There was no difference between the 

 relative numbers of different types of 

 pelecypods attacked by drills in tank Z 

 of Experiment I and check Tank 3 of 

 Experiment II. A X^ test on the data 

 supports these results (table 11). 



Differences occurred between the 

 distribution of first attacks on the dif- 

 ferent species of bivalves made by 

 drills in each of the four tanks of 

 Experiment I. This was confirmed by 

 a X^ value of 18.41 which was greater 

 than 16.92 at the 5 percent level of 

 significance, with 9 degrees of freedom 

 (table 12). When the results of tank 2 

 of Experiment I were excluded, a X 

 value of 3.64 was obtained (table 13). 

 This value indicates that the distribu- 

 tion of first attacks among the different 

 species of tanks 1, 3, and 4 of Experi- 

 mient I were not significantly different. 



No difference was found between the 

 distribution of first attacks on the dif- 

 ferent species of bivalves of tank 2, 

 Experiment I, and check tank 3, Ex- 

 periment II. This was confirmed by a 

 X^ test on the data (table 14). 



Tank 2 "continuation test" . --Of 16 

 drills that attacked food animals, 12 



made more than one attack during the 

 "continuation test" (table 3). Eleven 

 of the 12 drills attacked during the 

 second time period. Of these 11 drills, 

 three changed their choice of food from 

 Olympia oysters to Manila clams, and 

 the other eight continued to attack the 

 species they attacked last during Ex- 

 periment I. 



" Non-replacement of food" tests . -- 

 There was no difference between the 

 relative numbers of different types of 

 bivalves attacked by drills in tanks 1 

 and 4 of Experiment II (tables 6 and 8), 

 as confirnnedby X^ test (table 15). Also, 

 no significant difference was found 

 between the distribution of the first 

 drill attacks of the two tanks (table 16). 



The results of tanks 1 and 4 of Ex- 

 periment II were combined (table 17) 

 to show which species of food was pre- 

 ferred. Eighteen individuals of each 

 species of food were available to the 

 18 drills in each tank. As shown in 

 table 17, 33 Olympia oysters were 

 attacked during the first, second, and 

 third attacks by the drills. The bay 

 mussels and Manila clams were not 

 attacked as readily as the Olympia 

 oysters, and Pacific oysters were least 

 favored. 



Ikble 9. — Chi-sguare test of the relative numbers of different types of organisms 

 attacked in the four tanks of Experiment I 



Chi-square value of 33.04, with 9 degrees of freedom. 

 (Significantly different) 



15 



