Deep-sea hydrothermal vents as natural egg-case incubators at the Galapagos Rift

The discovery of deep-sea hydrothermal vents in 1977 challenged our views of ecosystem functioning and yet, the research conducted at these extreme and logistically challenging environments still continues to reveal unique biological processes. Here, we report for the first time, a unique behavior where the deep-sea skate, Bathyraja spinosissima, appears to be actively using the elevated temperature of a hydrothermal vent environment to naturally “incubate” developing egg-cases. We hypothesize that this behavior is directly targeted to accelerate embryo development time given that deep-sea skates have some of the longest egg incubation times reported for the animal kingdom. Similar egg incubating behavior, where eggs are incubated in volcanically heated nesting grounds, have been recorded in Cretaceous sauropod dinosaurs and the rare avian megapode. To our knowledge, this is the first time incubating behavior using a volcanic source is recorded for the marine environment.

including the Iguanas-Pinguinos site that is located 45 km north of Darwin Islands 14 . This particular hydrothermal site was first described in 2008 as being in a mature/waning phase with macrofauna dominated by crabs, bivalves, anemones and shrimp 14 . The area is characterized by vigorous, active venting and dispersing clouds of 'black-smoker' hydrothermal plumes.

Results
Remotely Operated Vehicle (ROV) Surveys. Our surveys using the Hercules ROV recorded a total of 157 egg-cases in and around the Iguanas-Pinguinos vent field ( Fig. 2; Supplementary material 1). Egg-cases were encountered right from the beginning of the dive when the ROV landed next to an active black smoker chimney located at 1660-1670 m depth (Fig. 3a). Over the duration of the 24-hour dive, most of the egg-cases seen were distributed within <150 m of two active black smoke chimneys (Figs 2, 3a). The largest deposition of egg-cases was on a basaltic ridge bathed in cloudy water venting from the nearby black smoker (Fig. 3b). The colors of live egg-cases ranged from golden to dark brown, suggesting that they were under different developmental stages ( Fig. 3c-f). The majority of egg-cases visible lacked evidence of fouling suggesting they were recently deposited, however there were often older spent egg-case remnants under these, indicating that this deposition site has been used for many years (Fig. 3d). Some egg-cases were located within less than a meter from an active vent chimney (Fig. 3g).

Environmental data.
A temperature probe and a CTD sensor (Conductivity, Temperature, Depth) recorded temperatures during the duration of the dive. The temperature probe was located ~10 cm above the bottom of the ROV, while the CTD was located 1.5 m above the bottom of the ROV. Mean ambient bottom water temperature was 2.76 °C ±.0.01 SD and salinity was constant at 34.6 psu. Water temperature measured between 0.6-7.1 m above the seafloor was generally higher closer to the active smokers, but diffuse venting was also observed in some areas of the vent field along the survey track. The highest number of egg-cases (58%) in both areas of the Iguanas-Pinguinos vent field were recorded within <20 m of black smoker chimneys (Table 1). Anomalous temperatures exceeding 3.1 °C were occasionally recorded as high as 2.78 m above egg-cases in some places (Fig. 4). Over 89% of the egg-cases were observed while the ROV measured temperature above the ambient bottom water temperature of 2.76 °C.

Species ID.
We collected a total of 4 egg-cases using the Hercules ROV manipulator arm (Fig. 3f). During this collection, the ROV altitude was <1 m above the seafloor and the temperature probe measured 2.9 °C. Egg-cases were large, measuring 110 mm in length (excluding horns), and the surface was rough and striated 15 . The lateral keels were narrow with 10% of the maximum egg-case width. Horns were flattened and tapered towards the tips with tips becoming thin but not filamentous. Anterior horns where shorter and wider than posterior horns, the latter being 2 times larger than the length of the anterior horns (Fig. 3f).
Based on visual examination, the egg-cases resembled those of Bathyraja spinosissima (Beebe and Tee-Van, 1941) a species previously reported associated with hydrothermal vents in the Eastern Pacific and with a depth range that extends beyond 2900 m 15 . Although we did not encounter any adult skate specimens on our survey of the Iguanas-Pinguinos vent site, B. spinosissima were observed during a previous dive at the Tempus Fugit hydrothermal vent site located about 750 km to the east, within the Galapagos Spreading Center (Fig. 3h).
In addition to the visual examination of egg-cases and ROV video footage, analysis of 603 basepairs of the 5′ mitochondrial COI region (GenBank Accession no. MF158863) resulted in a 100% identity match to a sequence in the BOLD Systems database (GenBank Accession no. FJ164384). Although this specimen captured off Vancouver Island (Canada) was reported as Bathyraja spinicauda, the reported range for B. spinicauda is restricted to the North Atlantic and a taxonomic re-assessment of this specimen confirmed it to be B. spinosissima 15 . Therefore, we conclude that the egg-cases collected were B. spinosissima.

Discussion
The presence of a B. spinosissima egg-case nursery (as described by 16 ) in an active hydrothermal vent field, where even the temperatures several meters above the substrate were often higher than ambient water, implies that this species is utilizing heat at the active Iguanas-Pinguinos vent site to incubate its egg-cases. We hypothesize that this behavior is directly targeted to accelerate embryo development times. Deep-sea skates have some of the longest egg incubation times reported for the animal kingdom, with species of the same genera, such as B. parmifera in the Bering Sea, having incubation periods of 1290 days at water temperatures of 4.4 °C 17 . Assuming the ambient water temperature of about 2.76 °C is relatively constant year-round, and a development rate similar to B. parmifera, even a conservative correlation between temperature and embryonic development would suggest an incubation time of over 1500 days. This direct relationship between temperature and development time has been reported for several deep-sea organisms, including other oviparous Chondrichthyans 17,18 . While we recorded temperature increases of <1 °C above ambient in the water above where egg-cases were abundant, these are conservative measurements considering that the temperatures reported here were collected at an average altitude of 3.5 m above the seafloor. The temperatures directly on the seafloor, where diffuse venting and conductive heating may occur, are likely to be considerably higher.
Previous evidence of egg incubation at hydrothermal sites exists in the fossil record, where a group of neosauropod dinosaurs used soil thermo-radiance and moisture of hydrothermal origin to incubate their unusually larger  eggs during the Cretaceous 19 . In contemporary times, megapode birds, such as the Polynesian megapode Megapodius pritchardii, burrow their nests in volcanically-heated soils on Niuafo' ou Island in Tonga 20 . Furthermore, several additional species of reptiles and birds actively seek specific soil temperatures to achieve optimal egg incubation. In the marine environment, aggregation of the egg-brooding deep-sea fish Pychrolutes phrictus and the cephalopod Granelodone spp. have been also recorded around "cold" seeps, where flows generate slight increases in temperature of 0.1-0.2 °C 21 . Despite these temperature anomalies, and the fact that Granelodone boreopacifica conducts the longest-known egg-brooding period of any animal (with over 1500 days of incubation time) 18 , egg occurrence was not correlated with temperature. This suggests that there may be other reasons for egg deposition in this area. For example, these cold seep sites could provide an additional food source that could influence the location of egg deposition 21 .
In addition to decreasing egg incubation periods, this is the first record of a hydrothermal vent habitat serving as an egg-case nursery site, a discrete habitat with extremely high densities of egg cases when compared to surrounding similar habitats 16 . Among the elasmobranchs, skates are the only group known to be strictly oviparous, where females produce large collagen egg cases containing a large yolk mass 22,23 . Egg-case nursery sites for members of the genus Bathyraja have been previously identified across most ocean basins and in diverse habitats like rocky reefs, submarine canyons, seamounts or even methane cold seeps 17,24,25 . Hydrothermal vent sites may also have other advantages as a juvenile nursery, although previous studies have revealed that juvenile skates of other species of the genus Bathyraja leave egg-case nurseries after hatching 16,26 .
One out of four species of Chondrichthyans are threatened with extinction, mainly as a result of over-fishing 27 . Deep-water Chondrichthyans species are among the least productive given their long turnover times (i.e. slower growth, later age at maturity and increased longevity) and, as a consequence, they may have higher extinction risk than other oceanic and continental shelf species 28,29 . Therefore, understanding their reproductive processes and key habitat requirements is vital to predict population stability and inform effective conservation strategies, especially under conditions of global change 17 . In March 2016, the Ecuadorian government created a 40,000 km 2 marine sanctuary to protect unique underwater communities around Darwin and Wolf islands 30,31 . This fully non-extractive reserve also protects adjacent seamounts, and it includes the Iguanas-Pinguinos vent site, thus protecting the first known nursery for deep-water predators associated with active hydrothermal vents. In 2015, the North Pacific Fishery Management Council designated eight deep-sea skate egg-nurseries in the eastern Bering Sea as habitat areas of particular concern, and this represented the first official recognition of this habitat type worldwide 16 . Further research should focus on identifying and promoting the protection of additional Chondrichthyan deep-sea nurseries, given the continuous expansion of fisheries towards the deep-sea and the intrinsic vulnerability of this group of species [32][33][34] .   ROV surveys. Exploration of the seafloor was carried out using the two-body Remotely Operated Vehicle (ROV) system Argus and Hercules, each rated for 4 km water depth. Video and still images of the sites were acquired using Insite Pacific Zeus Plus HD color video cameras on both vehicles, each equipped with a 10× mechanical zoom lens. CTD data was recorded using a calibrated Seabird FastCAT49 equipped with a circulating pump. The temperature probe used was an Omega PT100 RTD sensor housed in a custom titanium sheath (designed and fabricated by the Woods Hole Oceanographic Institution), which was calibrated against the Seabird FastCAT49 unit. These two units reported temperatures within 0.04 °C of each other. Seafloor navigation was performed using a combination of sensors including a LinkQuest Tracklink 5000 USBL system, RDI Workhorse Navigator 600 kHz DVL, IXSEA OCTANS 6000 fiber-optic gyrocompass, and a calibrated Paroscientific DigiQuartz depth sensor. DVLNAV software was used to process these data 35 .
Sample collection and genetic analysis. Egg-case samples were collected using the ROV manipulator arm and placed on sample boxes aboard Hercules ROV for recovery. Once aboard the ship, two egg-cases were opened to sample for molecular analyses and further examination. Both egg-cases were at a very early development stage, with no clear presence of the embryo. DNA was extracted from egg-cases with the QIAGEN DNeasy Blood & Tissue Kit (QIAGEN Inc., Valencia, CA). The 5′ region of the mitochondrial COI gene was sequenced on an AB 3130 genetic analyzer via the primer pair FF2d (5′-TTCTCCACCAACCACAARGAYATYGG-3′) and FR1d (5′-CACCTCAGGGTGTCCG AARAAYCARAA-3′) following the protocol outlined in 36 . Sequences were assembled in geneious 7.1.7 37 and subsequently compared to species level barcode data in the BOLD Systems database (www.boldsystems.org).