1978, p. 334) yet Pennak (1978, p. 334) goes on to write that a single female may produce 1-6 clutches in its lifetime. Brown and Carpelan (1971) observed clutch cycles of 2-4 days. Different species generally have different clutch sizes but clutch size can also vary in a single generation of fairy shrimp. Daborn (1976) counted 8-32 eggs in ovisacs of Eubranchipus bundyi which he interpreted as full over a period of 34 days. He inferred that most females produced only 1 clutch but that some may have produced 2. Rogers (2015b) reported clutch sizes of 2 to 48 for Branchinecta lindahli from 4 locations in California. 42 Branchinecta coloradensis females had clutches of 9-25 eggs each with a mean of 18.7, though non-normal (Bohonak and Whiteman, 1999). 100 Branchinecta mackini females each had “about 100 eggs” (Bown and Carpelan, 1971).
It takes energy to make eggs so any survival strategy based on eggs must find a balance between energy expended and eggs produced. Larger clutch sizes were found in fairy shrimp from less predictable rain-filled ponds than in those from more predictable snowmelt ponds in Arizona (Belk and Cole, 1975). A single storm could produce anything from a few drops to a gully-washer. Rain volume affecting TDS and time of year affecting temperature result in widely varying conditions for rain-filled ponds. Eggs have different hatching responses, even among those in the same generation or in the same clutch. Thus, the more eggs produced, the more likely that some will have hatching responses that match whatever conditions the next rain-filled pond has. Conversely, assuming egg hatching responses in a particular pond population adapt progressively from generation to generation, fewer eggs would be needed to assure the success of the next generation in a pond characterized by conditions that don’t change much from year to year.
In “Mono Lake”, clutch size increases with both female body length and the concentration of chlorophyll-a in the water, which is present in green algae and indicative of food abundance. Although the proportion of egg-bearing females increases with temperature (measured 15 days earlier) and body length, the proportion of females with non-shelled nauplii, or “live” eggs, decreases with temperature and body length (Stokes & Associates, 1993). The overall effect is for females in the cooler spring to produce mostly live nauplii and to have bigger clutches the longer they get and the longer the algae continue to increase. As the larger females of the first generation die off and algal productivity declines in the summer, the mostly second generation females produce smaller clutches of shelled eggs.
Resting egg diameters range from about 0.12 mm (0.005″) to about 0.55 mm (0.02″) and are loosely related to adult size. Artemia franciscana, a small species, has some of the smallest eggs and Branchinecta gigas, the largest species, the largest. Variation within a species is commonly more than 0.10 mm (0.004″) when more than 20 eggs are measured (Hill and Shepard, 1997). Egg sizes for a single species may differ by geographic location. Shantz (1905) measured diameters of 0.162-0.195 mm (0.0064-0.0077″) for 4 eggs of Branchinecta lindahli collected near Laramie, Wyoming whereas Hill and Shepard (1997) reported diameters of 0.19-0.32 mm (0.0075-0.013″) for 89 Branchinecta lindahli eggs from California. Shantz (1905) measured diameters of 0.308-0.373 mm (0.012-0.015″) for 25 Branchinecta coloradensis eggs from near Pikes Peak, Colorado. 37 Branchinecta coloradensis eggs from California had diameters of 0.23-0.38 mm (0.0091-0.015″) (Hill and Shepard, 1997) [as interpolated from graph by me]. Except for the smallest fairy shrimp eggs, they are large enough to be seen in an ovisac, particularly if they are a different color than the water. Once they leave the ovisac, it takes considerable effort to find eggs in, or separate them from, water or soil.
Fairy shrimp produce 2 types of eggs: thick-shelled eggs that have evolved to withstand dessication and to hatch when the parents’ water body refills after drying up and “thin-shelled” eggs that hatch before the parents’ water body dries up (Pennak, 1978, p. 334). Thick-shelled eggs are covered by a shell with a secretion from the shell gland while in the ovisac (Belk, 1982) and become resting eggs. In the genus Artemia, “thin-shelled” (Pennak, 1978, p. 334) or non-shelled (Belk, 1982) eggs begin developing into nauplii before leaving the ovisac. They subsequently leave the ovisac and develop into adults. This is typically the case for the first clutches of eggs produced in the perennial water bodies of “Mono Lake” and “Great Salt Lake”. Later clutches become resting eggs (as reported by Jones & Stokes Associates, 1993b, p. J-2).
The eggs of populations inhabiting ephemeral ponds dry out and experience wide temperature changes whereas those of populations inhabiting permanent ponds don’t. Does drying out become necessary for eggs from ephemeral ponds to hatch? Not in some cases. In a field experiment by Bohonak and Whiteman (1999), eggs naturally released by females from temporary ponds hatched at about the same rates when transferred to other temporary ponds or to permanent ponds in the same area. The eggs were probably thick-shelled as they were allowed to complete their development in the ovisac.
The hatching experiments of Brown and Carpelan (1971) included “dried eggs” and “ejected eggs”. They did not identify the eggs as thick-shelled or thin-shelled. Females with full ovisacs were placed in vials with water at various test concentrations and removed after their eggs had been released into the water. Hatching rates were then observed for different test concentrations. This is comparable to the natural hatching of “thin-shelled” eggs although the “ejected eggs” may not have been “thin-shelled”. Although Brown and Carpelan (1971) were primarily interested in how hatching rates varied with TDS, they noted that “ejected eggs” hatched easily without a period of drying or undergoing diapause; so easily that they “seem to require some inhibitor to prevent hatching” (p. 53). At the least, Brown and Carpelan’s (1971) experiments with “ejected eggs” indicate that even the desert-dwelling, clay-loving Branchinecta mackini can produce new generations in perennial water bodies.
Fairy shrimp eggs have different colors, which are probably not related to species. Colors range from pale to dark, generally in shades of white to brown. The key difference may be that eggs with shells are darker than eggs without. For example, Moore (1963) mentioned “[w]hite, unshelled eggs in lateral pouches of oviducts”. It may also be that eggs with shells get darker as the shells age. Moore (1963) also noted “[e]ggs confined to median ovisac, light or dark brown iin color depending upon the extent of shell deposition.” A serendipitous photograph illustrates the color change in the eggs of a single female (Branchinecta coloradensis) in Bivouac Lake (Wind River Mountains). There are 2 rows of white, unshelled eggs in the oviducts at the headward end of the abdomen and a rearward row of presumably shelled, yellowish-brown eggs in the ovisac. The same thing can be seen in the eggs of a Branchinecta paludosa female in “Coyote Lake” (Antelope Hills). However, another female in Bivouac Lake has white eggs in its ovisac. These may be non-shelled eggs like the spring eggs of Artemia monica (see above) although August is a bit late for hatching in an alpine pond. Alternatively, the white eggs could be awaiting fertilization before the shell gland goes into action.
The surfaces of resting eggs are extraordinarily varied. They may be relatively smooth, bumpy, pocked marked, dimpled like a golf ball, crumpled like a piece of paper, swathed by sharp or rounded ridges in polygonal or irregular patterns, or spiky [my descriptions of the photomicrographs in Hill and Shepard, 1997]. In some cases, the eggs are sufficiently detailed and distinct for the species to be identified (Hill and Shepard, 1997). Is this evolution run amuck or is there some reason why eggs of different species should have such different surfaces?
Fairy shrimp eggs are covered by a shell in the ovisac but this is only the start of a process to make the eggs resistant to physical changes that would normally kill. The covered eggs divide a few times and become embryos or, as described by Kaestner (1970, p. 88), “pass through total cleavage and blastoderm formation”. 2 cuticles are formed below the shell during 2 successive molts of the embryo (Belk, 1987). The outer cuticle ruptures when the shell does but the inner cuticle remains intact until the nauplius breaks it (Belk, 1987). The shell and the 2 embryonic cuticles are permeable to water so that the embryo can detect when it is in water (Belk and Cole, 1975). After some minimal level of development, the embryo becomes dormant and metabolism essentially stops (Belk and Cole, 1975). This state may be referred as diapause or cryptobiosis. The fact that the egg has become an embryo although dormant and still covered by a shell may be why some people use the term cyst. The term cyst refers to a variety of saclike biological features, like the one in my nose that hurt when I flew on an airplane.
Once the embryo switches off, the resting egg becomes a time capsule that survives in the face of extreme environmental assaults. This is an obvious adaptation to ephemeral ponds that dry up over the course of a season and may not refill again for years. Fairy shrimp eggs are tougher than that though. Maybe their toughness is the legacy of past environmental catastrophes.
Fairy shrimp eggs survive pond desiccation, obviously, and exposure to dry soil or an atmosphere with low humidity. However, there is a limit. Moore (1967) found that after storage at 0% relative humidity for 60-90 days, only 5 of 400 Streptocephalus seali eggs hatched (cited in Belk and Cole, 1975). But 5 did hatch!
Eggs routinely survive subfreezing wintertime temperatures and summertime temperatures greater than 30 C. They also survive -190 C for 24 hours and 81 C for one hour (Pennak, 1978, p. 335).
Resting eggs can survive storage at a pressure that is about one-billionth of atmospheric pressure (“0.000001 mm Hg”) for 6 months (Pennak, 1978, p. 335). Could there be fairy shrimp eggs on meteors or asteroids cruising through the cosmos? There were fairy shrimp eggs on Earth when the asteroid that killed off the dinosaurs hit.
There is anecdotal evidence of how long eggs remain viable. Eggs left in dried pond mud on a laboratory shelf for 15 years hatched (Pennak, 1978, p. 335; also cited by Donald, 1983). Donald (1983) hatched eggs that were 4 years old. Maybe some were carried off on the ejecta.
The outer shell protects eggs from fungus and, by analogy with conchostracan eggs, from abrasion and ultraviolet light (Belk and Cole, 1975) . Some abrasion would otherwise be expected when wind-blown eggs bounce or roll along the ground. Stock ponds with fairy shrimp prove that egg populations, if not all individual eggs, survive stomping by cows and horses (e.g., Belk, 1977). Eggs survive passage through the digestive tracts of birds and bird feces are one of the most important dispersal mechanisms for fairy shrimp (see section on Predators, above). That means they survive relatively acid, enzyme-rich environments and gizzard grinding, depending on the bird species.
The dispersal of fairy shrimp eggs is discussed under colonization in the section on Habitats.
Having eggs that survive everything the environment throws at them is not enough. They have to hatch. As long as temperature, TDS, and probably oxygen concentrations are in an acceptable range for the species, the eggs seem to do so readily (some experimental data on hatching is summarized in the section on Habitats, above). Some have suggested that a period of freezing or drying or both are necessary for the eggs of some species to hatch but there is experimental evidence of hatching without freezing or drying (Pennak, 1978, p. 335). Whether freezing or drying enhance hatching may be dependent on a population’s experience. There are examples of seasonal fairy shrimp populations where the ponds never dry out (e.g., “Great Salt Lake”, Bivouac Lake in the Wind River Mountains).
Water is a necessary but not sufficient condition for the hatching of resting eggs. Fairy shrimp eggs in Candelaria Playa Ponds and Garfield Flat Stock Tank Pond didn’t hatch until 10 days or more after my first visits verified the presence of water (see Winter 2022 Fairy Shrimp Hatching History in the Garfield and Candelaria Hills and the Gabbs Valley Range, on the Garfield Hills page). Donald (1983) found that in a pond in Alberta with multiple species of fairy shrimp, some species did not hatch at all during years when other species did hatch. In one case, he demonstrated viable eggs were present in the soil by hatching them in the laboratory. Moore’s (1963) observations of a pond in Louisiana demonstrated that Eubranchipus holmani hatched in the winter but not in the fall or the spring when Streptocephalus seali had hatched. A pond that lasted from January through April in 1961 had Eubranchipus holmani but not Streptocephalus seali even though S. seali had previously hatched in the spring. Selective hatching is an important adaptation to a variable, evaporative environment because the mere presence of water does not indicate whether it will persist long enough for fairy shrimp to mature or that the water chemistry and food sources will be conducive to growth and reproduction of a particular species of fairy shrimp.
Reports of hatching experiments imply that viable fairy shrimp eggs hatch immediately upon inundation as long as chemical conditions and temperature are suitable (e.g., Horne, 1967; Belk, 1977). Discussions of the time span of hatching and of why some eggs didn’t hatch are generally lacking.
There is at least one report of fairy shrimp eggs hatching well after initial inundation. Brown and Carpelan (1971) found that for the 1965-1966 winter pond in the Mojave Desert, newly hatched Branchinecta mackini nauplii appeared approximately 20, 35, and 55 days after the first hatch. Why? Although Brown and Carpelan (1971) noted that new hatches often occurred due to a decrease in TDS caused by rain or by the melting of ice, the hatches 20 days and 35 days later in the 1965-1966 pond could not be so explained.
In the winter of 2022, eggs in Stinking Springs Well Pond hatched on at least 3 different occasions over a period of about 50 days (see “Winter 2022 Fairy Shrimp Hatching History on Carson Lake Playa and Rawhide Flats” on the Rawhide Flats page). Estimated hatching dates did not correlate well with precipitation events or regional daily air temperatures. Although the hatching of generation I could have been triggered by melting ice, that could not have been the case for generations III and IV. For the first generations in Garfield Flat Stock Tank Pond and Candelaria Playa Ponds, I found the ponds had only surface ice and were barren about 10 days before their estimated hatching dates. Garfield Flat Stock Tank Pond had a third generation hatch about 20 days after the first (see “Winter 2022 Fairy Shrimp Hatching History in the Garfield and Candelaria Hills and the Gabbs Valley Range” on the Garfield Hills page). Melting ice can be ruled out as a trigger in these cases but the hatching at Candelaria Playa Ponds was likely triggered by rain. Rising air temperature doesn’t explain the first hatch in Garfield Flat Stock Tank Pond but it could explain 2 later hatches and a hatch in Win Wan Flat West Pond.
For ponds that do not dry up, something other than the presence of water must trigger hatching. “Mono Lake” is a permanent lake with high TDS and a lower water layer that has a very low concentration of oxygen. The fairy shrimp eggs sink and those that land in low oxygen sediment (most of the lake) mostly don’t hatch. Experiencing 3 months of water colder than 5 C seems to be required for eggs to hatch (Jones & Stokes Associates, 1993b). There is a long hatching period in the spring as the water warms. The food supply also increases as the water warms above 6 C. Although some eggs hatch below 6 C, those nauplii rarely survive. Higher TDS reduces experimental hatching rates but TDS doesn’t change much from year to year due to the large size of “Mono Lake”. At least in “Mono Lake”, low oxygen concentration inhibits hatching, low temperature conditions the eggs, and increasing temperature triggers hatching.
The temperature trigger in “Mono Lake” is not a simple threshold. Experiments have shown that eggs can hatch at low temperatures but with a delay compared to higher temperatures. The average number of days to hatching was 42 at 2.5 C and 3 at 20 C (described by Jones & Stokes Associates, 1993b). The use of an average here implies that eggs are actually hatching over several days of constant temperature.
Temperature effects on hatching in “Mono Lake” raise a few questions. Do fairy shrimp eggs respond more to the maximum daily temperature, the minimum daily temperature, or to some moving average of temperature? How long would eggs wait for the temperature to warm up? Do fairy shrimp eggs have a clock that can be sped up by temperature and that can measure off 3 months of cold water? If fairy shrimp eggs don’t respond to an instantaneous temperature threshold, do they respond to a cumulative deviation above a certain threshold, something like cooling degree days?
TDS of pond water affects hatching but how would a TDS trigger work? The quantitative experiments of Brown and Carpelan (1971) provide detailed answers. They used sample sizes of 200-900 eggs. Testing the effects of a sudden influx of water to a pond, Brown and Carpelan (1971) found that lowering TDS by dilution did in fact trigger hatching almost regardless of starting concentration and that the greater the dilution, the greater the hatch. At the high end, dilution of TDS by 80% resulted in about 80% of eggs hatching.
Brown and Carpelan (1971) also demonstrated that when the oxygen concentration in the water holding the eggs was reduced from a normal level of 7 mg/L to 1 or 2 mg/L, no eggs hatched. The unhatched eggs were viable though as 60-70% hatched when subsequently exposed to oxygen-rich water. Oddly, eggs exposed to very low levels of oxygen for 120 hours and then normal levels hatched at a greater rate than eggs not exposed to low oxygen. Very low oxygen concentrations thus promote hatching but don’t trigger it. Intriguingly, ephemeral ponds may experience diurnal changes in oxygen concentrations that are extremely low at night (Belk and Cole, 1975). This reflects the lack of photosynthesis by algae and intense oxidation of organic matter by bacteria and may be a sign of a favorable level of biological activity for a fairy shrimp to hatch into. Belk and Cole (1975) also noted that intial flooding of a pond is typically accompanied by a low oxygen concentration as bacteria burst into action on the organic matter left over from the previous wet phase.
In addition to demonstrating the effects of TDS dilution and oxygen concentration on egg hatching, Brown and Carpelan’s (1971) experiments demonstrated that more eggs hatch when conditions are more favorable. In experimental conditions with different TDS concentrations, the percentage of eggs hatched decreased as TDS increased. Because the TDS of a pond increases as a pond evaporates, higher TDS suggests a pond is closer to drying up than lower TDS. This is a critical hedge against pond duration variability. Eggs can’t sense or predict pond duration but Brown and Carpelan (1971) showed they can sense TDS.
To see how good Brown and Carpelan’s (1971) fairy shrimp eggs are at guessing pond duration from TDS, I concocted a variable for pond duration. I assumed that a TDS concentration equivalent to 20,000 micro-mhos specific conductance (micro-mhos are used here to maintain consistency with the data in Brown and Carpelan, 1971; Siemens are now the proper unit) was “lethal”, i.e., that fairy shrimp exposed to that level would die due to failure of osmoregulation or would soon die due to pond desiccation. I then subtracted the specific conductance of the experimental pond water from the “lethal” level to arrive at the amount the specific conductance could change before it became “lethal”. Dividing this by the daily rate of change of specific conductance observed in actual ponds by Brown and Carpelan (1971) gives the number of days it will take for the experimental pond water to reach the “lethal” level.
To arrive at the “lethal” specific conductance, I observed that Brown and Carpelan (1971) reported that only 0.9% of eggs hatched at 10,000 micro-mhos and only 0.24% at 20,000 micro-mhos. Using the graph in Brown and Carpelan (1971), 20,000 micro-mhos corresponds to a TDS of about 11,500 mg/L in a sodium chloride-only solution. Eng, Belk, and Eriksen (1990) listed 4,800 mg/L as the maximum observed TDS for ponds with Branchinecta mackini in California. Brown and Carpelan (1971) found fairy shrimp in water with TDS (my sum of 7 major ions) of up to 5,954 mg/L. Thus, the “lethal” level is about twice the highest TDS in which B. mackini was observed and also inhibits almost all eggs from hatching.
Brown and Carpelan (1971) didn’t report daily rate of change of specific conductance but it can be calculated from observations they presented. Using the text and graphs in the article:
- August 1965 pond: 55,000 micro-mhos on 18th (last) day for a rate of 3,055 micro-mhos/day,
- November 1965-March 1966 pond: 30,000 micro-mhos on 111th day and 1,000 micro-mhos on 44th day for a rate of 433 micro-mhos/day,
- August 1966 pond: 5,850 micro-mhos on 1.75th day (42 hours) for a rate of 3,340 micro-mhos/day (pond lasted only 62 hours),
- November 1966-February 1967 pond: 10,800 micro-mhos on 80th day and 2,000 micro-mhos on 20th for a rate of 147 micro-mhos/day,
- April 1967 pond: 20,000 micro-mhos on 20th day and 1,400 micro-mhos on 2nd day for a rate of 715 micro-mhos/day, and
- September 1967 pond: increase from 1,300 to 5,000 micro-mhos over 3.125 days (75 hours) for a rate of 1,184 micro-mhos/day.
The average of these 6 determinations is 1,479 micro-mhos/day. The wide variation of the rates does not inspire confidence but what the heck.
An example calculation for an experimental specific conductance of 5,000 micro-mhos is as follows. The number of days to reach the “lethal” level of 20,000 micro-mhos is (20,000-5,000)/1,479 = 10.1 days. In other words, a pond starting with specific conductance of 5,000 micro-mhos would have a specific conductance of 20,000 micro-mhos after 10.1 days if specific conductance increased at a constant rate of 1,479 micro-mhos/day.
The data in tables 4a and 4b of Brown and Carpelan (1971) were converted to Proportion (or decimal fraction) of Eggs Hatched and to Days to “Lethal” Specific Conductance as shown in Data Table for Fairy Shrimp Hatching vs. Specific Conductance Experiments of Brown and Carpelan (1971) below. Assuming specific conductance controls the proportion of eggs hatched, then Days to “Lethal” Specific Conductance is the independent variable, or x coordinate, and Proportion of Eggs Hatched is the dependent variable, or y coordinate. The r-squared (square of the correlation coefficient, or the coefficient of determination, which is a measure of how much of the variation of y is accounted for by variation of x) of a simple linear regression of the 2 variables from Table 4a is ridiculously high, at 0.9998 (Regression of Data in Table 4a for Fairy Shrimp Hatching vs. Specific Conductance Experiments of Brown and Carpelan (1971)). The maximum for a coefficient of determination is 1. With only 4 points, this result is somewhat suspect. I went to Table 4b which has more points even though the experimental conditions are not quite as natural. Again, r-squared is very high, at 0.8880 (Regression of Data in Table 4b for Fairy Shrimp Hatching vs. Specific Conductance Experiments of Brown and Carpelan (1971)).
There must be something wrong. I tried changing the “lethal” specific conductance, which is just a guess anyway, but that only changed the intercept. The slope and r-squared remained the same. I tried changing the rate of increase of specific conductance using only the summer data or only the winter/spring data. That changed the slope and the intercept but r-squared remained the same. I redid the regression using the original specific conductances instead of days to “lethal” levels. The r-squared was the same although the correlation became negative, as it should. I even eliminated the data point for a specific conductance of 10,000 (6.76 days pond duration) as that point has such a strong effect on the regression. That reduced r-squared to 0.9994. Because the data consist of so few points, maybe the statistics would show the results aren’t that great after all. For the Table 4a data, the t statistic (2-tailed) is 5,021 and the f statistic is 68,776. These imply that the probability that the data alignment is due to chance is very small (i.e., p less than 0.00002). The t and f statistics for the Table 4b data are not quite so crazy at 5.63 and 197 but they still imply a probability of less than 0.005 that the results are due to chance. To add to a correlation that is too strong to be believable, the regression line for the Table 4a data crosses y=1 (i.e., all the eggs hatch) between 14 and 15 days pond duration. It just so happens that fairy shrimp in Rabbit Lake ponds can reach maturity in 15 days (Brown and Carpelan, 1971, referred to individuals greater than 10 mm long as mature).
The “Rabbit Dry Lake” fairy shrimp proved their hatching expertise not only in the laboratory but they did even better on the playa. Brown and Carpelan (1971) reported their observations of 3 summer ponds with dramatically different hatching rates. The August 1965 pond lasted 18 days and had a “large” but uncounted hatch. The August 1966 pond lasted 3 days and only 74 individuals were recovered from all the water that could be scooped up 12 hours before the last bit evaporated. For context, “Rabbit Dry Lake” has at times supported populations estimated at 400,000 fairy shrimp, although that may have been for a long-lived winter pond. The September 1967 pond lasted 5.3 days. Only 24 fairy shrimp were recovered from the water that was “siphoned” off the playa shortly before final desiccation. Temperature, pH, and initial salinity were similar for all 3 summer ponds but “in these short-lived ponds few eggs hatched”. In all 3 ponds, the initial TDS was about 1,000 micro-mhos. That would have allowed a high hatching rate according to the constant TDS experiments but the eggs didn’t fall for it in the short-lived ponds. They may have been able to detect the rapid increase in TDS in the short-lived ponds as the ponds evaporated and decided to wait for something better.
It is generally accepted that populations of fairy shrimp eggs don’t all hatch at once when conditions are favorable (e.g., Belk, 1977; Horne, 1967; Rogers, 2015b). Brown and Carpelan’s (1971) experiments offer strong support as 17% of the eggs didn’t hatch at a quite low specific conductance of 360 micro-mhos. This is no fluke; this is critical to the population’s survival strategy. Even if the resting eggs can track TDS with a coefficient of determination of 0.9, there is always the possibility of a black swan event. An unusual warm spell could dry up the pond before the hatched fairy shrimp could reproduce. Or a herd of thirsty wild horses could dry up the pond much faster than evaporation alone (e.g., Wild Horse Rock Pool, August 19, 1987, Granite Mountains). It is consequently advantageous for populations of fairy shrimp eggs to hedge their bets by including some eggs that don’t hatch even when the future looks bright. Then, if one year’s hatch dies off before maturity in spite of favorable omens, there will still be eggs to hatch another time.
Eggs typically remain viable for at least a few years (e.g., Donald, 1983; Pennak, 1978, p. 335). The eggs that don’t hatch accumulate from year to year to form an egg bank in the pond soil. This greatly enhances the long-term survivability of a species (Rogers, 2015a). It also reduces the chances that an introduced fairy shrimp species could take over (Rogers, 2015a). A larger egg bank gives the original species chances for hatches over several years even if it has low reproductive success in a few particular years due to competition.
The females who lay the eggs and the eggs themselves are certainly not making decisions about when to hatch so how do they do it? Are the eggs somehow conditioned by microenvironmental factors that affect eggs of the same generation differently so that even if all the eggs share their responses to major factors like TDS and temperature, there is an element of chance involved in determining whether a specific egg hatches? Do differences between females cause differences in the hatching propensity of their eggs? Is there something in the shell gland that causes the hatching propensity to vary between females?
In an experiment by Rogers (2015b), eggs of Branchinecta lindahli were collected from 4 different ponds in widely separated geographic regions of California with different temperature regimes and were exposed in the laboratory to the temperature conditions of all 4 locations. The greatest hatching rate for each of the 4 populations was always greatest under temperature conditions of their home pond. At the extremes, the southernmost (warmest) population did not hatch at all under the coolest conditions and the northernmost (coolest) population did not hatch at all under the warmest conditions (Rogers, 2015b). This is wild. Do eggs have “memories” of their parents’ lives? Do eggs “remember” their own temperature experience? These experiments indicate that environmental factors can change the responses of eggs from different populations of the same species.
Broch (1965) found that eggs of Eubranchipus bundyi that remained covered by water from June through October did not hatch the next year while those where the pond dried up, but where the soil was still moist, did (as reported by Donald, 1983). In contrast, Branchinecta lindahli did best in Alberta after those years with below normal precipitation and consequently greater drying of the soil (Donald, 1983). These observations indicate microenvironmental factors, such as local soil moisture, can change the responses of eggs in the same population.
However fairy shrimp egg hatching works, it is a success. Most eggs hatch when conditions are favorable. Some don’t. The eggs that don’t can hatch under future favorable conditions. This allows a population to survive unusual adverse events.
Ultimately, eggs are the key to the survival of fairy shrimp species, families, and the order itself. Populations survive as long as some eggs remain viable. What would it take to kill off fairy shrimp eggs and put an end to the Anostraca order?
- Not an asteroid impact, at least as long as it left most of Earth intact. Fairy shrimp fossils have been found in rocks that are somewhere between 393 and 419 million years old (see Taxonomy and Origin of Anostraca). That oldest family of fairy shrimp is extinct but fairy shrimp are still here so some family or families survived the dramatic end-Cretaceous extinction due to an asteroid impact about 66 million years ago. That’s the one which ended the dinosaur fossil record. Of course no adult fairy shrimp survived; they generally don’t live longer than several months anyway. New generations arose from the eggs.
- Not a period of intense volcanism with voluminous ash and gases causing nuclear-winter-like conditions. That happened during the end-Permian extinction about 252 million years ago. Fairy shrimp inhabited Earth before and after the event.
- Not a period of warming that eliminates the polar ice caps. These were absent 145-34 million years ago (Cretaceous through Eocene) and at other times (e.g., O’Brien and others, 2020). Fairy shrimp are still here.
- Not cosmic ray bursts from a local supernova. There have been 2 such events in the last 10 million years (Thomas and others, 2016) and fairy shrimp are still here.
- Not a temporary increase in ultra-violet radiation due to a large solar storm. These occur every few years and fairy shrimp are still here.
- Not cyclical glacial coolings and interglacial warmings. These have occurred repeatedly over the last 1.6 million years, as documented by the Illinois State Geological Survey (see “Quaternary Glaciations in Illinois” at isgs.illinois.edu) and earlier in Earth history. Fairy shrimp are still here.
- Not a period of glaciation where ice extends down to the mid-latitudes. The Wisconsin glaciation 75-15 thousand years ago did that (see “Quaternary Glaciations in Illinois” by the Illinois State Geological Survey at isgs.illinois.edu). Fairy shrimp are still here.
- Not earthquakes, hurricanes, floods, droughts, forest fires, or sea level changes. These happen in various parts of Earth all the time. Floods could wipe out a fairy shrimp population by washing all the eggs downstream but the other events are unlikely to affect fairy shrimp eggs much. Even if they do, they are unlikely to be extensive enough to affect a large proportion of fairy shrimp ponds.
- Probably not a nuclear holocaust.
- The immediate impacts of direct radiation, heat flash/fireball, blast wave, and subsequent fires are unlikely to affect all fairy shrimp-hosting ponds globally as they are distributed from Antarctica and the Arctic to South America, southern Africa, and Australia in addition to the regions likely to be most affected (for a synopsis of the effects of a nuclear exchange see Wolfson, R., and Dalniki-Veress, F., unknown date, The Devastating Effects of Nuclear Weapons at thereader.mitpress.mit.edu/devastating-effects-of-nuclear-weapons-war/).
- The longer duration “fallout” of radioactive isotopes would be deadliest within kilometers/miles of detonation sites and would produce increased risk of cancer, and later cancer death, in metabolizing individuals elsewhere but might have less of an effect on cryptobiotic resting eggs. Radiation effects decrease exponentially over time but persist at low levels for decades. Effects also decrease as the isotopes diffuse into water/ice droplets and fall out of the atmosphere as rain/snow but then the receiving soil or water becomes more radioactive. This could be a problem for some ponds.
- Over periods of months and years, ozone would be depleted by nitrogen oxides and soot lofted into the stratosphere and the resulting increase in ultraviolet radiation at the surface would increase disease and cellular dysfunction (see Birks, J.W., and Stephens, S.L., 1986, Possible toxic environments following a nuclear war, in Solomon, F., and Marston, R.Q., editors, The medical implications of nuclear war: National Academies Press). The shells of resting eggs block some ultraviolet radiation by analogy with conchostracan eggs but percentage blocked was not mentioned by Belk and Cole (1977). Pond bottom dirt could block some more but the ultimate effects of increased ultraviolet radiation on resting eggs are uncertain.
- Toxic chemicals would be spread into the atmosphere and distributed globally by blasting or burning of petroleum products, building materials, road materials, asbestos, plastics, and chemical and other industrial facilities. Toxicity would be greatest closest to, and downwind of, urban centers and major industrial complexes. Like radioactive isotopes, they would eventually accumulate in soils and water as they are rinsed from the atmosphere. Some humans would die from the inhalation of highly toxic compounds like hydrogen cyanide and sulfur dioxide and those who survive the early months, if any, could succumb to cancers caused by a wide variety of chemical toxins over a longer term (see Birks, J.W., and Stephens, S.L., 1986, Possible toxic environments following a nuclear war, in Solomon, F., and Marston, R.Q., editors, The medical implications of nuclear war: National Academies Press). Although the effects of toxins on fairy shrimp are not well documented, even if all ponds within 30 km (19 miles) of urban centers and major industrial complexes become uninhabitable, that leaves plenty of others for fairy shrimp.
- Climate modeling indicates that the soot produced by firing about one third of the nuclear weapons available in 2006 would cool average global surface temperatures by 3-4 C (5.4-7.2 F) for 4 years and reduce average global precipitation by 20-27%. Local changes would be more dramatic. For example, July temperatures (after a May nuclear exchange) would be about 13 C (23 F) cooler in Iowa the first year and 10 C (18 F) cooler the second year. After 10 years, average global surface temperature would still be 1 C (1.8 F) cooler (see Robock, A., Oman, L., and Stenchikov, G.L., 2007, Nuclear winter revisited with a modern climate model and current nuclear arsenals: Still catastrophic consequences: Journal of Geophysical Research, v. 112, D13107, 14 p.). Although it could be several years before ponds warm up to temperatures suitable for the hatching of established fairy shrimp populations and receive sufficient precipitation, fairy shrimp eggs can wait that long.
- Global cooling in a nuclear winter is caused by the unusually dense concentration of soot and dust in the atmosphere reducing the incoming sunlight to 61% of what is normal in the first year and still 83% of normal at the end of the 4th year (Wolfson and Dalniki-Veress, 2007, estimated “change in global surface SW [shortwave radiation]” of -58 Watts/square meter maximum and -25 Watts/square meter at end of 4th year [my interpolation on tiny graph] compared to a global mean “net downward (or absorbed) surface SW radiation” for the period 1984-2000 of 149 Watts/square meter according to Hatzianastassiou, N., Matsoukas, C., Fotiadi, A., Pavlakis, K.G., Drakakis, E., Hatzidimitriou, D., and Vardavas, I., 2005, Global distribution of Earth’s surface shortwave radiation budget: Atmospheric Chemistry and Physics, v. 5, p. 2847-2867). Soot and dust would eventually be removed from the atmosphere in precipitation, which would be lower than normal for at least a decade due to the decrease in evapotranspiration as a result of cooling by soot and dust. Darker days for several years would greatly reduce photosynthetic activities. The ability of phytoplankton to come back after a nuclear exchange may be just as consequential for fairy shrimp as egg survival.
- The combination of increased ultraviolet radiation, radioactive fallout, toxic chemical fallout, and decreased temperature and sunshine could well affect different species differently and change the balances of predator-prey relations and food competition in ponds. Too little is known about the species involved to support a guess about whether fairy shrimp would fare better or worse than other aquatic animals.
- N.B. Robock, Oman, and Stenchikov (2007) also presented even more devastating model results for firing all the nuclear weapons known in 2006 (i.e., 150 Teragrams of soot) but I consider that less realistic than the 1/3 case (i.e., 50 Teragrams of soot) because it fails to account for weapons not mounted on launch vehicles when the opponent launches or for those destroyed or disabled in the first exchanges and because the Stockholm International Peace Research Institute estimated nuclear stockpiles in 2021 were about 63% of those estimated by Robock, Oman, and Stenchikov (2007). A large majority of humans will die within the first year in either case but the differences could be significant for more resilient species.
Does this mean Anostraca is the greatest Crustacean order ever? Not necessarily. Of the 50 Crustacean orders (following Moore, 1969, but splitting Conchostraca into 3 orders after Olesen, 2007, and Cladocera into 4 orders after Olesen, 2007, and Thorp and Covich, 2001), Moore (1969) indicated 14 are extinct. They are the losers now regardless of their accomplishments at their times of existence. Although some of the 36 extant orders may be relatively new arrivals, several crustacean orders have been around since before the end-Cretaceous extinction, including the other branchiopod orders Notostraca and Conchostraca (i.e., Spinicaudata) and some may be even more widespread and numerous than Anostraca. Some orders of barnacles (class Cirripedia, order Thoracica) and ostracods (class Ostracoda, orders Myodocopa and Podocopa) have been on Earth longer than Anostraca and are definitely contenders. A detailed comparison of orders is way beyond the scope of this web site but, if nothing else, we can opine that Anostraca is one of the most graceful of crustacean orders.
Humans have driven some species to extinction but, as far as I know, not an order with dozens of species. Ongoing fish-stocking zooicide (see Wind River Mountains, East Central Sierra Nevada) has likely eliminated thousands of fairy shrimp populations and, of course, we will never know if that included some entire species but it hasn’t come close to wiping out the order. Now that humans are raising Artemia fairy shrimp to feed fish and shellfish, the order doesn’t seem to be in any danger from humans beyond routine wanton slaughter in the near future.
So fairy shrimp might be here until the sun hits the red giant phase and boils off Earth’s water. That’s pretty good for such a small and superficially defenseless animal.
Other Crustaceans You May Find with Fairy Shrimp
If you go searching for fairy shrimp, there will likely be times when you don’t find any. Do not despair, there is more to ponds than fairy shrimp. Fish-free ponds have an abundance of animal species other than fairy shrimp and some of them are visible. Thorp and Covich (2001) wrote a book on freshwater invertebrates that is 1,056 pages long, and that is only for North America. Below, I try to provide enough information and photographs to allow you to recognize tadpole shrimp, clam shrimp, cladocerans, ostracods, and copepods. There are also lots of insect, mollusk, and other species you could find but I have to draw the line somewhere.
Tadpole Shrimp (order Notostraca)
Tadpole shrimp are branchiopods of the subclass Calmanostraca that are not likely to be confused with anything else. The bodies are 2-10 cm (0.8-3.9″) long (Belk, 1982). They look like little horseshoe crabs, in spite of the common name. They have a flexible, variably translucent, usually brownish, shield-like carapace covering the head and most of the body. Near the front of the carapace are 2 black, kidney-shaped compound eyes that are close together. The thorax and forward abdominal segments have more than 35 pairs of legs but they don’t stick out from underneath the carapace. The rearward abdominal segments lack legs and are not covered by the carapace. There are 2 long, thin, segmented, tail-like features that extend from the telson at the end of the body and are very conspicuous. In some species, the telson also has a short, stubby supra-anal plate between the 2 “tails” (Belk, 1982). The antennae are inconspicuous but the first (or first 2) pair(s) of legs look like antennae as they have long, thin, whip-like branches that generally stick out laterally from under the carapace. (Kaestner, 1970, p. 95-97; Pennak, 1978, p. 328-329).
Tadpole shrimp are generally bottom feeders and are found mostly on or near the bottom of a pond. They can swim as fast as fairy shrimp.
Tadpole shrimp, clam shrimp, and fairy shrimp on the white lid of a peanut butter jar. A whip-like branch of a front leg can be seen on the tadpole shrimp turned sideways at left. The tadpole shrimp at center has a supra-anal plate between its “tails”. DB identified samples from this lake as Lepidurus lemmoni.
View down through murky water at a tadpole shrimp in a white container. The eyes on the carapace and the 2 “tails” at the end of the abdomen are easy to see. The long, whip-like branches of the first legs can be faintly made out above the carapace and, with difficulty, below.
Ventral view of a tadpole shrimp that flipped over in the net. It obviously has lots of legs but good luck counting them. The pale supra-anal plate can be seen at the end of the abdomen.
Clam Shrimp (Conchostraca, now divided into orders Spinicaudata, Laevicaudata, and maybe Cyclestherida)
Clam shrimp really do look like small clams, but they swim. Clam shrimp carapaces are up to 2 cm (0.8″) long (Belk, 1982). There is not much visible detail. The bodies are flattened laterally within a translucent bivalve carapace. The 10-32 pairs of legs (Kaestner, 1970, p. 99) do not protrude from between the valves but pairs of bristly branches of modified second antennae do. These second antennae are used for swimming instead of the legs (Pennak, 1978, p. 329). The end of a clam shrimp body has a pair of stout anal spines (Pennak, 1978, p. 328) but they are generally not visible. Depending on species, concentric growth lines can be seen on the valves. Ostracods are also swimming clam-like animals but are smaller (see below). If it’s a freshwater swimming bivalve and it’s 0.5-2.0 cm (0.2-0.8″) long, it’s probably a clam shrimp. If you thought the clams with legs in the B.C. comic strip or the strolling oysters in Lewis Carroll’s “Through the Looking Glass” (poem recited by Tweedledee) were a surprising novelty, wait until you see clam shrimp.
Clam shrimp are generally not filter feeders like fairy shrimp and are more likely to be found near the pond bottom or in vegetation rather than swimming in open water. They can swim as fast as fairy shrimp.
See photograph Northeastern “Lewiston Lake” 1993-06-10, #0424c, above. The small, brown, oval-shaped animals in this photograph are clam shrimp. DB identified conchostracan samples from this lake as Caenestheriella belfragei.
A clam shrimp swimming near the bottom of a pond among sparse vegetation.
The pale, bivalve animal to the left of the tadpole shrimp is probably a clam shrimp.
Ostracods
Like clam shrimp, ostracods have 2 valves and swim. Unlike clam shrimp, they have a hard shell rather than a flexible carapace. The length of the shell can be as short as 0.2 mm (0.008″)(Thorp and Covich, 2001, p. 830) or as long as 2.5 mm (0.09″) (Thorp and Covich, 2001, p. 836). The length is significantly longer than the width. The hinged side is commonly convex and the other side straight or slightly concave (Thorp and Covich, 2001, p. 835-841). They commonly have colors of yellow, pale brown, or dark brown. They swim remarkably fast. I don’t have any photographs of ostracods because they are too fast. In my inadequate experience, there isn’t much to distinguish them from clam shrimp visually – just smaller and faster.
Cladocerans
Cladocerans are close to the limit of what one can identify visually without a microscope. They are small, 0.2-18 mm (0.08-0.7″)(Thorp and Covich, 2001, p. 851). Most are less than 6 mm (0.24″) long (Belk, 1982). In cross-section, they have a discus shape (Thorp and Covich, 2001, p. 851) and generally swim with the shorter dimension horizontal. They are round, ovoid, pear-shaped, or lumpy when viewed from the side and one end is rather pointed in some species (Thorp and Covich, 2001, p. 876-890). They have a carapace covering all but the head and a single eye (Dexter, 1959, p. 559) but the carapace may be too transparent and the eye too small to see.
With cladocerans, the best I can hope for is to distinguish them from copepods, which are also small and rather common in ephemeral ponds. In that light, the best distinguishing feature of a cladoceran is the pair of branched second antennae that extend laterally from the front, or top, of the animal and are used for swimming (good figures in Thorp and Covich, 2001, p. 852, 876, 886, 888, 889, 890). Copepods also have a pair of antennae (but first antennae, not second antennae) that extend laterally from the front of the body and are used for swimming but they are not branched (figures in Thorp and Covich, 2001, p. 919, 922). Because the cladoceran head is not covered by the fatter carapace, one rule of thumb that may be helpful in distinguishing cladocerans from copepods is that if the front end (based on forward motion) is narrower than the back, then it is a cladoceran. Copepods generally have the fatter end at the front.
These cladocerans swimming in a white container have obvious brown shells covering all but the head, discoid shapes, and swimming appendages that are branched, particularly visible on the cladoceran at lower right. They are 2-5 mm (0.08-0.2″) long but I didn’t make a good estimate of length at the time.
Cladocerans and fairy shrimp swimming in a white container. The animal at upper right has branched swimming antennae and is narrower at the front than the rear so it is almost certainly a cladoceran. The carapace is essentially transparent. I don’t know what the dark feature under the carapace is. At the time, I estimated these cladocerans to be 2-3 mm (0.08-0.1″) long. The reddish brown swimmer at top center is probably a copepod even though it is almost as big as the cladocerans. In a different photograph, there is one of these with unbranched swimming appendages on the fatter front end of the body.
Copepods
Swimming freshwater copepods are mostly 0.5-2.0 mm (0.02-0.08″) long but some species are up to 5 mm (0.2″) long (Thorp and Covich, 2001, p. 916). The copepods you see while looking for fairy shrimp are likely to be smaller than the cladocerans you see but unless you see both in the same pond that doesn’t help much. As discussed above under Cladocerans, my preferred criteria for distinguishing copepods from cladocerans is that copepods use an unbranched pair of antennae for swimming and cladocerans use a branched pair of antennae for swimming. Unfortunately, it is difficult to see the antennae, much less whether they are branched.
Copepods have an exoskeleton (Thorp and Covich, 2001, p. 915) so the legs are not enclosed within a carapace. Although they stick out laterally from the body they are too small to see. You might have better luck seeing the 2 tail-like appendages (rami) that extend from the rear end of the animal (Thorp and Covich, 2001, p. 915). I haven’t seen these either. Unlike cladocerans, copepods are approximately cylindrical in cross-section. Many species have a distinct change in thickness near the middle of the body with the front half fatter than the rear half (e.g., Thorp and Covich, 2001, Fig. 2 on p. 916 and Fig. 5 on p. 922). There is one feature that is a dead giveaway but it is not always present. When female copepods have eggs, the eggs fill 2 oblong pouches that stick out from the rear third of the body (e.g., Thorp and Covich, 2001, Fig. 5 on p. 922). Visually, it just looks like the rear of the body has a bulge or is just fat but a bulge due to eggs is shorter than the fat front part of the body. Cladocerans don’t have anything like this.
Both copepods and cladocerans swim with a jerky motion of short spurts in 1 direction and then another spurt in a different direction. Copepods’ motion, however, is more start-stop and cladocerans’ motion is more continuous (shorter pauses), in the examples I have seen.
One rule of thumb that may be helpful is that if you see an uncountable number of dark little specks swimming around in the water, they are more likely copepods than cladocerans. Unfortunately, I don’t know if they could be something other than copepods or cladocerans.
The 2 animals at right are swimming toward the left so they are fatter at the front. The long curved swimming antennae at the front of the animals are not branched so these are not cladocerans. They are about 2 mm (0.8″) long, which is big for copepods but small for cladocerans. My guess is copepods.
The 2 animals at right have reddish bulbous masses near their rear ends. I think these are egg pouches. If they are, these are almost certainly copepods. The copepods at upper left and lower center do not have these masses so they are probably males. Unbranched swimming antennae can also be seen. The greenish thing at the right edge of the photo is the head of a backswimmer which is back-side-up and trying to hide under a rock off the right edge of the photo but is too big for the space available.
Below, the cropped enlargements of photograph Smith Creek Cold Springs Ponds 2021-04-23, #09, show that making tentative identifications of cladocerans and copepods based just on size and shape is feasible and is right in some cases.
Animals from Smith Creek Cold Springs Ponds have been transferred to a white container for better viewing. There are a few fairy shrimp at left. The floating rush stem is 3 cm (1.2″) long. There are abundant, lenticular animals that are 3-4 mm (.12-.16″) long and look sort of like rice grains. Some are narrower at the front than at the rear so they are probably cladocerans. There are also a few dark specks that are only 1-1.25 mm (0.04-0.05″) long. Little detail can be seen at this scale but size alone suggests they are copepods.
This is a cropped version of photograph Smith Creek Cold Springs Ponds 2021-04-23, #09. Additional details can be seen in the suspected cladocerans. Most have a dark feature in the forward half of the body. I don’t know what it is. Pale features in the rearward half of the body give it a segmented appearance. While still too small for interpretation, these features at least do not contradict the tentative identification. The shapes of the 2 suspected copepods in this view (above and to right of center, left of center) can now be discerned. The one to the left of center is fatter at the front. The one to the right of center has a dumbbell-like shape that is fat at both ends. This is likely a copepod with eggs.
This is the cropped lower right part of photograph Smith Creek Cold Springs Ponds 2021-04-23, #09c. Branched swimming antennae and a single black eye on a narrow head can be seen on one of the 2 suspected cladocerans at lower left. In the suspected cladoceran to the right of center, the segmented appearance of the rear half of the body can be resolved into 3 rows of 4 spherical eggs. Cladoceran females carry their eggs under their carapaces. The fatter front end of the suspected copepod above center is more visible. At even greater magnification, a pair of unbranched swimming antennae can be seen extending laterally from the head. This confirms it is a copepod.