Stromatoporoids and their symbionts in the upper Silurian of Gotland: Implications for palaeobiological and palaeoecological analysis

Stromatoporoids in Silurian shallow marine limestones from Gotland are commonly symbiotic with syringoporid tabulates, rugose corals, shelly tubes and bioclaustrations (symbionts lacking shell walls). Syringoporid and rugosan symbionts occur almost exclusively in reefs, contrasting deeper facies where stromatoporoids have shelly tubes and bioclaustrations. In a Ludlow biostromal sequence at Kuppen, eastern Gotland, symbionts occur in stromatoporoids in both a marl unit and an overlying biostrome containing the same stromatoporoid taxa and symbionts, but in the marl, stromatoporoid growth forms appear more delicate. A minimum of 10% of a suite of 541 stromatoporoids contain symbionts: 1) two of the 16 stromatoporoid taxa (all Petridiostroma convictum and many “Stromatopora” venukovi specimens) contain abundant intergrown syringoporids; in others syringoporids are rare or absent. Rugosans occur sporadically in several taxa, commonly P. convictum. Some “S” venukovi specimens have syringoporids, rugosans and spiral shelly tubes close together; 2) symbionts mostly began growth after stromatoporoids established, rarely occurring at the base; 3) most stromatoporoid hosts and symbionts evidently had matched growth rates; 4) in some samples, symbionts terminated during stromatoporoid growth, reappearing later within the same host skeleton; and 5) at stromatoporoid growth interruption events, some coral symbionts survived and others did not. These results are interpreted to indicate that: a) symbionts discovered stromatoporoids by chance; b) in shallow locations, reasons for specific taxonomic relationships include suitable substrates for symbionts and a biological relationship; c) stromatoporoid hosts and symbionts responded differently to events; controls on growth were complex; d) there is little evidence that symbionts enhanced reef growth; e) in deeper facies, symbionts were largely overgrown by host stromatoporoids, thus appear unwelcome. In a-d, distinction between commensalism and mutualism remains unproven, but parasitism seems unlikely. The pattern found in this study may apply through stromatoporoid geological history and needs to be tested on other assemblages.