FISHERY BULLETIN: VOL 72, NO. 4 



quickly threatened or attacked. However, this 

 event was rare. 



When H. americanus pinched P. interruptus, 

 in both the shelter and no shelter condition, 

 the pincer claw appeared to be the only one 

 used. Panulirus interruptus that had been 

 attacked, and especially those pinched by H. 

 americanus, appeared to move away from an 

 approaching Hornarus in subsequent encounters 

 at greater distances than they did initially before 

 they had been attacked. 



Panulirus vs. Cancer with No Shelter 



There was little aggression shown in behavioral 

 interactions between P. interruptus andC. anten- 

 narius, species which commonly occur together 

 in nature. A total of 229 behavioral actions were 

 initiated by actors. Of these, 73% (167) were 

 classed as social contact, 5% (12) as Cancer 

 approach, 16% (36) as Panulirus approach, 2% (4) 

 as Cancer threat, 1% (3) as Panulirus threat and 

 3% (7) as Cancer attack (Table 2). There was a 

 total of 231 responses by the reactors. Eighty- 

 four percent (194) of these were classed as no 

 response, 9% (21) were classed as Panulirus walk 

 away, 5% (11) as Cancer threat, and 2% (5) 

 as Cancer attack. 



One of us (Krekorian) observed P. interruptus 

 walk over C. antennarius 21 times with no 

 response from the latter species. This was never 

 observed between P. interruptus and H. ameri- 

 canus. In addition, on 63 occasions noninteracting 

 groups were observed composed of two or more 

 P. interruptus and C antennarius within 30 cm of 

 one another, and on 50 other occasions similar 

 groups were observed consisting of C. antennarius 

 and one P. interruptus. In contrast, we never 

 observed groups composed of//, americanus and 

 P. interruptus. 



Discussion of Agonistic Behavior 



These results show that a large percentage of 

 //omaras -initiated behavioral interactions in the 

 shelter (44%) and no shelter (39%) conditions 

 involved aggressive behavior (Homarus threat 

 + Homarus attack. Table 2). The response of 

 P. interruptus to Homarus approach and aggres- 

 sive acts was usually defensive. In the shelter 

 condition P. interruptus was displaced by //. 

 americanus 61% of the time. Panulirus either 

 walked away or used an abdomen flex to remove 



itself from the area occupied by Homarus. In 

 the no shelter condition P. interruptus was dis- 

 placed 63% of the time. Although shelter reduced 

 the total number of behavioral interactions 

 initiated by//, americanus, the percentages for 

 aggressive and defensive acts remained about the 

 same (Table 2). 



The number of Pa/i«/irus-initiated behavioral 

 interactions was far less than the number of 

 behavioral interactions initiated by Homarus. 

 However, the result of the interactions which 

 followed were qualitatively the same. That is, P. 

 interruptus was ultimately displaced by threaten- 

 ing and attacking //. americanus in the shelter 

 (92% of the time) and no shelter (76% of the 

 time) conditions. 



Grasping of one//, americanus by another was 

 rarely involved in the agonistic encounters ob- 

 served by Schrivener (1971). In contrast to this, 

 we frequently observed H. americanus pinching 

 P. interruptus. Pinching by //. americanus is 

 typical of its predation behavior. At no time in 

 our study did we observe P. ifiterruptus attack 

 H. americanus or C. antennarius. 



Studies by other members of our group have 

 shown that the laboratory activity rhythms of P. 

 interruptus and //. americanus over a 24-h 

 period are very similar, both species exhibiting 

 their highest levels of activity during the first 

 4 h of darkness (Van 01st and Carlberg, pers. 

 commun.). This similarity in the timing of activ- 

 ity probably would intensify interactions between 

 the two species in the field. Such interactions 

 could result in P. interruptus being displaced 

 from areas occupied or frequented by //. ameri- 

 canus, and/or being placed under considerable 

 stress due to possible competition for food, suit- 

 able refuges, or other aspects of space within 

 the habitat. In the laboratory, //. americanus 

 displaced P. interruptus when there was only one 

 shelter for two lobsters (Lester, pers. commun.) 

 and inhibited P. interruptus from feeding when 

 food was limited (Needham, pers. commun.). 



Assuming that the behavior displayed by the 

 two lobster species in the laboratory would be 

 similar to that occurring in the field, one would 

 expect considerable agonistic interaction between 

 the two species if they were to occupy the same 

 habitat. Although there have been few studies 

 that have thoroughly compared the behavior of 

 animals in the laboratory with their behavior in 

 the field, there are some data that suggest the 



1154 



