September 17, 1908] 



NA TURE 



491 



protrusion, therefore, a volume of water corresponding with 

 that of the protruded organs is admitted into the com- 

 pensation-sac, the dilatation of which, by means of 

 radiating muscle-fibres, is the cause of the protrusion, and 

 is again evacuated when the polypide is retracted. These 

 alternate actions of filling and emptying the compensation- 

 sac with water from the outside are probably of import- 

 ance in the respiration of the animal. The advantages of 

 having avicularia in such a position that they can guard 

 the orifice from which the tentacles are protruded and 

 that of the compensation-sac are too obvious to require 

 detailed discussion. 



The avicularia probably afford little if any protection 

 against the attacks of the larger foes, such as Fishes, 

 Echinids, and Nudibranch Molluscs, which are said to 

 browse on Polyzoa. But there is one group of enemies 

 against which the opercula and the mandibles are probably 

 particularly effective. These are encrusting organisms, 

 including other species of Polyzoa ; and indeed the enemies 

 against which a Poh'zoon has to provide arc probably in 

 a special degree the members of its own class. 



In many Polyzoa which afford large surfaces suitable 

 foi the growth of encrusting organisms the older parts of 

 the colony, where the opercula and mandibles arc no longer 

 in working order, do actually harbour large numbers ol 

 encrusting Polyzoa, Sponges, Ascidians, and othei 

 organisms. These are usually absent in the active parts 

 of the colony nearer the growing margins. In these posi- 

 tions the only animals which obtain a footing are such 

 forms as the Infusorian Folliculina, adapted by its minute- 

 ness to find a place between the defensive appendages, or 

 such organisms as are attached by means of delicate 

 creeping stolons or rootlets that can find their way between 

 the opercula and mandibles without being damaged by 

 them. A branching species fixed by a narrow base may 

 do little harm to a Polyzoon on which it is growing. But 

 the effects of an encrusting species would be different, 

 since the orifices of the colony which is being overgrown 

 would be occluded, and the polypides entirely cut off from 

 the outer world. Although experimental evidence is at 

 present wanting to prove this view, I have little doubt that 

 the avicularia are specially important in preventing the 

 fixation of the larva? of encrusting species. The larva is 

 of course very vulnerable, and it cannot become the founder 

 of an adult colony unless it can find a secluded spot in 

 which it is safe from undue disturbance during the critical 

 time of its metamorphosis. The avicularia are well adapted 

 by their form for warding off larvs. Those that have the 

 so-called " duck-billed " or '* spatulate " form are in many 

 cases large enough to catch or crush a larva without 

 difficulty, while those which have a mandible terminated 

 by a spike-like projection must be even more destructive 

 to the life of any minute organism which is so imprudent 

 as to stray within their reach. In some of the avicularia 

 belonging to this latter type the mandible is strongly com- 

 pressed along the greater part of its length, and may then 

 assume the shape of a knife-blade, with a sharp cutting 

 edge and a thicker back. The blade shuts down into the 

 calcareous rostrum of the avicularium in such a way that 

 its action may be compared to that of a pair of scissors. 

 It cannot be doubted that this form of avicularium has 

 a high protective value. 



In some cases the mandibles or the opercula are toothed. 

 The teeth are specially strong in certain species of the 

 genus .Steganoporella, where the opercula become most 

 formidable weapons. The large development of the occlusor 

 muscles proves that the closure of these opercula must 

 take place with much force. 



The protective value of the avicularia may be illustrated 

 by the distribution of these organs in Rctepora, the species 

 of which usually grow in the form of a calcareous net- 

 work, with oval meshes or ** fenestra? " betw'een tlie 

 branches. These are furnished w-ith an elaborate armature 

 of avicularia. which usually occur in large numbers and 

 in considerable variety of form and size. .Some of them 

 are scattered over the front surface, on which alone the 

 zooecia open, while others occur on the more unprotected 

 backs of the branches, where there are no zocecial orifices. 

 To guard against an attack from the rear the margins of 

 the fenestrre are frequently furnished with avicularia, 

 among whi'^h some are usuallv of a specially large size, 



NO. 2029, VOL. 78] 



and are well situated to intercept any larva or adult animal 

 that might attempt to pass through a fenestra. 



A healthy Retepore is usually completely free from 

 encrusting organisms in regions where the avicularia are 

 functional. One of the few exceptions I have noticed to 

 this rule is specially instructive. In this case a small 

 encrusting Cheilostome colony is growing near the edge of 

 the Retepora frond. The primary individual or ancestrula 

 of the encrusting colony is readily distinguishable, and its 

 position shows that the larva from which it was formed 

 must have attached itself to the growing margin of the 

 Retepore, a region in which the avicularia were not fully 

 formed. Having thus established itself, the colony has 

 succeeded in invading a small region of the adjacent parts 

 where the zooecia are still vigorous and healthy. A dead 

 Retepora, on the contrary, forms a substratum which is 

 well adapted for the growth of various organisms, sucfi 

 as other Polyzoa, Sponges, Hydroids, Compound .Ascidians, 

 and Foraminifera. 



Although the avicularia are thus effective in preventing 

 the overgrowth of the colony by most of the organisms 

 that might otherwise settle there, there are one or two 

 animals of suitable habit which have succeeded in establish- 

 ing themselves in the very midst of the defensive works. 

 In species of Retepora from the Malay Archipelago ' I find 

 that a considerable proportion of the colonies are infested 

 by a Gymnoblastic Hydroid of Syncoryne-like appearance. 

 The association of this with its host is so intimate that 

 the hydrorhiza becomes completely included in tunnels 

 formed in the calcareous mass of the Polyzoon, where it is, 

 of course, safe from the avicularia. These tunnels, the 

 walls of which are secreted by the Polyzoon, open to the 

 exterior by tubular apertures situated on the margins of 

 the fenestras and on other parts ; and they are so definite 

 in their appearance, and often so regularly arranged, that 

 it might be difficult to believe that they were not a normal 

 feature of the Retepora were it not possible to demonstrate 

 their relation to the Hydroid." 



There is one other organism which has a definite relation 

 to colonies of Retepora in Malay w.'iters. This is Loxo- 

 soma, a stalked Polyzoon which leads a practically solitary 

 life owing to the fact that its buds break off as soon as 

 they have reached maturity. The Loxosoma no doubt 

 succeeds in enjoying a secure existence, even though it is 

 surrounded by avicularia, owing partly to its stalked form 

 and partly to its minute size. It is commonly found in 

 considerable numbers, and often attaches itself in such a 

 way that it projects into one of the fenestra;, where it is 

 as far as possible removed from the dangerous neighbour- 

 hood of the avicularia. 



We have thus seen that, while the species of Retepora 

 are adequately protected against many encrusting or 

 epizoic organisms, there are one or two animals that have 

 succeeded in evading the attacks of the avicularia, which, 

 it must be remembered, arc rigidly attached to the colony, 

 and cannot go in search of any enemy that keeps out of 

 their way. The efficient avicujarian protection may well 

 be responsible for the fact that Retepora is a common 

 and widely distributed genus, flourishing in both shallow 

 and deep water. Not only is it found in large numbers 

 in the most diverse localities, but it has differentiated 

 itself into a large number of species, among which 

 avicularia occur in great profusion and in a great variety 

 of forms. But so soon as the avicularia cease to he active 

 we find that numerous organisms settle on the unprotected 

 branches ; and a dead colony of Retepora is accordingly 

 usually found to be invaded by numbers of other animals. 



One other familiar case may be mentioned illustrative of 

 the means by which a Polyzoon may be protected from 



^ The greater number of the facts referred to in this Address have been 

 observed during my study of the Polyzoa collected during the Sibogn 

 expedition. 



- It maybe noted, as has already been done by Alcock ("Ann. Mag. 

 Nat. Hist.,"ser6, X., 1802, p. 207), that many other c^ses are known in 

 which there is an association between a Gymnoblastic Hydroid and some 

 other animal. The interesting case of the association of a Gymnoblastic 

 Hydroid (Stylactis) with a fish (Minous) described by .Alcock has also been 

 described, more recently, by Franz and Stechow {Zool. Anzeiger, xxxii., 

 1Q08, p. 7S2). Another ca^e of the assoc'ation of a Coslenterate with a 

 Polyzoon has been recorded by Haswell and by Kirkpatrick, who have 

 called attention to the occurrence of a small '* .\ctinid " which forms definite 

 cavities in a massive calcareous Cheilostome from Australian waters. 

 There is in this case no satisfactory evidence to show what the Ctelenterate 

 really is. 



