A 470 m long ghost net, first observed in June, 1984 was 

 resurveyed for two consecutive years with Johnson Sea-Link and 

 Delta . The vertical net profile during the first year of study 

 varied from < 0.2 to 1.0 m and in the succeeding years from < 0.2 

 to 0.7 m. Dogfish was the predominant vertebrate species caught in 

 the net; cancer crabs represented the most common invertebrate 

 catch (Table 1). Oceanpout, Macrozoarces americanus , cunner, 

 Tautogolabrus adspersus, and redfish, Sebastes marinus , were 

 frequently seen near the net, but none were observed caught in the 

 net. 



Codfish were not seen in the ghost gillnet, nor were there 

 identifiable remains of cod seen at the base of the net. Fishermen 

 in the area notified us each year on the timing of the decrease in 

 cod and increase in dogfish several weeks prior to our survey. Cod 

 were not present during any of the surveys. However, if the ghost 

 nets observed had caught cod within a month prior to 

 our survey, at a time when cod were known to be in our study area, 

 we would expect to find skeletal remains within or at the base of 

 the net. We observed on only one occasion the skeletal remains of 

 a fish belonging to the gadid family. Since cod are an 

 economically important species associated with the New England area 

 and one of the key species targeted by commercial gillnet 

 fishermen, this observation is important. Impact of these ghost 

 gillnets on juvenile groundfish, other than flatfish, should be 

 minimal because of the size of the mesh ( = 15 cm) and its 

 selectivity (Stewart, 1987; Kawamura, 1972). 



Both the 470 m and 700 m ghost nets were stretched out in a 

 horizontal mode. Ghost nets found in this configuration appear to 

 have greater impact on the fishery resources than those nets that 

 are twisted into a more vertical, "balled-up" configuration. 

 During the first year of the study fishermen identified ten ghost 

 gillnets which we examined; four were twisted into a vertical tight 

 bundle, four were stretched out horizontally, and two nets (Fig. 

 4) had a combination of vertically-aligned twists and horizontal 

 lengths. Within these ten nets one dogfish was observed caught in 

 a vertical bundle: a greater number of species and individuals of 

 a given species were found in the stretched horizontal sections. 



Net-animal-substrate interactions are portrayed in Figures 3 

 and 4, which present typical ghost gillnet configurations and 

 related animal associations (entangled, taking shelter, and 

 attached to) as documented on color video. A higher than normal 

 concentration of animals are illustrated to portray various aspects 

 of the ghost gillnet problem. Figure 3 illustrates distinct net 

 profiles, stretched horizontal, wrapped around boulders and rocks, 

 and vertical twists. The horizontal stretched sections of net had 

 a vertical profile of one meter or less. Vertical thrusts in these 

 nets rose to a profile of 2.5 to 3.0 meters. Dogfish were the most 

 frequent vertebrate species caught; they usually became wedged or 



437 



