contraction behavior. Results are purely qualitative, since the 

 laboratory conditions were not optimal, but the sea pens were 

 observed to retract completely into the sediment and reappear on 

 a random or, at least, asynchronous rhythm with regard to other 

 individuals in the tank. The animals survived several months in 

 the laboratory and were observed to release gametes in August and 

 September. After this, the condition of the animals deteriorated 

 markedly. Attempts were made to feed the sea pens both zooplank- 

 ton and algal cultures but no feeding response was observed. 



DISCUSSION 



Two species of sea pen of the genus Pennatula (i.e., P. 

 aculeata and P. grandis = P. borealis ) are reported to occur 

 along the north Atlantic coast from the Davis Straits to 

 Chesapeake Bay (Deichmann, 1936). They occur in deeper waters, 

 usually greater than 60 fathoms (Miner, 1950). Although rela- 

 tively common in the Gulf of Maine, the details of their distri- 

 bution have not previously been examined (Figure 1 and 2). In 

 fact, the magnitude of their abundance was not fully appreciated 

 until the submersible based macrobenthic surveys began in 1983. 

 Pennatula is reasonably ubiquitous in the Gulf, with the excep- 

 tion of Wilkinson Basin. Submersible dives in this basin, with 

 the specific objective of searching for sea pens, found no 

 animals. The one previous submersible study in Wilkinson Basin 

 ( Rowe et al . , 1975) also did not include Pennatula in the faunal 

 list. The apparent absence of this species from Wilkinson Basin 

 raises a fundamental question about environmental conditions in 

 the basins of eastern (Jordan and Georges) and western (Wilkin- 

 son) Gulf of Maine. The pattern of water circulation, originally 

 described by Bigelow in 1927, has recently received much atten- 

 tion (e.g., Hopkins and Garfield, 1979; Brooks, 1985, 1987, Ramp 

 et al., 1985) and this may well be the ecologically controlling 

 factor. However, the link between the physical environment and 

 faunal characteristics in the basins has yet to be examined. 



Accurate estimates of sea pen density are confounded by the 

 behavior of this group of animals. These otherwise sessile 

 invertebrates are capable of retracting completely into the 

 sediment thus avoiding predators as well as visual or photo- 

 graphic counts. Contraction and extension behavior is well 

 documented for shallow water species (Birkeland, 1974; Buisson, 

 1964, 1980; Dickson, 1978; Hoare and Wilson, 1977; Imafuku, 1976, 

 1980; Magnus, 1966; Mori 1944a, & b, 1945, 1947a & b, 1949, 1950, 

 1960; Pavans de Cecatty and Buisson, 1965; Pavans de Cecatty et 

 al . , 1963) but no data are available on deeper water forms like 

 Pennatula . Results reported here confirm that deeper water 

 animals show a similar behavior. Further study is needed to 

 understand when and why the animals contract before we can 

 accurately estimate density. In shallow water, pennatulids 

 appear to expand to feed on a diel cycle in response to light or 

 increases in zooplankton. In deep water, light is greatly 

 reduced or totally absent so this cannot serve as a cue for 

 extension and contraction. The population observed over one 



127 



