Siphonophorae
Methods

Most of our knowledge concerning the zoogeographical distribution of siphonophores has come from animals caught in a variety of nets, ranging from simple surface dip-nets to sophisticated deep-water trawls. However, the complexity and fragility of these animals mean that the whole structure can easily be broken apart, and typically this is what happens during net collection. This also means that many pieces can be lost through the meshes so that there can be problems in quantifying their abundance. This is particularly the case for physonect siphonophores as many of their potentially myriad parts can be lost. The number of floats, or pneumatophores, can give some indication of the number of specimens collected but the physonect pneumatophore is relatively small and is rarely retained in net samples. This problem is further discussed by Pugh (1984).

The majority of siphonophore species belong to the Order Calycophorae and these are the ones that are most frequently caught by nets. These species do not possess a pneumatophore but, fortunately, most of them possess other unique "individual" parts that can enable a reasonably accurate estimate of their abundance to be made. This applies to both the polygastric (asexual) and eudoxid (sexual) stages. There are a few species, in the subfamily Prayinae, that develop two identical swimming bells or nectophores. In this case the numbers of nectophores need to be halved in order to quantify the number of specimens. Similarly, in the family Hippopodiidae multiple (up to ca. 15) nectophores are developed. In this case, the exact number per animal can vary, and some of the less well developed ones may be lost, and so quantification is more complicated. Again this is discussed further by Pugh (1984).

Since nets do not sample all species adequately then the importance of the total siphonophore population can be greatly underestimated. In general, net samples indicate that calycophoran siphonophores are more abundant in superficial waters and, in the open ocean, may reach densities of ca. 1 per mö3, while in the top 1000 m of the water column they average about 10 per 1000 m^3. However, in inshore waters concentrations of a single small species can exceed 500 per mö3 (e.g. Greve, 1994).

The best, and most exhilarating, way to study and collect siphonophores is to use an in situ technique such as SCUBA diving or submersibles. It is the only way by which the full beauty of these animals can be appreciated. In the case of SCUBA, the divers take down wide mouthed jars into which, with a lot of skill, the animals can be induced to swim. Shipboard experiments can then be carried out on the specimens, using large tanks. Nonetheless, siphonophores like their freedom of movement and so do not usually survive long under these conditions. "Blue water" SCUBA diving techniques have been described by Hamner (1975) and Madin and Swanberg (1984). Many supposedly rare, but actually very fragile siphonophore species have been collected by this means, as well as several new species.

At deeper depths, submersibles with sophisticated collecting devices need to be used. Ones that have been used extensively in the recent past are the Johnson-Sea-Link I and II because of their ability to collect many delicate specimens intact (see Youngbluth, 1984), although they do have a depth limitation of ca. 1000 m. Even so, some gelatinous species still cannot be described because they simply disintegrate "before your very eyes", probably as a result of turbulence produced by the submersible. Such in situ studies have shown that even relatively large physonect species can be quite abundant; for instance Nanomia cara, which is about 20-40 cm in length, can occur at densities of 7-8 per m^3 (Rogers et al., 1978).

Personal observations (Pugh, 1989) have shown that in contrast to nets, where calycophoran species predominate, over 60% of the siphonophore species and 70% of the specimens collected by submersible belonged to the order Physonectae. There is an obvious reason for this, in that the physonects are generally larger and more highly pigmented than the calycophoran species and, thus, much easier to observe. Nonetheless, it is also true that over half of the physonect, and a quarter of the calycophoran, species that have been collected have proved to be new to science. Again, species that had been thought to be rare are shown to be common. For instance, Pugh and Harbison (1986) found that Lychnagalma utricularia (Claus, 1879) was the commonest physonect species collected in the vicinity of the Bahamas, despite the fact that there have been no other substantiated records for this species since it was first described.

When siphonophores have been collected in nets it is best to fix and preserve the whole sample as early as possible in 5% formalin buffered by borax. After a few days, the formalin should be changed, but it is best otherwise not to disturb the sample so as to allow proper fixation of the animals. Further details can be found in the Ostracoda chapter (this volume). The siphonophores inevitably will shrink, and volume loss is rapid, with about a 50% reduction within the first 2 hours of preservation (Pugh, pers. obs.). However, most of this is due to a loss of turgor so that the interstitial water that previously was retained within the main cavity of the nectophore, the nectosac, is lost. Nonetheless, and in contrast to a previous statement (Kirkpatrick and Pugh, 1984), it is best to leave the specimens to fix completely before sorting them from the remainder of the sample. As formalin is an unpleasant substance, during the sorting of our samples they are transferred into Steedmann’s preserving fluid, which contains only 1% formalin (Steedman, 1976). Nonetheless, some species are so delicate that they disintegrate when preserved in formalin. In such cases better success has been achieved using ca. 3% glutaraldehyde. Alcohol is not recommended for siphonophores.

It is very difficult to preserve siphonophore specimens intact. The suggested method is first to anaesthetise the specimen by the slow, dropwise, addition of an isotonic magnesium chloride solution to the sea water, followed by a further dropwise addition of 10-20% formalin.

Once preserved and sorted, the parts of the siphonophore specimens can be examined under a low magnification binocular microscope, preferably using dark background illumination. Many of the component parts of a siphonophore species have characteristic morphologies and, with a practised eye, can be readily identified. They can also be stained using a variety of stains, for instance Steedman’s Triple Stain or Borax Carmine. Staining can be useful if dark background illumination is not available, or when there is a difficulty in identifying the specimen, as the stain will help to show up certain critical morphological features.