Both E. berryi and E. morsei were obtained from vendors or from wild collections in southern mainland Japan. Adults are available from vendors seasonally. Adults survived long-distance shipping with commercial couriers using oxygen-saturated seawater and styrofoam-insulated packaging. Adult females of E. berryi and E. morsei were collected from Mie prefecture, Japan from February to June with a set net, and transported using overnight shipping services. Animals were individually packed in 15 L round-bottom transparent 3 mm plastic bags containing 5-7 L of oxygen saturated filtered seawater with excess volume filled with pure oxygen and shipped in expanded polystyrene foam boxes similar to E. scolopes (Hanlon et al., 1997; Cecere and Miyashiro, 2022). Transit time until arrival in the lab was less than 48 h. Upon arriving in Okinawa, animals were acclimated to the temperature (20°C or 23°C) and salinity (~35 gL ^-1 i.e., parts per thousand) of our culture system. Adults from each species were housed separately. Upon spawning, eggs were removed to two-liter tanks like previous methods (Sanchez et al., 2019). Recently spawned eggs – within the first ten days post spawning - were shipped internationally within 72 h using similar methods to shipping adults apart from using 5 L bags containing 1.5-2 L oxygen saturated filtered seawater. To prevent widespread fouling, eggs were monitored on a daily basis and nonviable eggs removed. Hatchlings were housed in two liter tanks for approximately the first month after hatching.

The tank system assembled for culturing bobtail squids is a closed tank system that consists of a 200 L filter tank, five 70 L tanks (60 cm x 35 cm x 35 cm), five 2 L tanks (20 cm x 13 cm x 13 cm), two protein skimmers, an ultraviolet sterilizer, and contains filtered natural seawater from the OIST Seragaki Marine Science Station ( Supplementary Figure 1 ). Flow rate for larger tanks is 4.5 to 5 L min^-1, and 160 to 180 ml min^-1 for smaller tanks. Cleaning and partial exchange of 10% seawater are performed daily, and larger 50% water changes are performed biweekly. Water temperature from 2017 to February 2018 was maintained at 23°C with a chiller and heater and thereafter maintained at 20°C. All measurements were taken from animals kept at 20°C. The following water parameters were maintained between the ranges and monitored daily: salinity - 33 to 37 gL^-1, pH - 8.2 to 8.4. The following parameters were measured at least weekly nitrate - 0 to 20 mg L^-1, i.e., parts per million (ppm), nitrite - 0 to 0.5 mg L^-1 (ppm), ammonia (NH₃/NH₄) - 0 to 0.25 mg L^-1 (ppm).

Artificial plants, coral rubble, and PVC pieces were added to tanks to provide egg laying substrate and refuge. Beach sand collected from nearby beaches was added to aquarias. Sand was autoclaved and rinsed in reverse osmosis-treated water before introducing into aquaria. Enough substrate was given to allow the animals to bury completely. Without any burying substrate, skin lesions can form on the ventral side of the animal from friction with the aquarium floor. Aquaria were spot cleaned daily with siphons. The use of coarse sand collected from beaches or from vendors is adequate to allow the animal refuge while remaining easy to siphon.

Blue light-emitting diodes (450 nm wavelength) were used to create twelve hour light-twelve hour dark diurnal cycles in the laboratory. The photoperiod was shifted similarly to previous methods (Franklin et al., 2014) so “night” begins at noon local time to facilitate feeding and experimentation. We used red light-emitting diodes (665 nm) to observe and feed animals during “night” when they were most active.

Animals were fed ad libitum, with new shrimp added once daily. From hatching to 40 dph (days post-hatching), both species were fed mysids (Neomysis spp.). Mysids were maintained in a separate tank and fed Artemia sp. nauplii once a day prior to being fed to hatchlings. After 40 dph, both species were fed glass shrimp (Palamonetes spp.), and the freshwater marsh shrimp-Caridina spp.

Both species were also trained to consume cut frozen shrimp (e.g., black tiger prawn) upon reaching maturity.

Frozen shrimp were thawed and cut before presenting to a squid with forceps and moved to simulate living prey. Afterward, contact was made between frozen food and the inner portion of the squid’s arms until the squid either showed signs of stress or voluntarily grasped the frozen food ( Supplementary Video 1 ). After approximately one week of continuous training, squids could attack falling frozen food spontaneously.

In ongoing culture we provide adequate space, refugia, and burying substrate to minimize stress. Adults are kept in a ratio at or greater than 1:1 males to females. Our heuristic for determining appropriate space is to ensure each animal has at least two mantle lengths distance between animals. For our tank dimensions (approximately 70 L, 60 cm x 35 cm x 35 cm and water height of 31 cm), we do not exceed eight fully mature E. berryi individuals per tank (eight squids per 2,100 cm² floor area, 65,100 cm³ water volume). Assuming a maximum potential mantle length of 6 cm at maturity (Okutani and Horita, 1987), each animal is therefore given a cube of water that is 12 cm on each side, equivalent to 1700 cm³. E. morsei reaches a smaller size at maturity, with a dorsal mantle length (DML) less than 4 cm (Okutani and Horita, 1987; Reid and Jereb, 2005), and therefore can be kept at higher densities than E. berryi. To avoid reproductive attempts and aggression from males, females can be separated from males after mating. We have not observed any overt changes in behavior when adults of either species are isolated.

Thirty F1 hatchlings of E. berryi and E. morsei were isolated on the first day after hatching and reared to monitor their survivorship and growth rate. To measure growth rate, wet weight (WW, g) and DML (mm) were measured in five randomly selected individuals approximately every 10 d to maturity to prevent additional stress due to handling. We measured only five individuals each to minimize handling stress on the cohorts. Squid were placed in a transparent reservoir containing aquarium water atop graph paper and imaged using an Olympus TG-5 camera. The FIJI variant of ImageJ (Schindelin et al., 2012) was used for image calibration and DML measurement. All data is expressed as Mean ± SD. Survivorship was calculated (Leverich and Levin, 1979) as the percentage of surviving individuals, I(t) by:

Where N_s(t) is the number of survivors at time t and N_o is the initial cohort size.

At 115 dph, eleven E. berryi were removed from the study due to lack of space and prey items, and N_owas then adjusted from 30 to 19 for E. berryi. All 27 surviving E. berryi were first removed from the aquaria, then eight of each males and females were selected and placed back in the aquaria. The animals selected were the first animals to be collected. The other 11 E. berryi were removed and euthanized via overdose to the anesthetic ethanol (Abbo et al., 2021). Animals were immersed in a bath of 1% ethanol in filtered seawater. Over a period of thirty minutes, ethanol was gradually introduced until reaching a final concentration of 5% followed by mechanical destruction of the brain (Fiorito et al., 2015; Abbo et al., 2021).

Behavioral observations of E. berryi were made visually from March 2017 to May 2020. E. morsei observations were from March 2018 to May 2020. Both species were cultured during that time for other experiments.

We cultured E. berryi and E. morsei under conditions similar to those used for other bobtail squid (Hanlon et al., 1997; Nabhitabhata et al., 2005; Jones and Richardson, 2010; Sanchez et al., 2019) (Methods). We spawned wild-caught adults of both species and cultured E. berryi to the third filial generation and E. morsei to the second filial generation. Growth rate and survivorship were tracked for the first filial generation ( Figure 3 ).

Both E. berryi and E. morsei were collected during the Japanese spring season and we observed mating and spawning of wild-caught adults in the laboratory. We shipped recently spawned eggs internationally and they hatched and were raised without overt abnormalities. Both species laid a large number of eggs per clutch, usually exceeding 200 eggs for E. berryi and 100 eggs for E. morsei ( Figure 2A ). Wild-caught E. berryi laid an average of 235 ± 75.8 eggs per clutch (n=7), while wild-caught E. morsei laid 153 ± 26.5 eggs per clutch (n=4). Eggs are encapsulated within a jelly coat and laid individually in a clutch. The jelly coat of eggs of both species have an orange tint due to a dye secreted by the maternal accessory nidamental gland. Non-viable eggs have an opaque white appearance.

The period between spawning events for wild E. berryi at 20°C was 6.7 ± 2.7 d (n=18) and was not recorded for E. morsei. Both species demonstrated intermittent terminal spawning and spawned separate clutches continually once reaching sexual maturity ( Figure 3A ). On one occasion one isolated wild-caught female E berryi laid 9 fertilized clutches over a period of 59 d without any additional mating in the laboratory (presumably using stored spermatangia). E. berryi was observed to live longer in captivity after capture than E. morsei and benefited from more spawning events. E. berryi has higher survivorship and fecundity than E. morsei ( Figure 3 ).

We observed no physical or behavioral abnormalities in sequential generations; however, survivorship of later generations of E. berryi and E. morsei immediately after hatching was noticeably reduced for some clutches. On some occasions, we observed eggs laid outside of the jelly coat and some clutches with many unfertilized eggs.

Growth of E. berryi and E. morsei, measured by WW or DML, was approximately exponential for the first 90 and 60 dph ( Table 1 ) before reaching species-specific plateaus by 80 and 140 dph, respectively ( Figures 3B, C ).

We observed a large range of both DML and WW at later stages in both E. berryi and E. morsei ( Figures 3B, C ). Average E. berryi WW and DML were 17.32 ± 3.82 g and 32.14 ± 4.90 mm for males (n=16), and 23.06 ± 4.90 g and 36.26 ± 3.67 mm for females (n=14). Average E. morsei WW and DML were 1.41 ± 0.15 g and 11.45 ± 1.24 mm for males (n=6), and 3.70 ± 0.42 g and 18.95 ± 2.20 mm for females (n=6). The average female E. berryi weighed 1.33 times larger than males and were 1.13 times longer. The average female E. morsei weighed 2.63 times larger than males and were 1.65 times longer.

Survivorship was 93% for E. berryi and 80% for E. morsei for the first 30 days after hatching ( Figure 3A ). Survivorship was stable until shortly after spawning began. Thereafter, survivorship declined steadily from ~101 dph in E. morsei and ~148 dph in E. berryi. The oldest E. berryi and E. morsei in our laboratory culture were males and lived 265 dph and 169 dph, respectively. E. berryi outlived E. morsei and took longer to reach spawning age by 42 dph.

Sexual maturity was noted when males became aggressive towards conspecifics. No courtship behavior was observed in either species. Aggression appeared similar to mating, i.e., a male would assault, grapple, and possibly bite a conspecific. At 20°C this was first observed at 90 dph in E. berryi and 70 dph in E. morsei. During mating the female is first attacked by the male and the male attempts to grab the ventral head of the female, i.e., the male-to-female neck position. The male maintains control of the female and inserts the hectocotylus holding spermatophores into the mantle of the female ( Figures 2F, G ).

Spawning events began shortly after the night cycle began. Females spawned on the substrate provided including the tank walls, PVC pipes, rocks, and on imitation plants. As described above, for both species, the female laid each egg individually as part of a large clutch ( Supplementary Video 2 ). Spawning began at night and continued into the day. Females have been observed laying eggs cooperatively on the same substrate simultaneously. Females typically consumed less prey one day before spawning. Females repeatedly laid egg clutches every few days until reaching a late senescent life stage ( Figure 3A ). Some individuals spawned within three days of shipment. Mature E. berryi females were observed spawning fertilized clutches repeatedly over a period of 100 days at 20°C. No parental care was observed in either species. Egg clutch morphology was different for both species. E. morsei eggs were more densely packed in a clutch, whereas E. berryi eggs were more spaced out ( Figure 2A ).

Sexual dimorphism was visually evident at 100 dph for E. berryi and 70 dph for E. morsei. The sex of the animal can be determined by its side profile, size, suckers, and the morphology of the first left-arm ( Supplementary Figure 2 ). Males are smaller than females for both species. The size difference is more pronounced in E. morsei than E. berryi. The side mantle profile in males is sharper in males than in females ( Supplementary Figures 2A, B ). Fully mature females generally have a bulbous mantle, because of the presence of oocytes in their ovaries, and can be distinguished from males visually in a minimally invasive manner. Males of both species can further be distinguished from females by observing the first arm pair ( Supplementary Figure 2C ). Males have a modified first left arm known as the hectocotylus which is shorter than the opposing arm and curls slightly outward. Males of both species have large suckers on some rows of certain arms and modified suckers on the hectocotylus ( Supplementary Figure 2D ), whose patterns can be used to discriminate species of Euprymna (Norman and Lu, 1997). Females have uniform sucker sizes. Female first arms and suckers ( Supplementary Figure 2E ) are indistinguishable from one another (Norman and Lu, 1997).

Males of both species generally outlived females and displayed similar signs of senescence. Characteristics of early senescence include nonfunctional and faded chromatophores, greater susceptibility to infections, and loss of appetite. Signs of later stages of senescence include complete cessation of eating and burying, loss of equilibrium, and continuous labored ventilation.

The utility and prominence of E. scolopes as a model cephalopod species was discussed by Lee et al. (2009c) who also suggested that, as genomic information becomes available for different cephalopod species, the availability of broodstock and embryos becomes a primary factor in choosing a model system. Here we have explored the culturing of two related Japanese bobtail squid species, E. berryi Sasaki, 1929, and E. morsei Verrill, 1881. We find that E. berryi and E. morsei have comparable life cycles in captivity to E. scolopes (Hanlon et al., 1997), E. hyllebergi (Nabhitabhata et al., 2005), E. tasmanica (Nabhitabhata and Nishiguchi, 2014), E. parva, and E. brenneri (Sanchez et al., 2019) ( Table 2 ). Both E. berryi and E. morsei can be raised in laboratory settings and are intermittent terminal spawners, spawning repeatedly once reaching sexual maturity ( Figure 3A ). They are therefore well-suited for evo-devo studies, physiological assays, behavioral assays, laboratory culture, and have the potential to be used for gene editing (Crawford et al., 2020). E. berryi has higher survivorship and fecundity than what is reported for other sepiolids including E. morsei, E. scolopes (Hanlon et al., 1997), and S. atlantica (Jones and Richardson, 2010). These characteristics are crucial for establishing genetic lines with mutations that potentially decrease fitness.

Adult wild E. berryi and E. morsei can both be shipped using commercial couriers from their native range in southern mainland, with oxygen-saturated seawater and styrofoam-insulated packaging as described for E. scolopes (Cecere and Miyashiro, 2022), and acclimate well to aquarium conditions. E. berryi and E. morsei are usually caught using a set net round 30m deep, although both species have been collected with dip nets at night near the surface. E. berryi is also caught from the shore by recreational fishermen, and by commercial fishermen by trawling for sale to fish markets. An existing commercial fishery is potentially useful to obtain large numbers of specimens either living, for seeding propagation in captivity or studying behavior, or dead animals for morphological comparisons, isotope analysis, and population genetics. Because E. morsei is a relatively smaller cephalopod, this species is less familiar to the fishing community, which hinders the collection of wild specimens. E. morsei is also similar in size to the adult forms of the sympatric species Lusepiola birostrata and Eumandya parva making identification challenging for nonexperts (Takayama and Okutani, 1992; Bello, 2020). Differences in egg clutch morphology have been used to distinguish sympatric species (Sanchez et al., 2019), and can aid in identifying and collecting eggs in the field.

Proper care of sepiolid eggs is necessary to prevent fouling and maintain high hatching rates (Lee et al., 2009a). Females of E. scolopes host bacterial consortium in their accessory nidamental gland that is secreted to eggs during spawning to protect them from predation (Kerwin et al., 2019). We indirectly observed the same feature of E. berryi and E. morsei; specifically, we noted orange-dyed accessory nidamental glands, whose pigments are generated from carotenoids produced by symbiotic bacterial communities (Pichon et al., 2005). Eggs can be kept in incubating tanks with constant water flow in dark conditions to further inhibit microbial growth (Choe, 1966a). Our eggs were maintained at constant conditions with minimal disturbance as failure to do so can cause premature hatching and decreased survivorship (Hanlon et al., 1997).

The stocking density and male:female ratio are important to consider in cephalopod culture. Crowding has been shown to induce stress and decrease fecundity in Sepia officinalis (Forsythe et al., 2002), Sepioteuthis lessoniana (LaRoe, 1971; Boal and Gonzalez, 1998) and Euprymna scolopes (Hanlon et al., 1997). As most bobtail squid adopt benthic lifestyles quickly ( Table 2 ), the ratio of animals to floor area is also relevant. A low male:female ratio can reduce stress from mating events and prevent forced copulation. We achieved mating and spawning with a low (1:1-1:2) male to female ratio though it is preferable to separate males from females as males become aggressive, similar to E. tasmanica (Nabhitabhata and Nishiguchi, 2014).

Few studies exist on the effects of inbreeding depression on cephalopod culture. Sepia officinalis grown for seven consecutive generations developed decreased fertility in later generations, and the seventh generation failed to produce viable offspring (Forsythe et al., 1994). There are also accounts of a decreased size at maturity for cephalopods cultured to multiple generations (Iglesias et al., 2014). Thus, maintaining the genetic diversity of a colony, by careful interbreeding of separate subpopulations, or the introduction of new alleles by the steady addition of wild animals to the culture, may be necessary to support healthy laboratory colonies. Euprymna hyllebergi and Euprymna tasmanica were cultured for three generations without the introduction of wild-caught specimens. Growth rates were similar across generations and no obvious abnormalities relating to inbreeding were observed (Nabhitabhata and Nishiguchi, 2014). No physical or behavioral abnormalities were observed in both species in sequentially cultured generations; however, sometimes survivorship immediately after hatching was noticeably reduced for some clutches due to some unknown phenomenon similarly reported in E. scolopes (Hanlon et al., 1997). Additionally, rare clutches contained many unfertilized eggs and aberrant jelly coats, and more work should be done to understand and improve these traits. Based on our findings, it should be feasible to maintain a culture of both E. berryi and E. morsei for several generations. Genetic diversity can be maintained by introducing wild caught individuals seasonally (February to June) when vendors in Japan are able to supply additional animals.

Activity patterns were similar to what is described for other Euprymna species (Hanlon et al., 1997; Nabhitabhata et al., 2005; Nabhitabhata and Nishiguchi, 2014; Sanchez et al., 2019; Drerup et al., 2020) and animals became active at “night” - emerging from the substrate and discarding the sand coat for hunting and mating. During the “day”, animals bury themselves under the sandy substrate for shelter and to avoid predators (Hanlon et al., 1997; Rodrigues et al., 2010; Drerup et al., 2020). An alternating 12 hour light-dark cycle is sufficient to mimic natural diurnal cycles (Franklin et al., 2014). As for other Euprymna, E. berryi and E. morsei cover their body, head, and arms with sand but leave their eyes exposed (Nabhitabhata et al., 2005; Hanlon and Messenger, 2018; Sanchez et al., 2019; Drerup et al., 2020).

While adults can be fed frozen shrimp, hatchlings and juveniles require live food, similar to other Euprymna spp. (Choe, 1966b; Hanlon et al., 1997; Ikeda et al., 2003; Nabhitabhata and Nishiguchi, 2014; Sanchez et al., 2019). To feed hatchlings of either species, mysids can be collected from inshore locations and reared on a diet of Artemia spp. (Lussier et al., 1988). Hatchlings of both species attacked adult mysids often larger than themselves similar to other species of Euprymna (Okutani and Horita, 1987; Hanlon et al., 1997; Nabhitabhata et al., 2005; Sanchez et al., 2019). While adults were taught to spontaneously grasp frozen prey, they would sometimes ignore non-moving prey. Fully mature adult females consumed more food relative to adult males possibly due to ongoing egg production.

For both E. berryi and E.morsei, hatching from an egg clutch occurs over a period of several days. E. berryi settled and established a benthic lifestyle shortly after hatching in agreement with (Choe, 1966a) and similar to Euprymna tasmanica (Nabhitabhata and Nishiguchi, 2014) and Eumandya parva (Sanchez et al., 2019). E. berryi and E. morsei exhibited a brief nektobenthic paralarval stage similar to what is described for Euprymna hyllebergi (Nabhitabhata et al., 2005) and unlike hatchling behavior of Euprymna scolopes, Eumandya pardalota, and Euprymna brenneri (Hanlon et al., 1997; Sanchez et al., 2019) which displayed surface swimming phototaxic paralarval stages the first month after hatching ( Table 2 ). In our culture, E. berryi and E. morsei could consume prey within 24 hours after hatching, two days earlier than previously reported for E. berryi (Choe, 1966a).

E. berryi and E. morsei followed similar growth patterns to other Euprymna spp., including early growth stages of E. scolopes (Hanlon et al., 1997) and E. hyllebergi (Nabhitabhata et al., 2005) ( Table 2 ). For comparison, E. scolopes raised at 23°C experienced exponential growth from hatchling to 83 dph (Hanlon et al., 1997). E. hyllebergi demonstrated an exponential growth phase the first 30 dph when raised at 28°C (Nabhitabhata et al., 2005). E. tasmanica raised at 20°C experienced exponential growth from 7 to 44 dph was followed by approximately linear growth from 58 to 140 dph (Moltschaniwskyj and Carter, 2010). As with other bobtail squid, adult males of E. berryi and E. morsei can be definitively distinguished from females by their characteristically modified first left arm, the hectocotylus, which is shorter than the opposing arm and curls outward; female left and right first arms are indistinguishable (Okutani and Horita, 1987; Norman and Lu, 1997). Sexual dimorphism becomes visually evident ~90-100 dph for E. berryi and ~70 days for E. morsei, concurrent with aggressive behavior in males.

Both E. berryi and E. morsei recorded higher survivorship in the first 30 dph (93% and 80%) compared to the 73% survivorship reported for E. scolopes (Hanlon et al., 1997). Neither species exhibited long-lived pelagic paralarval stages after hatching, which could contribute to higher survivorship in captivity relative to E. scolopes. E. morsei was previously reared at 22.5°C and survived for 97 to 128 dph (Ikeda et al., 2003). Our E. morsei grew more slowly and lived longer, possibly due to being cultured at a lower temperature (20°C) and thus having a lower metabolic rate (Iglesias et al., 2014). E. berryi recorded the longest lifespan of any cultured Euprymna spp. ( Table 2 ).

Mating behavior is similar to what was observed in other Euprymna spp. without any obvious courtship behavior (Moynihan, 1983; Hanlon et al., 1997; Nabhitabhata et al., 2005; Squires et al., 2013; Sanchez et al., 2019; Drerup et al., 2020). Females stored spermatangia deposited during matings similarly to other Euprymna spp. (Hanlon et al., 1997; Squires et al., 2013). As females were observed laying eggs cooperatively on the same substrate simultaneously; it is necessary to separate females to track parental lineage without genotyping. Male squids were sometimes aggressive towards conspecifics; therefore, crowding should be avoided especially as squids reach sexual maturity.

Both E. berryi and E. morsei are multiple spawners similar to other members of the genera Euprymna and Sepiola (Huang, 2006; Rodrigues et al., 2011; Squires et al., 2013). Adult females of both species laid egg clutches every few days until reaching a late senescent life stage similar to other Euprymna species (Hanlon et al., 1997; Nabhitabhata et al., 2005; Squires et al., 2013). Spawning events were observed within three days of shipment of wild animals, possibly stimulated by the stress of transport (Cecere and Miyashiro, 2022). Mature E. berryi were observed spawning fertilized clutches repeatedly over a period of 100 days at 20°C. Egg clutch morphology differs across sepiolids, and may be used to differentiate sympatric species (Sanchez et al., 2019). Similarly, we found eggs more densely packed in E. morsei clutches than in E. berryi ( Figure 2A ).

Protocols established for E. scolopes are readily adapted for E. berry and E. morsei. Some existing protocols, including in situ hybridization (Lee et al., 2009e), micro-CT (Kerbl et al., 2013), immunohistochemistry (Lee et al., 2009d), and hemocyte collection (Collins and Nyholm, 2010) have already been described in E. berryi (Gavriouchkina et al., 2022), and protocols for infection with symbiotic bacteria (Naughton and Mandel, 2012), behavioral assays, injury treatment, and electrophysiology (Howard et al., 2019) are expected to be transferable from E. scolopes other bobtail species.

Established cultures of E. berryi and E. morsei will allow for comparative studies among bobtail squids in the genus Euprymna. Genomic and transcriptomic data are publicly available for both E. berryi and E. morsei (Sanchez et al., 2019) and other related species (Sanchez et al., 2021), and a genome sequence of E. berryi has recently been reported (Gavriouchkina et al., 2022). The widespread distribution of E. morsei and E. berryi in conjunction with the ability to ship adults and recently spawned eggs should allow more researchers access to these model bobtail squids, and also offers opportunities to find adaptations acquired by different populations.