In all cases, the Pacific oysters were 

 the least preferred. 



Tanks which differed in the numbers 

 of various species attacked, also dif- 

 fered in the distribution of first attacks. 



In Experiment I there were 103 pos- 

 sible chances (total possible pairs 

 of attacks, tables 2 through 5) for a 

 drill to attack the same species that 

 it had attacked previously. Of these, 

 84 pairs were attacks of the same 

 species. This, as well as the tank 2 

 test, suggested that a drill might con- 

 tinue to attack the same species of 

 food organism that it had attacked 

 previously, rather than move to an 

 organism of a different species. If 

 this hypothesis is accepted, it might 

 be expected that the number of attacks 

 would be greatest upon Pacific oysters, 

 since the drills were taken from this 

 species of MoUusk originally. This 

 was not the case as noted above. It 

 should be mentioned that when the 

 drills were collected off of Crassostrea 

 gigas, nnany drills were depositing egg 

 cases. Thus, it does not necessarily 

 nnean the drills were attacking Pacific 

 oysters when collected. According to 

 Chew and Eisler (1958), someOcinebra 

 deposited egg cases in the early days 

 of an experiment. These egg cases 

 were cemented on the shells of several 

 Pacific oysters and on the bottoms and 

 sides of two experimental aquaria. They 

 noted that no egg cases were found on any 

 other test animal including Olympia 

 oysters, bay mussels, and Manila 

 clams, indicating that the Japanese drill 

 may be able to differentiate between 

 species of bivalves. Possibly this could 

 have occurred in the field, when the 

 drills were collected for this study. 



Positive correlations existed between 

 the thickness of prey's shells and the 

 duration of attack on bay mussels, 

 Olynnpia oysters, and Pacific oysters. 

 No correlation was found between dura- 

 tion of attack and thickness of clann 

 shells. The narrow range in shell thick- 

 ness might have accounted forthislack 

 of correlation. Ocinebra took the least 

 number of days to drill and eat the 

 bay mussels (4-5 days), followed in 

 increasing order by the Pacific oysters 

 (5-6 days), Olympia oysters (6-7 days), 

 and Manila clams (7-8 days). The 



mean shell thickness of the four food 

 species used in Experinnent I fell in 

 this same order (bay mussels, 0.0220 

 inch; Pacific oysters, 0.0316 inch; 

 Olynnpia oysters, 0.0418 inch; and Ma- 

 nila clanris, 0.0548 inch). Therefore, it is 

 probable that the variations induration 

 of attacks were the result of differences 

 in thickness of bivalve shells. 



When daily observations were made, 

 one or two drills were usually found 

 on the walls of the tanks. Some as- 

 cended the wall to the surface of the 

 water. When they descended to the edge 

 of the floor of the tank, they were not 

 confronted with an equal choice of the 

 four species of food. However, a drill 

 usually moved down the wall and tem- 

 porarily on to an animal at the peri- 

 phery of the group of food species, 

 then moved within the group before 

 making an attack. It is believed that 

 the error due to this factor is not 

 great enough to affect the results of 

 these experiments. 



An Ocinebra frequently drilled half- 

 way through the valve of a victim and 

 then stopped to move to another prey. 

 Another drill sometimes moved onto 

 the first prey and continued drilling 

 on the vacated hole. The second drill 

 usually finished the hole started by the 

 first. For example, as noted in the 

 daily record, drill number 12 (tank 3, 

 Experiment I) attacked and drilled 

 halfway through the shell of a clam 

 (location L5) from August 1 3 to 17. 

 After this drill moved off the bivalve 

 (August 17), another drill moved onto 

 this sanne clam on August 20 and con- 

 tinued drilling in the same hole. 



It was not uncommon to observe two 

 or three drills attacking the sanne 

 victim, but only two cases (both Manila 

 clams) were observed where two com- 

 plete perforations were found on the 

 same food animal. In other cases, one 

 completely perforated hole and one or 

 two half-drilled holes were found. 

 Federighi (1931) and Galtsoff et al. 

 (1937) also have reported that two or 

 more Urosalpinx citierea may attack 

 an oyster simultaneously. 



No study has been made on the feed- 

 ing processes of Ocinebra; however. 



24 



