|
KARP AND TYUS-COLORADO SQUAWFISH INTERACTION 29
<br />by two large, dominant individuals which dis-
<br />placed others from the central and lower por-
<br />tions of each tank, respectively.
<br />Ptychocheilus Lucius, R. balteatus, N. lutrensis,
<br />and L. cyanellus fed predominantly in the upper
<br />half of the tank, whereas Pimephales promelas,
<br />A. melas, and C. latipinnis fed in the lower por-
<br />tion.Pimephales promelas, N. lutrensis, and R, bal-
<br />teatus exhibited a frenzied feeding behavior im-
<br />mediately after food addition, the latter two
<br />species darting repeatedly between the surface
<br />and lower area of the tank. In contrast, the
<br />feeding behavior of Ptychocheilus Lucius appeared
<br />to be more directed and less frenzied. These
<br />fish initially exhibited brief bouts of intense ac-
<br />tivity following food addition, but quickly re-
<br />sumed roaming movements in al] tanks. All
<br />species exhibited some changes in spatial dis-
<br />tribution with food addition, but few were sig-
<br />nificantly different from nonfeeding distribu-
<br />tions (P. Lucius [C. latipinnis tank] X~ = 8.798, df
<br />= 2, P < 0.05; P. Lucius [N. lutrensis tank] x2 =
<br />14.39, df = 2, P < 0.01; P. Lucius [A. melas tank]
<br />x2 = 15.115, df = 2, P < 0.01; C. latipinnis, X2
<br />= 17.671, df = 2, P < 0.01; R. balteatus, x2 =
<br />14.99, df = 2, P < 0.01; A. melas, X2 = 17.95,
<br />df = 2, P < 0.01). These differences were due
<br />to an upward shift by P. Lucius and R. balteatus
<br />and downward shift by G. latipinzis and A. melas
<br />during feedings. Pimeplales promelas, N. lutren-
<br />sis, and L. cyanellus did not significantly alter
<br />their overall distribution with food addition.
<br />Aggression and activity.-We observed only two
<br />incidents of obvious physical injury due to in-
<br />terspecific interactions. These included a L. cy-
<br />anellus nipping a L. cyanellus and a P. promelas
<br />nipping a Ptychocheilus Lucius. Chases and threats
<br />were the predominant forms of aggression ob-
<br />served. Lepomiscyanellus, Pimephales promelas, and
<br />N. lutrensis initiated the greatest number of ag-
<br />gressiveacts, more than three times the number
<br />initiated by Ptychocheilus Lucius (Table 1). These
<br />non-native fishes were frequently observed
<br />chasing smaller P. Lucius or conspecifics across
<br />several cells. This form of aggression was most
<br />often noted in the L. cyanellus tank. Obvious
<br />territorial behavior was observed only for L.
<br />cyanellus and Pimephales promelas. We attributed
<br />some of the P. promelas aggression to repro-
<br />ductive behavior because all fish developed sex-
<br />ual characters (breeding coloration in all, and
<br />tubercles in the two largest individuals), and
<br />larvae were found during the initial acclimation
<br />period.
<br />Richardsonius balteatus, C. latipinnis and Ptycho-
<br />cheilus Lucius initiated relatively few agonistic acts
<br />and generally roamed with little apparent re-
<br />gard for neighboring fish. The number of ag-
<br />onistic behaviors initiated by P. Lucius was rel-
<br />atively low in all tanks (x = 14.25, range 4-24;
<br />Table 1) and significantly decreased in the tank
<br />containing fathead minnows (X2 = 6.857, df =
<br />1, P < 0.05). Only N. lutrensis and Pimephales
<br />promelas showed a significant increase in ag-
<br />gressivenesswith feeding (P. promelas: X2 = 24.5,
<br />df = 1, P < 0.01; N. lutrensis: X2 = 7.52, df =
<br />1, P < 0.01). Although overall aggression by L.
<br />cyanellus did not increase significantly with feed-
<br />ing (X2 = 2.81, df = 1, P > 0.05), the number
<br />of aggressive acts directed at Ptychocheilus Lucius
<br />during feedings was significant (X2 = 4.03, df =
<br />1, P < 0.05).
<br />All non-native fishes exhibited relatively high
<br />amounts of interspecific aggression; 80% of R.
<br />balteatus, 63% of L. cyanellus, 53% of A. melas,
<br />50% of Pimephales promelas, and 42% of N. lu-
<br />trensis agonistic behaviors were directed at
<br />Ptychocheilus Lucius. In contrast, less than 10%
<br />(all tanks averaged) of P. Lucius agonistic behav-
<br />iors were interspecifically directed. We ob-
<br />served ahigh number of "nudges" (bumping)
<br />between adjacent A. melas and attributed these
<br />to their "shelter seeking" behavior and clumped
<br />distribution. This behavior, directed mostly at
<br />conspecifics (91 %) was not considered a direct-
<br />ed act of aggression. We noted no initiation of
<br />interspecific aggression by C. latipinnis.
<br />Ptychocheilus Lucius and R. balteatus were the
<br />most active and widely ranging species, and L.
<br />cyanellus and A. melas the least (Table 2). Ptycho-
<br />cheilus luciusmovements were similar within and
<br />between all tanks, except for the tank contain-
<br />ing Pimephales promelas. In that tank, Ptychochei-
<br />lus Lucius movements appeared to be somewhat
<br />restricted relative to those in the tank contain-
<br />ing C. latipinnis (activity: Student's t = 2.41, df
<br />= 64, P < 0.05; range: Student's t = 2.09, df
<br />= 64, P < 0.05).
<br />We detected no significant differences in ac-
<br />tivity level between P. Lucius and N. lutrensis, R.
<br />balteatus, Pimephales promelas and G. latipinnis
<br />(Student's t < 1.65, df = 64-67, P > 0.05), but
<br />Pimephales promelas and N. lutrensis used less
<br />space than Ptychocheilus Lucius (Pimephales pro-
<br />melas: Student's t = 2.69, df = 64, P < 0.01; N.
<br />lutrensis: Student's t = 2.29, df = 64, P < 0.05).
<br />Activity level and range of movement of Ptycho-
<br />cheilus Lucius in the tank containing C. latipinnis
<br />was most similar to that in the R. balteatus tank.
<br />
|