KREKORIAN, SOMMERVILLE, and FORD: LOBSTER BEHAVIORAL INTERACTIONS 



two are very similar, at least where species other 

 than primates are involved. For example, a recent 

 study by Myrberg ( 1972b), involving such a direct 

 comparison, showed that the agonistic and other 

 social behavior of the bicolor damselfish, Eupo- 

 macentrus partitas, in the laboratory was very 

 similar to its field behavior both qualitatively 

 and quantitatively. In another study, Hazlett 

 and Bossert (1965) similarly detected no quali- 

 tative differences between the laboratory and field 

 behavior of some pagurid and diogenid crabs. 

 As a result of this, they chose to study the behavior 

 of these crabs in the laboratory where conditions 

 were more uniform and controllable. 



There have been surprisingly few studies con- 

 cerned with interspecific aggression. This is 

 especially true of many decapod crustaceans for 

 which most of the studies have been rather 

 superficial (Reese, 1964). It is rapidly becoming 

 apparent from the few studies that have been 

 made, however, that interspecific aggression may 

 be very important in regulating the distribution 

 and abundance of many marine animal popula- 

 tions. Myrberg (1972a) has found that in the bi- 

 color damselfish interspecific chases accounted for 

 approximately 40% or more of the chases carried 

 out by territorial males. Interspecific aggression 

 was displayed by both sexes and occurred through- 

 out the year. 



In laboratory experiments. Teal (1958) found 

 that the fiddler crab, Uca pugnax, reduced the 

 number of burrows dug by U. pugilator and 

 U. minax. In field studies, described by Aspey 

 (1971), it was found that where U. pugnax and 

 U. pugilator existed in overlapping areas, the 

 number of burrows per square meter was signifi- 

 cantly less than in areas inhabited only by U. 

 pugilator. The reduction in the number of burrows 

 dug by other species of Uca when paired with 

 U. pugnax appears to be due to the greater 

 frequency of agonistic display exhibited by U. 

 pugnax (Aspey, 1971). 



In contrast to the results of the Panulirus- 

 Homarus experiments, very little behavioral 

 interaction or aggression occurred between C. 

 antennarius and P. interruptus. Seventy-three 

 percent of actor behavior involved nonaggressive 

 actions (social contact), and 84% of the reactor 

 responses involved no change in behavior (no 

 response). In addition, C antennarius frequently 

 remained in close proximity to P. interruptus, 

 and P. interruptus frequently walked over C 

 antennarius when moving from one area in the 



tank to another area. These events were never 

 observed inHomarus-Panulirus experiments. The 

 laboratory observations on Panulirus-Cancer 

 interactions agree with observations made in the 

 field where C. antennarius is sometimes found 

 sharing the same refuges with P. interruptus. 



Our data also show that even though the 

 number of Homarus-Panulirus interactions 

 decreased with time, there was still a large 

 number of behavioral interactions between P. 

 interruptus and H. americanus on day 10 in the 

 no shelter condition. This is most clearly seen 

 when the number of interactions on day 10 are 

 compared with the total number of Cancer- 

 Panulirus interactions. The percentage of aggres- 

 sion shown by Homarus {Homarus threat + 

 Homarus attack) on day 10 (40% ) was very similar 

 to the total percentage of aggression (39%). 

 Likewise, the amount of fleeing (Panulirus walk 

 away -I- Panulirus abdomen flex) shown by 

 Panulirus on day 10 (55% ) was similar to the 

 total percentage of fleeing (63%). Thus, although 

 the absolute number of behavioral interactions 

 decreased with time, the relative amounts of 

 Homarus aggression and Panulirus fleeing 

 remained the same as the total percentages. 

 These data suggest that, even if the number of 

 encounters in the field between introduced H. 

 americanus and P. interruptus were small, the 

 behavioral actions by H. americanus would be 

 largely aggressive and the responses by P. 

 interruptus defensive. 



Locomotion 



Homarus vs. Panulirus with No Shelter 



The total numbers of P. interruptus observed 

 roaming during the three observation periods, 

 precontrol 1,277, experimental 1,271, and post- 

 control 1,171, were not significantly different 

 (Table 4; P < 0.05, Kruskal-Wallis One-Way 

 Analysis of Variance by Ranks). A comparison of 

 the three observation periods showed that the ma- 

 jority of this roaming by both P. interruptus and 

 H. americanus occurred during the 1510-1700 h 

 observation period. In P. interruptus, 62% of the 

 roaming occurred during this period. In//, ameri- 

 canus, 81% of the roaming occurred during the 

 1510-1700 h observation period. 



The relationship between the amount of time 

 spent roaming hyH. americanus and the number 

 of i/omarus-initiated behavioral actions is shown 



1155 



