The Associates of Four Species of Marine Spongesof Oregon and Washington

EDWARD R. LONGl

ABSTRACT : Four species of sponge from the coasts of Oregon and W ashingtonwere studied and dissected for inhabitants and associates. The four species differedin texture, composition, and habitat, and likewise, the populations of associates ofeach differed, even when samples of two of these species were found adjacent to oneanother. Generally, the relationships of the associates to the host sponges were offour sorts: (1 ) inquilinism or lodging, either accidental or intentional; ( 2) preda-tion or grazing; (3) competition for space resulting in "co-habitation" of an area(i. e., a plant or animal growing up through a sponge) ; and (4) mutualism. Fish eggsin the hollow chambers of H omaxinella sp. represented fish-in-sponge inquilinism,which is the first such instance reported in the Pacific Ocean and in this sponge.The sponge Halichondria panicea, with an intracellular algal symbiont , was foundto emit an attractant into the water, which Archidoris montereyensis followed, inbehavior experiments, in preference to other sponges simultaneously offered. Atotal of 6,098 organisms, representing 68 species, were found associated with thesamples of Halicbondria panicea with densit ies of up to 19 organisms per cm3 ofsponge tissue. There were 9,581 plants and animals found with Mic rociona prolif era,and 150 with Suberites lata.

SPONGES FREQUENTLY SERVE as hosts for manyplant and animal associates. The relationshipsof such organisms vary from loose lodging orinquil inism to mutualism, a close type of sym-biosis. The first account of sponge symbiosiswas that of Radcliffe (19 17) in which he re-ported the goby Garmannia spongicola livingwithin unidentified sponges off N orth Carolina.Other early reports include those of Vosmaer(191 1) who found invertebrates in the canalsof tropical sponges, and Ridley and Dendy(1887) who found a very abundant oscillator-ian alga in H alicbondria panicea.

Perhaps the most significant reports concern-ing sponge symbioses were those of Pearse( 1935, 1949) who described the inhabitantsof various Caribbean sponges, Gudger (19 35)who found many fish living in Atlantic sponges,and Forbes (1964, 1966) who described themutualistic relationship between the spongeStellate gm bii and the oyster Ostrea permollis

1 Department of Zoology, Oregon State Un iversity.Present address: U.S. N aval Oceanographi c Office,W ashington, D .C. 20390. Manuscript received Jul y18 , 1967.

from the Gulf of Mexico. Caullery (1952) ,Dales (19 57) , Nicol (1 964 ) , and Henry( 1966) have summarized the symbiotic relation-ships of marine animals and included in theirdescriptions many previous reports of spongesymbioses. The following is a description of theassociates of four species of sponge, threesubtidal and one intertidal in Pacific N orthwestmarine waters.

METHODS AND MATERIALS

Because the sponges studied occupied variedhabitats, the methods of collection differed.Mic rociona prolifera and Suberites lata, whichwere subtidal, were obtained with SCUBA. Theywere pulled or scraped off their substrate andimmediately placed into zippered plastic bagswhich were sealed underwater. Halicbondriapanicea, which occurred intertidally upon un-protected rocks, was scraped off the rocks by useof the fingernails and placed into plastic bags.H omaxlnella sp. was collected from 10 to 15fathoms with an otter trawl. The collected speci-mens were subsequently taken to the laboratory

347

348

to recover the associates. The sponges werepulled apart and examined under a binocularmicroscope (120 X ) . The inhab itants foundwere preserved in AFA for further examinationand identification. The results are summarizedin Table 1.

OBSERVATIONS AND EXPERIMENTS

Suberites lata

Snberites lata occurred in Hood Canal, W ash-ington, in a distinct zone of its own at a depthof 5- 10 meters (MLLW ) on nearly verticalrocks that were lightly covered with a fine mud.The water conditi ons in this narrow inlet werequite stable and calm.

The consistency of the sponge was very toughand impenetrable. It harb ored very few orga-nisms. The 15 samples studied had 150 associ-ates, representing 25 species and 6 phyla. Thedensities of the associates varied from 0.001 to0.03 (mea n 0.1) organisms per ern" of sponge(Table 1) . The sparsity of inhabitants was mostlikely due to the toughness of the sponge. Thepred ominant organisms were the small spionidpolychaete Polydora socialis which was foundin every Hood Canal sample, the gammaridA oiroides columbine, the ectoproct Crisia sp. ,and unidentified, filamentous green algae.

P. socialis was the only organism actuallyliving within the sponge, the remainder weresimply attached to the surface. The largeranimals, such as the crabs, were usually foundhiding among the convolutions and contortionsof the sponge.

One clump of S. lata was found in an un-protected rocky coast situation at Yaquin a Head,

PACIFIC SCIENCE, Vol. XXII, July 1968

Oregon. It harbored only a few more organismsthan the Hood Canal samples. The organismsfound on the Yaqu ina Head sample were thoseof the local community .

M icrociona prolifera

M icl'ociona prolij er« was found attached tosubtidal rocks in Willapa Bay, W ashington. Thissponge was erect, branching, and bushy, some-times forming a mass of branches 1 foot or morein diameter and 6 or 7 inches high .

M . pl'olifera provided a place to live for manyanimals and plants; the population densityvaried from 0.27 to 3.64 (mean 0.80) orga-nisms per cm3 of sponge. The organisms mostcommonly found were the gammarid Coropb iumacberusicnm, various caprellids, the polynoidH arm otboe imbricate, the sabell id Sabellamedia, and the anemone Diadumene lncine. The15 samples of sponge dissected harbored 9,581organisms of at least 52 species from 9 phyla.The most common and abundant of this arraywere the amphipods Coropb ium acherusicnmand various caprell ids.

The majority of the inhabitants were foundattached to the surface of the sponge and rela-tively few were found within the sponge tissue.Some of the amphipods, sabellids , nereids, andnematodes occurred within the sponge. It wasapparently difficult for the associates to pene-trate into the tough and fibrous M . prolifera.

Due to the branching and bushy morphologyof the sponge, mud often collected on it andthis mud harbored many of the associatesthat were found . Therefore, there is a questionwhether the associates were attracted to the

TABLE 1

A VERAGE THICKN ESS, V OL U ME, T OTAL A SSOCI ATES, AN D D ENSITY OF P OP ULATION F OR

SAMP LES OF Suberites lata, Microcion« prolijera, AND Halicbondria panicea

AVERAGE AVERAGE AVERAGE

SPECI ES N O. OF T H ICK NESS VOLU M E TOTAL AVERAGE

OF SPON GE SAM PLES (CM) (CM3) ASSOC IATES DENSITY

Suberiteslata 14 4.5(3 -7 ) 503( 195-820 ) 10(1-59) 0.2(0.001-0.8 )

Min'ocionaprolij ern 15 5(2-13 ) 594 ( 24-2,340 ) 640 ( 21- 1,885) 1.2 (0 .27- 3.64 )

Halicbondriapanicea 33 1(0.5-3) 104 ( 20- 540 ) 188(1 8-670) 1.8 (0.126- 19.14 )

Associates of Marin e Sponges-LONG

sponge itself or to the mud collected thereupon.The fact that relatively clean samples of thesponge also harbored many associates indicatedthat there was a definite attraction to it. Otherspecies of sponge occurring near or upon M .prolij era were completely devoid of associates,suggesting that the associates preferred M.prolifera .

H alicbondria panicea

GENERAL OBSERVATIONS : H alicbondria pani-cea occurred as an encrusting sheet 1-5 cm thickupon unp rotected rocky coast areas in violentsurf. It was green in its upper layers and yellowbelow, the coloration being due to an intra-cellular single -cell alga.

H. panicea harb ored many associates of agreat variety. A total of 6,098 organisms of 68species were found in 32 samples with densitiesof population of from 0.26 to 19.0 (mean 1.15)organisms per em", The most common associ-ates were amphipods (Jassa falcata and caprel-lids); the barnacles Balanus gland,tia and B.nubilis; the isopods D ynamen e sbeareri and D.dilata; and the coralline alga Coralline gracilis.

The relationships of these associates to thesponges varied considerably. The associates canbe grouped in four general headings : ( 1) thosethat were caught in the sponge as larvae andsubsequently matured; (2) those that were onthe sponge accidentally at the time of collection;(3) those that grew simultaneously with thesponge on the same substrate ; and (4) thosethat clung to the sponge because it was abundantand provided purchase.

Such organisms as algae, hydroids, ectoprocts,some crabs, barnacles, mussels, and some mol-luscs and annelids probably were caught amongthe sponge spicules as planktonic larvae and,having survived digestion, developed into adultsthereupon and therein . Many of the nereids,large crabs, caprellids , gammarids, and fishwere found by chance upon the particular spongecollected when they were feeding or using itfor temporary shelter . It is most likely thatsome of the algae (e .g., M icrocladia borealis)grew simultaneously with the sponge.

Several animals were grazing on the sponge.Acmaea asmi, M opalia lignosa, Oedognatbnsinetm is, Pachycheles rndis, Cancer productns,and , particularly Arcbidoris montereyensis were

349

found to have H . panicea spicules in theirstomachs.

The only possible examples of truly mutual-istic associations were those concerning theintracellular algae of H. panicea and the clean-ing action of the crustacea associated with H.panicea, as well as with Snb erites lata andM icrociona prolifera. The algae may have pro-vided needed oxygen or other nutrients to thesponge . The sponge without algae could sur-vive, but it usually appeared as thin , rubberysheets. The numerous crustacea fed on thematerial wh ich collected upon the sponge .

Generally, the samples of H . panicea fromdifferent areas harbored the same species, andthe various members of the community wereusually found in the same proportions fromsample to sample. H owever, the exact composi-tion of the sponge communities varied enoughamong the samples to suggest that a g iven in-dividual community was related to the fauna ofthe immediate area.

EXPERIMENTS: Sponges of similar consistencyliving immediately adjacent to Halicbondriapanicea normally had very few associates of anykind , while the clumps of H . panicea were wellprovided with associates. Also, when the nudi -branch A rcbidoris montereyensis was placed intoa laborat ory tank upon a piece of Sub erites latadownstream from H. panicea, the animal wouldinvariably crawl to the H . panicea, indi cating achemotaxis toward the H . panicea.

An apparatus similar to that of Davenport(1 950) was set up to determine if there was achemical attr action of A. monterevensis to H .panicea. The apparatus consisted of a series ofwaterways which gave the nudibranchs a choicebetween water passing over H . panicea and Sub-erites lata. The majority of the nudibranchs al-ways moved toward the H . panicea water outlet :9 individuals of 15, 10 of 15, 13 of 15 in thethree experiments went to the H. panicea outlet.N one ever moved to the S. lata outlet.

H omaxinella sp.

This sponge had much the same shape asM icrociona prolifera, but it was branched fewertimes, and its branches were thicker and hollow.H omaxin ella sp. occurred at a depth of 20-30meters off the Oregon coast.

350

The hollow branches were found to containclusters of 20- 30 fish eggs with living embryosinside. The eggs were probably those of a smallblenn y or goby that normally deposits its eggswithin sponges . All the eggs found died beforehatching.

Assuming that these eggs are from a sponge-dwelling fish, th is example of fish-sponge in-quilinism is the first recorded for H omaxinellasp. and for the Pacific Ocean.

DISCUSSION

The four species of sponge studied showedgreat variation in the community of associatesf ound. These variations were due to the differ-ent habitats of the sponges and also to differentpr operties of the sponges that acted as attrac-tants and inhibitors. Other species of spongeadjacent to those studied harbored fewer associ-ates, probably because they lacked factors thatattracted the associates, or they possessed some-thing that discouraged them.

In comparing H alicbondria panicea, M icro-ciona prolij era, and Suberites lata, which oc-curred in progressively deeper waters, the com-munity of associates was pr ogressively morestable, consistent , and predictable and its com-position fluctuated less from sample to sample.The intertidal H . panicea showed great varia-tion in community density and composition, in-dicating that these parameters were related tothe fauna of the immediate area. When twospecies of sponge (e.g., S. lata and H . panicea)occurred next to each other, the tougher, moreimpenetrable species harbored fewer associates.

The majority of associates did little harm tothe host sponge. The numerous crustacea mayhave benefited the sponges by cleaning them .Such a relationship can be considered mutu alist-ic. The intracellular algae of H . panicea werealso pr obably mutualistic with the sponge. Themost deleterious associate was Arcbidoris mon-tereyensis, which, along with other pred ators,fed voraciously on H . panicea. Some of the tubeworms and amphipods associated with M.prolifera caused minor structural damage to thesponge. The sponge-dwelling fish whose eggswere deposited within H omaxinella sp. obvious-ly benefited from the water currents and shelterof the sponge, while it did not appear to inhibitthe growth or functi on of the host.

PACIFIC SCIENCE, Vol. XXII, July 1968

CONCLUSIONS

Four species of sponge from the coasts ofOregon and W ashington were stud ied forassociates. A large number and great variety ofassociates were found . The great numbers ofsponge associates suggests that sponges are im-portant to the ecology of the numer ous associ-ates. The relationships of these associates fallinto four general categories: (1 ) inquilinism,or lodging, with in or upon the sponges; (2) co-existence of two organisms on the same substrate as a result of simultaneous growth ; (3)predation or grazing ; (4 ) mutu alism.

The 32 samples of Halicbondrin pnniceaharbored 6,098 organisms, representing 68species, with a mean density of 1.15 organismsper em" of sponge. The 15 samples of Micro -ciona prolijer« had 9 ,581 associates with a meandensity of 0.80 organism per ern". The 14 sam-ples of Suberites lata had 150 associates at amean density of 0.1 organism per cm3 ofsponge.

Significant differences in composition anddensity among the sponge communities wereattributed to the different habitats of the speciesstudied, and to differences in the exudates andconsistencies of the species of sponge. The nudi-branch A rcbidoris montereyensis, which fre-quently occurred on Halicbondria panicea, ex-hibited a distinct preference for that species inlaboratory experiments.

The tubular sponge H omaxinella sp. harboredthe eggs of a fish, which was assumed to be anunidentified, sponge-dwelling inquiline.

ACKNOWLEDGMENTS

Space and facilities were provided by Prof.Ivan Pratt at his laboratories in Corvallis and atthe Marine Science Center , Newport. The fol-lowing persons provided identifications of manyof the sponge associates: Dr. Joel Hedgpeth( insects), Dr. J. Laurens Barnard (amphi pods) ,Mr. Danil Hancock (annelids) .

REFERENCES

CAULLERY, M. 1952. Parasitism and Symbiosis.(A. M. Lysaught, trans. ) Sidgwick and Jack-son, London. 340 pp.

DALES, R. PHILLIPS. 195 7. Commensalism. In:

Associates of Marine Sponges-LONG

Joel W . Hedgpeth, ed., Treatise on MarineEcology and Paleoecology, Vol. 1. Mem.Geol. Soc. Am. 67, pp. 391--412.

DAVENPORT, DEMOREST. 1950. Studies in thephysiology of commensalism. I. The polynoidgenus Arctonoe. BioI. Bull. 98(2) :81- 93.

FORBES, MILTON 1. 1964. Distribution of thecommensal oyster, Ostrea perm ollis, and itshost sponge. Bull. Mar. Sci. Gulf and Carib-bean 14(3) :453--464.

- - - 1966. Life cycle of Ostrea permollis andits relationship to the host sponge, StellatagmbH. Bull. Mar. Sci. Gulf and Caribbean16(2) :273- 295.

GUDGER, E. W . 1955. Fishes that live as in-quilines (lodgers) in sponges . Zoolog ica35:121-126.

H ENRY, S. MARK. 1966. Symbiosis. Vol. 1.Academic Press, N ew York. 429 pp.

351

N ICOL, I. A. COLIN. 1964. The Biology ofMarine Animal s. Interscience Publishers, Inc.,New York. 707 pp .

PEARSE, A. S. 1935. Inhabitants of certainsponges at Dry Tortug as. Papers TortugasLab., Carnegie Inst. W ashington 28 ( 7) :117-124.

- --1949. N otes on the inhabitants of certainsponges at Bimini . Ecology 31 (1) :149-1 51.

RADCLIFFE, LEWIS. 1917. Description of a newgoby, Garmannia spongicola, from NorthCarolina. Proc. U.S. N atl, Mus. 52 :423-425.

RIDLEY, S. 0. , and A. DENDY. 1887. Reporton the Monaxonida . In: Report on the voyageof the H .M.S. Challenger (Zoology) 20:1-263.

VOSMAER, G.c.J. 1911. The genus Spirastrella.In: The Porifera of the Siboga-Expedition.Vol. 64, pt. 2, pp . 1-69. E. J. Brill, Leyden.