Breeding Yamato (a.k.a. Amano) shrimp, Caridina japonica.

By Mike Noren.


Mid-2003 I got hold of my first few Amano shrimp, Caridina japonica, although sold to me under the (wrong) name Caridina serrata.

I immediately started planning for breeding, placing them in a densely planted tank, but it was not until October, when I bought 6 more shrimp, that I started getting some action - as it turned out, my 4 original ones were all males!



As I found out, you must have both males and females to breed Amano-shrimp. While there is a definite difference in size (males are much smaller, see figure 2) and body shape (females have broader tails, distinctly convex if seen from the dorsal side), there is a much easier way of telling the sexes apart: you look at the second row of spots along the side of the shrimp. Females have elongate spots, more a broken line than individual spots really, while males have round spots (see figures 1 & 3).


 Figure 1. The spots in the second row of spots are round in males (left) and elongate in females (right).


Once you have determined you have males and females, you just have to make sure the animals are well-fed, and wait. Soon you will be able to see the developing eggs as a yellowish-greenish 'filling' dorsally in the females - up to one third of the females volume will eventually be occupied by developing eggs!

As the female is getting ready to breed, she releases pheromones into the water, and the males become increasingly frantic. Prior to the first mating, the males "swarmed" around the tank, much like moths around a lamp, trying to copulate with everything which moved, including each other and my poor fish!

Copulation normally takes place late in the evening, with the male first swimming around, then landing on top of, the female. If she does not fight him off, he then climb in under her, belly to belly, to deposit the sperm. The whole procedure is over in a matter of seconds. If the female is not ready, or if there are many males trying to mate with her, fights may break out. With many males present, one may even get "communal" spawnings, with one female surrounded by numerous of males.


Figure 2. A male (top) attempting to breed with a female shrimp. If she does not (quite forcefully) reject him, he'll climb in under her, and, using a pair of legs modified as copulatory organs, deposit his sperm in the females sperm receptacle.


A couple of days after copulation, the female lays the eggs, gluing them onto her abdominal swimmerets. Literature suggests the female may lay as many as 2000 eggs, but although my females are large and well fed my guesstimate is that they've never had more than 200 - 400 eggs. When freshly laid, the eggs are a dark moss-green color, but become progressively lighter and more khaki in color as they mature. The female carries the eggs for about 5 weeks before they hatch. The color of the eggs can be used as a rough guide to when they're going to hatch, but a better estimate can be obtained by looking closely at the eggs (e.g. with a hand lens, or, as I do, via a macro photo): if you can see the eyes of the developing embryos through the eggshell, then the eggs will hatch within a few days.


When the eggs are close to hatching, the eyes of the embryos are visible through the eggshell.

           Figure 3. These eggs are close to hatching: they're light in color and the eyes of the embryos are visible through the eggshells. The bright yellowish stuff dorsally inside her body are her gonads, filled with developing eggs - the female starts producing the next batch before the last one is hatched! Contrast has been enhanced to show eyes more clearly.


You may let the eggs hatch in the aquarium, the larvae are positively phototactic (swim towards light) and can be collected by shining a flashlight into the aquarium at night, and siphoning out the larvae as they are drawn to the light. However, if you've got fish in the aquarium, they're likely to eat the larvae. I'd recommend you remove the female when hatching draws close and place her in a jar or small breeding tank, where the eggs may hatch in safety.

The eggs do not all hatch at once. In my experience, most hatch during the night, and the remainder throughout the next day.

You should return the female, who does not eat her young, to the main aquarium as soon as the eggs have hatched. Within days of releasing the larvae, the female will have mated, moulted, and be carrying a new batch of eggs. Mature females are always producing eggs, and nearly always carrying some as well.

Note: do NOT place the female in brackish water! While I found out that adults survive quite high salinities, the eggs fail to hatch if the water is brackish - I lost two batches this way.



The newly hatched young are whitish and exceedingly small, approximately 1.8 mm long. The first time I saw the larvae, I for a moment mistook them for small water fleas - that's what they look like. The larvae are planktonic, and hatch at a fairly late larval stage, known as the mysid or mysis stage, since the larva resembles a mysid shrimp, and like mysis swim with the legs on the thorax, and spend the first weeks swimming in a curious vertical, head-down, position, after which they adopt a more horizontal, if slightly bent, swimming position.


Figure 4. These larvae are just hours old, and still in the 2 liter jar where I move my females to hatch. They're transparent to faintly whitish, and hard to see - contrast has been improved in the magnified section to show the larvae more clearly. The larvae are here slightly less than 2 mm long.


Figure 5. Light-microscope image of a freshly hatched larva. This is an early mysis stage larva - note absence of legs on the abdomen. The grainy substance in the larva is presumably yolk.


In nature, adult Amano-shrimp live in mountain streams, but the larvae are washed out into the sea, where they feed on marine plankton and grow. After metamorphosis they migrate back up into the streams. We must mimic this cycle in order to breed the shrimp. Therefore the larvae have to be transferred to salt water as soon as possible, at the latest at the 8th day after hatching, because after that they become unable to live in fresh water - or in insufficiently salty brackish water, for that matter. There is no need to gradually increase salinity, the larvae have no problem being unceremoniously dumped straight into salt water. To make the salt water, I would suggest using either filtered natural seawater, or a quality commercial salt mix intended for coral reef aquaria, e.g. Instant Ocean, which should be aerated vigorously prior to use.

While there are numerous reports of Amano-shrimp reproducing in freshwater, my own experiments at using pure freshwater met with total failure. There does appear to exist a closely related species occasionally for sale which is virtually indistinguishable from Caridina japonica, and whose larvae can complete development in freshwater, however I believe most reports of freshwater breeding are cases of mistaken identity (e.g. Tow Fui's guide, which surely concerns some smaller species with direct development, not Caridina japonica).

One guide suggest that 17 ppt (parts per thousand) is a good salinity for raising the shrimp, but when I first tried 17, and then 25 ppt, I saw >97% losses in the first three weeks, with the bulk of losses occurring day 8 and 9, and very slow growth. By contrast, when I used full marine salinity (35 ppt), I have not had any noticeable losses, and growth has been rapid. Another guide on the net who, like me, experimented with different salinities came to the same conclusion: any salinity below 30 ppt will result in heavy larval losses!

I've used an airstone with reduced flow for circulation. One guide advised against this, on the grounds that the shrimp larvae get trapped in the surface tension. With my first batches I had no such problem, but with a later batch I did indeed experience losses because larvae got trapped in the surface and died. The main difference was that I had not changed water as frequently, and a surface film had formed. I removed the film mechanically and increased circulation and airflow, and larvae no longer got trapped. Although larvae do not appear to be particularly sensitive to water quality, I suggest frequent, small, water changes to maintain water quality - I tried to change 20% every two days.


Figure 6. The 15 liter acrylic tank I used to raise the larvae. Circulation: an airstone with reduced flow. Black cardboard was taped to the sides and bottom of the aquarium to keep larvae swimming. Lighting: an 11 W compact fluorescent desk light left on 24 hours per day, 7 days per week. Temperature is 22 - 24 degrees Celsius.



The larvae require fine-particulate food for the first weeks, after which they are capable of accepting larger food particles. Judging by published reports, the larvae can subsist on pretty much any food small enough - successful rearing have been reported using e.g. fine-crushed Spirulina-flake, and dry yeast.

I fed mine about 5 times per day, initially with small amounts of brewers yeast, generous helpings of greenwater (containing a green alga which to me, in my microscope, looks much like Chlorella), Baby Star II (50-100 microns microencapsulated feed for egg-laying freshwater fish), and after a week also Golden Pearls (microencapsulated food for marine invertebrate larvae, 50-100microns).

The golden pearls seem to be a nearly ideal food - the larvae eagerly accept it, grow fast, and the strong color of the feed makes it easy to see that the larvae are eating (the guts turn brownish red, and eventually the whole larva turns orange). However, variation is always a good thing, so I suggest also giving other feeds, at least occasionally. Also the golden pearls clearly pollute the water a lot, so be careful not to overfeed!

I tried feeding the shrimp freshly hatched Artemia, but although I could see them catch & eat some artemia, most of the artemias quickly grew too big for the shrimp to eat, and actually started reproducing in the rearing tank. As the Artemia are clearly competing with the larvae for food, I'd advice against feeding Artemia to Amano-larvae.


Figure 7. The white animals are Artemia, originally introduced as food for the shrimp but now grown & reproducing; the orange animals are late mysid stage shrimp larvae, about 8 mm long. The orange color presumably comes from carotenoids in the Golden Pearls diet. The lines in the image are scratches in my old acrylic 15 liter rearing tank.




Figure 8. Close-up of late mysid-stage larva. This picture was taken just days before the first larvae metamorphosed into postlarvae. The dark reddish-brown thingy underneath the head of the shrimp is a clump of food the larva is holding. The larva is about 8 mm long.

(Incidentally, I apologize for the quality of the images - neither the jam-jar nor the acrylic tank are optically very suited for macro photography. I hope to replace them with higher-quality images later on.)



My shrimp larvae started metamorphosing to postlarvae (juveniles with the same body-shape as the adults) at day 30. The early postlarvae are about 8 mm long, and no longer planktonic, instead sitting on the walls/bottom of the tank like the adults. They've also lost their orange color, and are now transparent/whitish like the adults. My experiences suggest they are capable of being transferred to freshwater immediately after metamorphosis.

Some web resources claim that it takes 80 - 90 days for larvae to reach postlarval stage, but my shrimp metamorphosed between days 30 to 60. Probably those other reports can be explained by to too low salinity, inadequate feeding and/or poor quality feeds.


Figure 9. Freshly metamorphosed postlarvae. The "rope" is the nylon string I've used to tether the airstone. The diameter of the string is 2 - 3 mm, the shrimps are about three times longer than the diameter of the string, making the shrimps 7 - 8 mm long.



Figure 10. Close-up of two freshly metamorphosed postlarvae.  Notice that the legs on the thorax are now walking legs. They're photographed through the acrylic they're sitting on, but even seen from the underside one can tell they are now miniature versions of their parents.




The larvae do not metamorphose into adult form all at the same time. It took almost two weeks until the majority had completed metamorphosis, and even after three weeks there were a few mysis-stage stragglers. Worse, I estimate I got about 100 postlarvae, meaning I lost 50 - 75% of the larvae during metamorphosis. Clearly there is some problem associated with metamorphosis. My guess is that they no longer tolerate full-strength seawater after metamorphosis, and that these losses might be avoided by lowering salinity to, perhaps, 17 ppt when the first larvae start to metamorphose.

Day 48 I did an experiment and moved 10 postlarvae to the freshwater aquarium the adults came from. I introduced them via a floating net, so I know they survived the first 12 hours in freshwater, but as soon as I released them from the net they, predictably, vanished: given their small size and the amount of plants and hiding places in the tank, there's little chance of spotting them. It was not until months later I found that three had survived and were now almost adult.

All the remaining postlarvae were transferred to a 30-liter grow-out tank, decorated with rocks, sand & plants to offer more hiding spaces, thereby hopefully avoiding losses due to cannibalism. Initial salinity 17 ppt, after which all water changes were done with freshwater. I fed Spirulina pellets, crushed flake, and shrimp mix.

The first week after transfer the postlarvae were extremely secretive, staying hid at all times. After that they adopted about the same behavior as the adults, spending a lot of time sifting the sand and occasionally swimming.


At Day 65 an estimated 50 postlarvae were still alive. The biggest approach 15 mm total length, the smallest are about 10 mm.

An interesting observation: Whenever I change water in the aquarium, the shrimp start swimming around like crazy, and continue doing so for a few hours. Presumably the water change triggers their migratory instincts in some way.

At day 127 after hatching, the largest of my shrimp were about as big as the smallest Amanos one see in shops, i.e. about 25 mm, and although I will keep them until they've grown a centimeter or so more, I now consider this breeding project to have come to a successful end. With heavier feeding, and a larger grow-out tank, the shrimp would have reached marketable size faster.

As it turned out there were more shrimp hiding in the aquarium than I thought: I've sold a total of 82 shrimps, and have kept 15 for myself. This means that losses after being transferred to freshwater has been negligible.


Figure 11. Young shrimps gathering around food. The largest shrimp in the image are about 25 mm. I now, day 127 post-hatch, consider my shrimp to have reached marketable size, and with that my attempt to breed Caridina japonica has been brought to a successful end.



Addendum: After the completion of this article, I received a copy of "The complete larval development of Caridina japonica De Man (Decapoda, Caridea, Atyidea) reared in the laboratory" by Ken-Ichi Hayashi & Tatsuo Hamano. 1984. Zoological Science, 1: 571-589, and this article disagrees with what I write above on two key issues:

The size of freshly hatched larvae according to Hayashi & Hamano is 1.7 mm, which, coarsely measured against a ruler, is approximately the same as I see; my initial guesstimate of 2-3 mm was simply mistaken, and I have corrected my article accordingly. Their drawing of first larval stage correspond perfectly to my larvae (cf my fig. 5 and their fig. 1).

Secondly, and more importantly, the optimal salinity. Hayashi & Hamano report total failure to survive in salinities up to and including 8.5 ppt; optimal survival at 17 ppt (80% survival rate); and suboptimal survival (11%) at salinities up to 35 ppt (full marine salinity). This is quite different from what I found, with zero survival at salinities lower than 25 ppt, suboptimal survival at 25 ppt (3 larvae, out of at least 200, survived for four weeks without metamorphosing at which point the attempt was aborted), and high survival at 30-35 ppt (an estimated 80% reaching postlarval stage). Quite possibly (see addendum 3!) the requirements of the larvae are different in different populations.

Of note is also that salinities just slightly higher than 35 ppt seem to cause losses.

I would suggest using a salinity of 30-34 ppt for the zoeal stages; I do however suspect the losses I saw after metamorphosis was due to the postlarvae not being able to survive in full-strength seawater, and I would recommend lowering salinity to 17 ppt as soon as the first larva metamorphoses.

The reader might also find this information interesting: Hayashi & Hamano report that larvae did not begin to feed until the fourth day (first ecdysis), after which they successfully raised their larvae on a diet of "adhesive diatoms, Cymbella and Navicula, or a grated artificial diet, mixed with a small amount of rice bran", and note that the larvae were capable of foraging on the tank walls and bottom (a behavior I too have witnessed). Hayashi & Hamano failed to raise larvae on a diet of just rice bran, or a diet of Chlorella phytoplankton + rotifers.

I would suspect that diatoms is a good food for the developing larvae.


Addendum 2: I've gathered that many people find the method I've used here too complicated. I've therefore tried using a very low-tech approach, using natural light, no water changes, no aeration, no high-tech feeds. The result was a failure, but the larvae survived & grew well for about two weeks in seawater which had been added with fertilizer and kept in sunlight (basically marine "green water"). To make the larvae survive to metamorphosis supplemental feeding seems necessary - like Hayashi & Hamano report. I also accidentally found out that the larvae survive daytime water temperatures as high as 37 celsius (!!!), and nighttime temperatures as low as 16 celsius.


Addendum 3: In the AG Wirbellose mailing list, several people, e.g. Andreas Karge, have reported a shrimp which looks very similar to Caridina japonica, but which is capable of reproducing in fresh water. Allegedly this unidentified shrimp comes from Taiwan, and can be separated from C. japonica by the size of the rostrum, and on the fact that it lacks the two small blue spots C. japonica has on the uropods. Most likely this species/race/cultivar/whatever it is, will become increasingly common in the hobby; it should be much easier to breed than the "normal" Caridina japonica. I have changed the article to reflect this.


Addendum 4: The eggs hatched the 30/1 2004, and on the 13/8 the first of the new females was berried (carries eggs). That is, the generation time of Caridina japonica may be as short as six months.


Addendum 5:

An amazingly good page with information on the keeping and breeding of this shrimp:

Another very good report on a successful breeding:



Back to Projects page

horizontal rule


Mike Noren, 2004.