|
<br />~ ~
<br /> 0
<br /> I
<br /> >- 25
<br /> u
<br /> z
<br /> w 20
<br /> =>
<br /> 0
<br /> w 15
<br /> a::
<br /> u.
<br /> w 10
<br /> >
<br /> I- 5
<br /> <[
<br /> -1
<br /> W 0
<br /> a::
<br /> 1 3 5 7 9 11 13
<br /> COMPARTMENT NUMBER
<br />
<br />
<br />FIG. I.-Relative frequency distribution of Colorado squawfish in the lest trough Willi II!',.
<br />~em,p~rature equal lO respenive acclimar.ion temperature. Data on 60 observations pn I:
<br />IOdl\'ldual fish combined.
<br />
<br />hours after placing the fish in a thermal gradient, and is strongl~' in[h...".
<br />by prevIOUS thermal acclimation. However, when given sufficiem 1111"
<br />the gradient (usually 24 hr), the fish will gradually move to a region 01 r
<br />ferred temperature that is considered to be species-specific and nOI 1111
<br />enced by previous thermal acclimation (Reynolds and Casterlin, 19i1J ,
<br />(194 i) defined the latter to be the "final preferendum." He further delll'
<br />final preferendum as "that temperature at which the preferred temper~llll'
<br />equal to the acclimation temperature." If acute preferenda are pilll"
<br />against acclimation temperature, and a curve visually fit through Ii,'
<br />points, the intercept of that curve and the "line of equality" cOrreS~)OIH:-
<br />the final preferendum (Richards et aI., 1977; Jobling, 1981).
<br />
<br />METHoDs.-Yearling Colorado squawfish used in this slUdy were hatched in .JulY 19;" ,.
<br />hatchery-reared adults at the Willow Beach (Arizona) National Fish ){atchery. In Juh 19'" ,.
<br />were transferred to the Utah Water Research Laboratory. Logan, fed dry commercial tTOU' "n' ,
<br />libitum twice per day, and kept on a 14L x IOD photoperiod. The fish ranged in leng,h:
<br />[rom 7.5 to 10 cm,
<br />
<br />Fish were held al an initial temperature o[ 20 C and were brought to their acclimation t.,,"~'
<br />lUres by a change o[ I C per day (Reynolds. 1977). Fish were held one week afrer acclimallo!' '
<br />higher temperature and one month for a lower acclimation temperature before use in prri",~
<br />experiments (Richard et a1.. 1977). All fish were kept in flow-through systems o[ healfd, drrh
<br />nated city waler. Dissolved oxygen was 90-100% salUration, ammonia below 0.2 ppm. and "I' .
<br />8.0.
<br />
<br />Tests o[ acute preferenda were conducted in a horizontal gradient trough which measurrd; r
<br />25 cm x 5-10 cm depth. Range in temperature was controlled by adjusting the flows 01 hol .,.
<br />cold waler at opposite ends o[ the trough. Flows were I liter per minute or less. Water enrrr~ '"
<br />trough through perforated lUbing secured along the bottom. Seven air stones were auached ar .-
<br />form intervals along the bottom of the trough '0 prevent gas supersaturation and vertical "..".
<br />cation. Thirteen thermistor probes (YSI teleth, '. mometer) were similarly attached at uniform ,nr-
<br />vals, with the telethermometer on an observation plat[orm. Fish were observed through a ..".
<br />opening in a curtain which surrounded the gradient trough, Uniform, low-level lighting (appl'"
<br />
<br />--'
<br />
<br />, I lux at the water surface) was provided, and all fish were tested within the daylight
<br />~. The range o[ temperatures in the gradient was varied according to the acclimation temper-
<br />"'~oI .he [ish: 13-330C [or 140C fish, 16-350C [or 200C fish, and 14-340C for 260C fish, Twenty
<br />..,. r Jested per acclimation temperalure.
<br />... ;'dual [ish were not fed on the day of lesting. A single fish ,:"as placed in the gradient
<br />h al a poinl where the temperature most closelv corresponded to 1Is acclimatIon tempera lure
<br />--'h tI al.. 1979; Hall et al.. 1979). Observations o[ fish position and temperature al that posi-
<br />:.,tor .....r taken over a 3-hr period, at one-minute intervals. [rom the moment the fish was placed
<br />........ Irough. Only the last hour was used to estimate lemperature preference, but the first two
<br />.. "",tributed information on the tempera lUre selection behavior of individual fish.
<br />~ Ihe lest ~riod, the gradiem was shifted by shulling of[ either the hot or lhe cold water.
<br />nr lith "'3S then observed for 30 minutes at one-mimlle imervals to see if it followed its selected
<br />..-prraIUrt' down the trough or i[ il remained in a particular preferred area o[ the trough
<br />,a.unl' and Ibara, 1978; Hesthagen, 1979). An additional control was used because it was some-
<br />..... d,fhrult to shift Ihe I!"radient significantly. Five [ish from each acclimation temperature and
<br />. crouP were individually tesled in the trough wilh the water held at a uniform temperature
<br />,.ulltllht"ir respecU\'f acclimatIon tempera lure.
<br />,-"",dual modes, a mean o[ individual modes, a pooled mean and mode, and Pearson's coeHi-
<br />_ 01 .kewness were calculated at each acclimation tempera lUre, The pooled mode is often
<br />...-rd a. the statistic that best describes the preferred temperature (Reynolds and Caslerlin, 1976;
<br />......,.,1. tt al.. 1977; HeSlhagen. 1979). However, lhe mean o[ the individual modes is less sensi-
<br />_ If' .krwnes, which can be caused by one of twO fish remaining in lhe same spot (particularly
<br />_.,,1 onr end o[ lhe trough). lIse o[ the mean of the modes also gives each fish equal weight in
<br />.:a<.1 rotTlputation. Final preferendum was estimated by plouing acute preferenda against acclima-
<br />".,.. ,.",perature and visually filling a curve through the points.
<br />fJ1Nllr, oi arrhmalion temperature on selected temperature were tested by analysis of variance. H
<br />.......n".' were significant, means were further compared with Fisher's (protected) leasl signifi-
<br />...... d,ll..-enee (LSD) test (Steel and Torrie, 1960).
<br />
<br />Rf_~t'LTS AND DISCUSSION.-Some fish appeared to select a particular posi-
<br />t.on In the trough rather than the temperature associated with that position.
<br />~I\h position data from tests and controls were used to effectively screen
<br />IIxhvidual fish for positional effects. Figure 1 represents the results obtained
<br />wtM-n fish were placed in the trough at a uniform temperature equal .to their
<br />'"1~live acclimation temperature. Data from five fish per acclimation
<br />It'Tt1peralure (140, 200, 260C) were combined. The theoretical frequency of
<br />l11unbulion for random behavior over the fourteen compartments was 7.14%
<br />pet compartment. The distributions were not random but incorporated some
<br />nld.ol.umk bias" (Badenhuizen, 1967). There was an obvious preference for
<br />Ibr rllll compartments.
<br />fi,h position was plotted against time for each individual fish. Individual
<br />inti placed in the trough at a uniform temperature (controls) exhibited ran-
<br />dom movement and/or a preference for either end. There was some variabil-
<br />M' an the behavior of fish placed in a temperature gradient. A few fish exhi-
<br />"It'd behavior similar to that of the controls; it was assumed that they were
<br />~K'lecting temperature. Other fish clearly showed a preference for a par-
<br />brutar area within the trough other than the end compartments. Data of
<br />.arvidual fish were rejected for analysis if position modes occurred at either
<br />flail of the trough. if fish position did not shift with shifts in temperature, or
<br />III lbe variance of fish position was not significantly different from the var-
<br />I&Qc:n of the controls. Th majority of the test fish appeared to select
<br />~ptrature over position in the gradient trough. Application of criteria for
<br />~n of data rejected only three of sixty individuals.
<br />
<br />111........,..,....
<br />~;
<br />
|