Life in the Open Ocean. Joseph J. Torres

Siphonophores are quite delicate, particularly prone to damage from midwater trawls and even gentle plankton nets. The damage usually involves breaking the animal into its constituent parts, and with particularly delicate forms, the damage may preclude identification altogether. Delicate forms can literally pass through the meshes of a scientific net. Thus, conclusions drawn from a net sampling survey will tend to be biased toward the more robust species. Surveys done primarily with visual observations, by either SCUBA or submersible, will also have bias, usually toward slower moving, larger species. Both types of sampling have their shortcomings, but without question, far more sampling has been done with nets. Trawling techniques have been around longer, trawls sample greater volumes, and they are far cheaper to execute. It is important to recognize the pros and cons of all sampling methodologies and to take away from each what is most useful.

      Pugh (1999) observed that calycophophoran species tend to dominate in net samples, but that over 70% of the specimens collected by submersibles are the larger, more delicate, and more highly pigmented physonects. An interesting, though disturbing, corollary to his observations on relative numbers was that about half of the physonects collected by submersibles were new to science. Clearly there remains a lot to learn about the siphonophores.

Data from Pugh (1999) for 93 widely distributed Atlantic siphonophore species are summarized in Table 3.9. Forty‐one species are considered to be mainly epipelagic (0–250 m), 17 are epipelagic–upper mesopelagic (<100 m to >250 m), 31 are mesopelagic (200–1000 m), and 4 are bathypelagic (>1000 m). The best information to date suggests that the majority of siphonophores reside in the most productive upper 250 m of the ocean.

Table 3.9 Vertical distribution of Siphonophore species in the South Atlantic.

      Source: From the data in Pugh (1999).

Primary depth range	Number of species
Order Cystonectae	Order Physonectae	Order Calycophorae
Epipelagic (0–250 m)	3	4	34
Epi‐upper Mesopelagic (<100 m to >250 m)	0	5	12
Mesopelagic (200–1000 m)	0	7	24
Bathypelagic (>1000 m)	0	0	4

      Diurnal Vertical Migration

      As discussed, swimming ability within the siphonophores varies widely. The idea that siphonophore populations move up to the surface at dusk and back to depth at dawn in response to the waning and waxing illumination in near‐surface waters is not a compelling one, especially for the suborders with more limited mobility. Nonetheless, there is a substantial amount of data for many species that suggest precisely that. Moore (1949, 1953) reported vertical excursions of 30–40 m for many species of calycophorans in both the Florida current and in the vicinity of Bermuda on a day–night basis. Similar results were obtained by Musayeva (1976) for calycophorans in the Sulu Sea.

An alternative to a directed vertical migration triggered by sunset and sunrise is a slowly undulating change in vertical profile over a 24‐hour period (Pugh 1977). Siphonophores gradually move up and down with the changing photoperiod. This sinusoidal pattern of migration would explain the unusual depth profiles obtained for some weakly swimming species such as Hippopodius hippopus. The theory is still somewhat speculative.

      Without question, a changing vertical distribution over the diel cycle is a characteristic of many siphonophore species. However, even among the calycophorans the vertical excursions are usually quite limited in scope (<50 m), a situation to be expected in an order that is morphologically adapted more for ambush predation than long‐distance swimming. Because of their float, the physonects are not only good acoustic targets, they face the same problems that fish with swimbladders do when moving vertically: expansion and compression of gas in their flotation system when moving up and down in the water column.

      Geographical Distribution

Siphonophores are pan‐oceanic; they are found from the equator to the polar oceans in all the major ocean basins. Mackie et al. (1987) concluded that siphonophore distributions coincide reasonably well with the biomes and provinces described in the biogeography literature (e.g. Briggs 1995; Longhurst 1998), which in turn coincide with the earth’s major climatic zones. The best data describing faunal relationships with latitude and longitude come from Atlantic waters. Margulis (1976) recognized seven faunal groups among the siphonophores of the Atlantic, which correspond well to the latitudinal zones of Briggs (Figure 3.36): Arctic species, Northern Boreal species (cf. cold temperate), Antarctic species, Bipolar species (found in the Arctic and Antarctic), Tropical species (which include equatorial and warm temperate components), Eurybiotic species (those which live in all biogeographic areas), and Neritic species. An excellent compendium of Atlantic species’ latitudinal ranges may be found in Pugh (1999). Most (60 of 96 species) were very wide‐ranging, with distributions encompassing 80° of latitude or more, roughly half on each side of the equator.

      Mackie et al. (1987) provide a summary of diversity and numbers for 21 common species of calycophoran siphonophores in the North Atlantic. The data show a peak in both numbers and diversity at about 18 °N with a gradual decline in species numbers further north. A second peak in abundance is obvious between 40 and 53 °N.

      Organization and Sensory Mechanisms

      No sensory apparatus has been detected in the siphonophores, i.e. no ocelli, statocysts, or mechanoreceptors such as those observed in the medusae. However, siphonophores are sensitive to touch, light, chemicals, and, in some cases, to waterborne vibration. How? It is likely that the nerves themselves act as receptors although mechanisms effecting the receptor‐like responses are undescribed.

      Two types of conduction are recognized in the siphonophores, epithelial and neural. Both contribute to coordinated movement and responses. Epithelial conduction is similar to the spread of depolarization in myogenic hearts and is present in the nectophores of physonects and calycophorans. Epithelial conduction was effectively demonstrated in Nanomia when its nectophores remained coordinated after severing their nervous connection to the stem (Mackie 1964).

Epithelial conduction as a mechanism for propagating impulses is confusing at best. There is no obvious morphological distinction between epithelia that are capable of conduction and those that are not. Before the techniques of neurophysiology were available, the presence of epithelial conduction was inferred by the absence of nerves coupled with the presence of coordinated activity (Mackie et al. 1987). Presumably, the epithelial cells themselves possess the membrane channels that are necessary for ion movement and signal propagation. Since all animal cells maintain an ionic disequilibrium with the medium bathing them, whether that medium is seawater or blood, the basic ion transport “equipment” is already in place within the membrane. The signal must spread from cell to cell to function as a conductive pathway. How it happens is still undescribed.