Starfishes and brittlestars are anatomically very similar and were formerly grouped together in a single class, the Stelleroidea, which included all star-shaped echinoderms. This simple classification, which was still in use in 1900 (when Gregory employed it in Lankester's Treatise of Zoology) was subsequently displaced by another, in which the starfishes (Asteroidea) and brittlestars (Ophiuroidea) were elevated to rank as distinct classes. The change was introduced on the grounds that embryological investigations suggested that asteroids were more closely related to holothuroids, since their larvae were similar, whereas ophiuroids had larvae which resembled those of echinoids. This view, although still maintained by Hyman in 1955. is no longer acceptable to many students of echinoderms. Fossil studies have shown that it is extremely unlikely that asteroids are more closely related to holothurians than to ophiuroids, or that ophiuroids could be more closely related to any known echinoids than they are to asteroids. Indeed, it can hardly be disputed that both asteroids and ophiuroids descend from common Palaeozoic ancestors which are now known to us -- namely, those star-shaped forms called Somasteroidea. Thus the larval evidence must have been misinterpreted, and requires reassessment.

After a lifetime of more than fifty years devoted to the study of fossil echinoderms, W. K. Spencer (Phil, Trans., 1951) set out in his last paper a new classification of the star-shaped echinoderms; in it, he revives the older grouping Stelleroidea, ranking it as a single class, and including in it three subclasses, two of them corresponding to the familiar asteroid and ophiuroid groupings, the third one comprising the Somasteroidea. This treatment has been adopted in the Anglo-American Treatise of Paleontology, now in preparation by a group of specialists, and probably will remain standard for some time to come. It may be noted that these authors prefer to use the name Asterozoa instead of Stelleroidea, parallelling another assemblage of echinoderms termed the Echinozoa. Diagnoses of the three groups of Asterozoa are too technical for presentation here but the following notes may serve to illustrate their character:

Class: ASTEROZOA— Free-living echinoderms with a depressed, stellate body; the ambulacral skeletal plates lie internally with respect to the radial ambulacral vessels; the madreporite is externally visible, but is not incorporated into any system of apical plates.

• Subclass (1): SOMASTEROIDEA— Extinct forms with no ambulacral groove, and with the arms barely differentiated in any way from the rest of the body; probably ‘ suspension-feeders ‘, i.e. stationary animals resting upon thepage 128
• sea-floor, depending for food upon the plankton-shower or upon the organic content of the uppermost layer of the sea-bed, the food particles being swept into the mouth by centrally directed water-currents; Lower Ordovician onward, probably ancestral to the two following subclasses.
• Subclass (2): ASTEROIDEA— Carnivorous or detrital-feeding forms, usually with rather stout arms which are not sharply demarcated from the centre of the body, and with an open ambulacral groove traversing the lower side of each radius; alimentary organs, and often also the reproductive organs, usually extending into the arms; food seized by the arms or engulfed into the everted stomach; Palaeozoic to present time.
• Subclass (3): OPHIROIDEA— Selective detritus-feeders, usually with a disc-shaped central body from which radiate the long, clearly demarcated arms; earlier types had an open ambulacral groove, like that of asteroids, whereas later types and all extant forms have the groove closed over and reduced to an internal canal; alimentary and reproductive systems usually confined to the disc; Palaeozoic to present time.

The foregoing paragraphs have referred to the general systematic position of starfishes. The following are some notes on the habits and ecology of starfishes, and their relation to other animals.

Starfishes usually creep about the sea-bed with the aid of their tube-feet. No swimming forms have yet been recorded. Those species which have suctorial tube-feet are able to climb and descend obstacles with their aid, and capture food. A number of genera have peculiar peg-like tube-feet (Fig. 2A), without suckers; they usually occur on submarine mud-banks, where they seem to use their tube-feet as oars or as stilts perhaps— I imagine them as tottering or paddling through the semi-fluid, slimy mud.

Although starfishes cannot see, since they have no eyes, they are able to detect light-fluctuations with the aid of a photosensitive eye-spot at the outer end of the ambulacral groove where it reaches the tip of the arm. Starfishes cannot be said to ‘ hear ‘, though they evidently detect the grosser vibrations often associated with sound. At the tip of the arm are some special tube-feet in some starfishes, which are held erect when the animal is active, and probably serve as taste-organs (chemoreceptors).

The sex of a starfish is seldom obvious without dissection, but in those forms which carry or brood the young, the female can be recognised during the breeding season (see Fig. 42). Their breeding habits are very varied. Apart from brood-protecting forms such as Calvasterias in New Zealand littoral waters, there are deep-sea forms such as Pteraster (Fig 28) with a dorsal brood-chamber developed on the upper side, provided with a water-conditioning system not unlike the water-circulatory system of a sponge, and having a large central osculum for the outflow of the water-current. One Greenland genus, Leptychaster, somehow contrives to hatch its eggs inside its stomach without accidentally digesting them. Many starfishes have free-swimming larval stages, others lay yolky eggs which undergo a direct development.

page 129

Starfishes are probably short-lived animals with a life-span of only a few years. Some, at least, are known to be sexually mature after one year, though growth continues for about four years. Some kinds can regenerate lost arms; Luidia (Fig. 7) fragments upon even gentle handling, and must spend much of its life regenerating its lost members, to judge by the unevenly matched arms its species often exhibit. In New Zealand waters another genus. Allostichaster (Figs. 38-40) regularly reproduces by asexual transverse fission. In the tropics Linckia can regenerate the whole animal from a single arm. producing curious ‘ sea-comets ‘ in the process; in Cook Strait a species of Sclerasterias sometimes shows the same habit.

The brilliant colours of starfishes are due to biochromes which usually fade on preservation, and are chemically changed by alcohol. The colours include orange or vermilion (the two commonest), crimson (deep-sea forms), yellow, green, blue, violet, brown, as well as variegated patterns.

There are usually five arms, but many species have more than five, extreme cases occurring among the deep-sea Brisingidae, where as many as 44 long fragile arms may be present. Square starfishes with only four arms are abnormalities, but by no means rare. The largest starfishes (Linckia spp.) reach a metre in diameter; the biggest New Zealand species, Astrostole scabra (Fig. 41), is about half this size.

Starfishes are fast-growing animals and require a great deal of food. The carnivorous types will attack any animal which they can swallow whole, and will feed upon pieces of larger animals if the stomach can be applied to the food. Some species with suctorial tube-feet attack bivalve shellfish and sea-urchins. The valves of shellfish are forced apart by long-sustained suction exerted by the tube-feet and the arms; in this way members of the family Asteriidae do much damage to oyster fisheries.

Starfishes are not known to be dangerous to man, though tropical species occasionally cause septic wounds in the foot, as in the case of the Fijian genus Acanthaster, a coral-reef dweller with sharp spines. No New Zealand species is likely to cause injuries of any kind. Starfishes are not known to be poisonous to eat, though they are said to be bitter and unattractive. Of the few animals which do use them as food the walrus is perhaps most notable.

A number of parasites attack starfishes. A sea-snail, Stylifer, bores into their skin; a polychaete worm, Achloe, inhabits the ambulacral groove, a cirripede destroys the sex-gland of Coscinasterias and others, and in the tropics a slender fish, Fierasfer, inhabits the body cavity (apparently entering the mouth of its host, and boring through the stomach wall).

There are about 1,500 known species of starfishes, inhabiting the coasts of all known seas, but avoiding brackish and fresh water. Some live at low-tide level, or float on sea-weed at the surface of the sea (Calvasterias. Fig. 42); others burrow into the sea-floor. Twelve families are known to range into waters more than two miles deep, and one species (Porcellanaster caeruleus) has been recovered from the sea-bed at a depth of four miles. Next to holothurians, starfishes are the deepest-dwelling animals of the sea —

page 130

probably because, like holothurians, they can extract nutriment from mud. Individual species usually have rather restricted bathymetric ranges, but where the continental slope is relatively steep, as it is off New Zealand, deep-water forms often ascend the slope, and shelf forms often tumble into deep water; thus the content of Cook Strait hauls is rather unpredictable.

From New Zealand waters 35 genera, embracing 46 species, are now known; six families have been added in the last few years, and one other family. the Porcellanasteridae. will probably be discovered here soon (it has been included in the key). It would now appear that there is a larger Indonesian and Australian element in the fauna than was formerly supposed, especially in the northern half of New Zealand. The supposed South American affinites seem to be negligible, and are perhaps a result of the epiplanktonic drift of seaweed-inhabiting forms, or the dispersal of planktonic larvae. Asterodon, a genus well represented in New Zealand waters, may have contributed a species to South America by way of the west-wind drift and Humboldt current. Certainly the main affinities of the New Zealand asteroid fauna are with the Indo-West-Pacfic, not with South America. A few deep-water species resemble North Atlantic forms; perhaps they are really cosmopolitan.

To use the key which follows, take each numbered paragraph consecutively. so long as the characters of the specimen in hand agree; if a character differs, go directly to the number shown in brackets. The key is complete to date, and is substantially systematic, so that the taxons occur in their accepted natural order. The length of the arm is given in the form of the major radius (R), measured from the centre of the disc to the tip of the arm, and is for an adult specimen. The colour in life is given, when known, and omitted if only preserved material has been described. The series of illustrations is representative, though not complete. In many cases it should be possible to make an approximate determination merely by inspection of the illustrations; the identification should then be checked against the key. Since many species undoubtedly remain to be discovered, a final determination can only be made against a complete description of each species. Unfortunately the literature is too scattered for citation here.

Subclass Asteroidea

page break page 134