Response profile of the Calafia pearl oyster, Pinctada mazatlanica (Hanley, 1856), to various sedative therapies related to surgery for round pearl induction.

ABSTRACT Anesthetic drugs are common in management sequences of commercial aquaculture that imply relaxation of specimens. Studies on physiologic response and relationships with functions of body parts are particularly centered in fishes to improve transport, handling, or surgery (i.e., spawning stimulation, sampling of blood and fluids). In pearl oysters, relaxation therapies are important to induce round pearl formation with nucleus-mantle insertions, but the physiologic effect of drugs are poorly known. In the current work, respiration rate (Rr, in [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1]) was used as index to establish recovery profiles in Pinctada mazatlanica and evaluate relaxation therapies of three doses and two exposure times using propylene phenoxetol (PP), MS-222 (MS), and quinaldine (QN), in comparison with an untreated (CTRL) model. Recovery profiles were explored with median Box-Whiskers pints. Intercept position, dispersion pattern, and range values of Rr, and slope attributes, were inferred through correlation and regression methods. Relationships among these elements were analyzed with three-way ANOVA (time x dose x exposure time); recovery profiles were compared within and between drugs through multiple covariance and cluster analysis to explain physiologic responses. Two-way ANOVA at start and end of each trial provided further elements for therapy evaluation. Specimens responded dynamically in this experiment, sharing an overall background variation of Rr strongly influenced by time but with wide differences of significance P level, between [10.sup.-18] in CTRL and [10.sup.-10] to <[10.sup.-24] between drugs. CTRL, MS, and PP had negative correlation of Rr over time, whereas it was positive for QN. The influence of dose had grouped P levels between [10.sup.-11] to [10-.sup.-15]; exposure time and interactions with dose shared P levels of similar significance between [10.sup.-3] to [10.sup.-5]. In comparison with the CTRL model, the use of drugs apparently contributed to decreased stress response in treated specimens, and this was particularly notable in early recovery phases. Variations in dose and exposure time defined specific recovery profiles within drugs. An adequate presurgery therapy should sustain low intercept, low dispersion and range value of Rr, and horizontal slope. QN at doses of 45 mL and PP at 5 mL propiciated low-activity profile during recovery.

KEY WORDS: MS-222, Pinctada mazatlanica, propylene phenoxetol, quinaldine, relaxation therapy

INTRODUCTION

Surgery techniques for round pearl formation in pearl oysters have been examined since the early 1940s in publications, reports, manuals, congresses, reunions, and training courses (Gervis 1992, Fassler 1994, SPC-POIB). Videotapes, photography, and the World Wide Web have also become important sources of information. The general agreement recognizes that a pearl culture technician may qualify as "trained" after thousands of essays; even then, differences of success rate (nucleus retention, survival, number and quality of pearls) are notable among technicians and species (Alagarswami 1987, Shirai 1994, Monteforte 1996, Haws 1998, Haws et al. 1999, Nava et al. 2000). Many inherent factors have their own influence, such as the biologic stale and health of pearl oysters, response profile to presurgical and healing therapies, quality and profilaxis of workspace, and so forth. Field culture management and environmental conditions also play an important role. Pearl culture technicians coincide in two major requierements that specimens should fulfill before the surgery: (1) newly spawned gonads, empty but prominent enough for assuring nucleus and mantle piece insertion, and (2) adequate relaxation, valves can be easily gapped and body contraction will be slow during the intervention. Thermal shocks and other strategies are applied for spawning induction; anesthetic compounds are commonly used as relaxation treatments (Alagarswami 1987, Ehteshami 1993, Monteforte et al. 1994, Mills et al. 1997, Norton et al. 2000). We investigated the recovery profiles of the Calafia pearl oyster, Pinctada mazatlanica (Hanley, 1856), treated with propylene phenoxetol, MS-222, and quinaldine in different doses and exposure times. P. mazatlanica needs adequate presurgery therapies for round pearl stimulation because specimens react defensively if certain level of relaxation is not reached. On the other hand, norms for evaluation of response profiles in mollusks under anesthesics, including pearl oysters (Norton et al. 2000), have been related mainly with behavior (reply level of body parts as a result of mechanical stimulations, time for specimens to attain visible signs of "normal" state, and results in nucleus-mantle retention. survival after cases of expulsion, and total failures). We used the respiration rate as index to describe recovery profiles of the presurgery therapies tested in this study.

MATERIALS AND METHODS

Selection of Therapies

Previous essays with menthol and benzocaine gave deficient results in presurgery sedation of P. mazatlanica. Propylene phenoxetol was tested in P. maxima (Mills et al. 1997) and P. margaritifera (Norton et al. 2000). Both species had suitable relaxation with 2.5 to 5 mL [L.sup.-1] at exposures of 15 to 20 min, but total failures in round pearl induction in P. margartifitera were associated with this drug (Norton et al. 2000). Heasman et al. (1995) found no apparent effect in Pecten fumatus at doses of 0.6 mL [L.sup.-1]. In P. radiata, Ehteshami (1993) obtained good results with 1 mg [L.sup.-1] of MS-222. The effect of this substance has been investigated in a wide range of doses and exposure times for several species of fishes and mollusks (i.e., Oncorhynchus mykiss by Wagner et al. 2002; Ostrea edulis by Culloty & Mulcahy 1992; Haliotis iris by Aquilina and Roberts 2000; Elliptio complanata by Lellis & Plerhoples 1997; and Lellis et al. 2000). Heasman et al. (1995) reported hyperactivity and hiperextension of tentacles in Pecten fumatus when exposed to 0.1 g [L.sup.-1] of MS-222. Quinaldine is commonly used in fish transport and broodstock handling; as in the previous drugs, there are different exposure times or doses reported as efficient, depending on the species and type of studies (i.e., Sparus aurata, Kumlu & Yanar, 1999; Massee et al. 1995 in larvae of Sciaenops ocellatus and Carassius auratus). Studies on the effect of quinaldine in mollusks are rare. Heasman et al. (1995) observed valve closure in Pecten fumatus when exposed to this substance. Apparently, there is no information regarding pearl oysters.

Experimental Design and Data Treatment

Approximately 3300 cultivated P. mazatlanica (wild spat collection) 14 to 16 mo old, were selected for this experiment in September 2001. We confirmed 39/45 specimens in rest or post-spawning stage as expected in the end of reproduction season (Saucedo & Monteforte 1997). A total of 342 were size-selected and divided in three groups of 108 individuals corresponding to each drug: propylene phenoxetol (PP), MS-222 (MS), and quinaldine (QN), plus 18 as controls (CTRL). CTRL monitoring was made in the first place to establish the "standard" profile, then PP, MS, and QN. In each session, pearl oysters were cleaned, measured for shell height (mean = 108.8 [+ or -] 6.7 mm SD), and left in starvation for 24 h in a 1500 L tank with filter-aerated seawater stable at 24 [+ or -] 1.5[degrees]C SD throughout the experiments, which is optimal for the species (Saucedo et al. 2003). With exception of CTRL, trials of 18 individuals were exposed to 3 different doses for each drug during 30 and 60 min into 20 L containers filled with filtered seawater (Table 1). These variables were deduced from the therapies previously outlined. When concluded, a particular exposure time, wet weight (mean = 161.2 [+ or -] 33.1 g SD) was measured after placing pearl oysters in ventral position and open for few seconds to drip water excess. Along with 9 untreated individuals per "exposure" time, they were transferred into respective respiration chambers already stabilized on oxygen concentration (Saucedo et al. 2003). Three YSI-58 oxygen-meters provided records every 10 or 20 min (Table 1), ending when the variation between subsequent measurements trended to stabilize and/or to appear alike CTRL. Calibration of YSI meters was checked at start and end of each experimental session in comparison with blank respirations chambers (1 per session). Normality for shell height was confirmed with a K-S test (d = 0.10, P > 0.2). Wet weight had differences, but the significance level was conservative (d = 0.15, * P < 0.05). Correlation of both dimensions was positive and significant (F = 278.75, R = 0.87, *** P < 0.001). Saucedo and Monteforte (1997) and Saucedo et al. (1998) considered wet weight acceptable as identification tag for individuals this size/age. In our study, similar overall reproductive stage among individuals would have contributed to reduce wet weight outliers. Respiration rate (Rr) was analyzed as index of recovery within and between drugs, and related to CTRL, through the formula:

Rr([micro]L [L.sup.-1] [O.sub.2] [g.sup.-1] [min.sup.-1] = [(OxTo - OxTn)/Wet weight]/Tn

OxTo and OxTn are starting and subsequent YSI records in mg [L.sup.-1] of oxygen, converted to [micro]L [L.sup.-1] of [O.sub.2] (Lehninger 1982). This unit was convenient for the strategies of data management and definition of plot patterns in which the purposes of this work were based. Wet weight (in g) is the individual tag, and Tn is the corresponding recovery time stage expressed in logs (T) of 10 or 20 min. Response profile was explored with median Box-Whiskers tests on raw Rr data (Fig. 1), this underlines individual behavior and its influence on profiles which represents important criteria for therapy evaluation. Slope attributes and intercept of data sets were obtained by regression methods. Their reliability was inferred with one way ANOVA F and corresponding regression coefficient (R). Covariance analyses were calculated to assess differences in slopes (Fcalc > Ftab, * P < 0.05) or elevations (Fcalc < Ftab, P > 0.05) (Table 2). The influence of recovery time on effects of dose, exposure time, and their interactions, was examined with three-way ANOVA (time x dose x exposure) and compared with CTRL based on their respective P level of significance (Table 3). To recognize main effects of dose and exposure time within drug, the post-hoc HSD Tukey sequence was applied with recovery time as contrast background. Further elements for therapy evaluation were provided by two-way ANOVA and HSD Tukey on Rr records at the start and end of each experiment (Table 4) (Sokal and Rholf 1981, Dillon and Goldstein 1984, Zar 1999, CSS-Statistica 1999). Survival was monitored 6 mo in field culture installations.

[FIGURE 1 OMITTED]

RESULTS

Blank respiration chambers remained stable in all sessions between 6.55 and 6.70 mg [L.sup.-1]. Calibration of YSI meters was seldom needed. Untreated (CTRL) P. mazatlanica presented highly significant negative correlation of Rr over time (F = 121.2, R = 0.85, *** P < 0.001). Intercept was 0.20 [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1] and slope -0.004 (Table 2); the P level of "recovery" time layed on [10.sup.-18] (Table 3). Overlap of time-logs in CTRL occurred between T40 and T50 (Fig. 1). Final record of Rr at T50 was 0.05 [+ or -] 0.01 [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1] (Table 4). Some individuals at T30 also had low Rr, whereas T10 was grouped highest (Fig. 1). Significant negative correlation was also detected for pearl oysters treated with PP and MS (F = 28.39 and 177.62, ** P < 0.01), whereas for QN the correlation was positive (F = 114.69, ** P < 0.01). The general response profile in all drugs was highly dependent on recovery time with P levels of [10.sup.-10] to < [10.sup.-24]; particular patterns were noted among treatments and in comparison with CTRL (Fig. 1, Tables 2-4). Trials into PP and MS seemed to share a common model near to CTRL, whereas notable differences were detected in pearl oysters exposed to QN (Fig. 1). R coefficient was low in PP and better in MS (0.37 and 0.74, respectively), whereas in QN was 0.61; CTRL had 0.85. This review indicated intrinsic effects of drug in the recovery profiles of P. mazatlanica. Dose and/or exposure time delineated particular patterns within and between drugs. Results in CTRL were essential to understand these profiles. We examined each experiment.

Propylene Phenoxetol

Significant negative correlation of Rr related to recovery time was confirmed for most cases (** P < 0.01). In these, Rr intercept was from 0.12 to 0.21 [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1], R settled into 0.52 and 0.64, and slope into -0.003 and -0.001. Individuals at dose of 5 mL in both exposure times were exceptions (Fig. 1, Table 2). The regression test and R values seemed sensible to individual differences within PP, particularly in these exceptions. However, the overall Rr range was lower than in MS or CTRL (Fig. 1, Tables 2 and 4) and tendency to horizontality seemed more notable (Fig. 1, Table 2). The covariance analysis indicated highly significant differences in slopes (Fcalc = 4.40 > Ftab [approximately equal to] 2.27, *** P < 0.001) and elevations (Ftotal = 33.55 > Ftab [approximately equal to] 2.81, *** P < 0.001). The effect of dose had important influence in these estimates, recovery time shortly followed as evident background, whereas exposure time had less significance; P level of dose and exposure time was similar as in other drugs (Table 3). Interaction of dose with exposure time, and of recovery time with dose also had rather equivalent significance as in MS and QN, while nonsignificant influence of recovery time was detected on the effect of exposure time or the combination of all effects (Table 3). Dose of 5 mL at both exposure times tended to horizontality from T10. Recovery time had no influence as all-effect, confirming a common background response profile for all PP treatments without major Rr outliers (Fig. 1. Table 3). The three-way post-hoe HSD Tukey test with contrasted recovery time indicated that pearl oysters had low Rr in high dose or long exposure. On the contrary, low dose and short exposure introduced data dispersion, high intercept value and more slope inclination (Fig. 1, Table 2). Exposure for 60 min promoted homogeneity of recovery profiles for all doses; the general Rr range in each time stage tended to be low and narrow (Fig. 1, Tables 2 and 4). Pearl oysters exposed for 30 or 60 min into 5 mL had stable Rr during recovery, with nearly horizontal slope and low intercept value (Fig. 1. Tables 2 and 4). The two-way ANOVA-HSD Tukey at T10 of recovery indicated highly significant differences of Rr values for dose (F = 15.22, *** P < 0.001) but not for exposure time (F = 3.54, P > 0.05), neither their interaction (F = 2.78. P > 0.05). In this stage, individuals treated with higher dose and/or longer exposure time had lower and narrower Rr, coinciding with low intercept and almost horizontal slope (Fig. 1, Table 4). At T40 there was no effect of dose (F = 2.8, P > 0.05), whereas exposure time had moderated importance (F = 5.23, * P < 0.05). Their interaction was not significant (F = 2.0, P > 0.05). In this stage, differences of Rr values were significant only between batch of 1 mL-30 min (highest Rr) and the one of 5 mL-60 min (lowest Rr) (Fig. 1, Table 4). Survival was 100%.

MS-222

Correlation of Rr over time was negative and significant in all treatments (*** P < 0.001). The intercept varied between 0.18 and 0.30 [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1] and R between 0.71 and 0.90. Slope values predominated in -0.003 and -0.004, with exception of one only case of -0.002 (Fig. 1, Table 2). The overall range of R coefficient appeals for more uniform variation of Rr within MS, frequently over 0.79 (Table 2), but this may not be clinically conclusive. Covariance analysis revealed conservative differences in slopes (Fcalc = 2.32 > Ftab [approximately equal to] 2.27, * P < 0.05) while elevations appeared more diverse (Ftotal = 28.31 > Ftab [approximately equal to] 2.81. *** P < 0.001) (Fig. 1, Table 2). The effect of time in the recovery profiles under MS seemed most important than in other drugs (Table 3). Dose, exposure time, and their interaction were significant with similar P level as in PP and QN. All-effect interaction was significant but at lowest P level: although code mixing was frequent at T50, overlap order of T30 and T50 occurred seldom within sets and most cases of T10 tended to group at higher Rr (Fig. 1). This suggests that the influence of recovery time was probably overestimated: the inclusion of a T40 log would have increased overlapping events between this and T50 (as in PP and CTRL), while the inclusion of a T20 would have scarcely affected the first (higher Rr) position of T10 logs (i.e., T10 of datasets 2.5 mg-60 min and 7.5 mg-30 min). The sudden rise of Rr at T30 and T50 in trial of 7.5 mg-60 min corresponds to three specimens (Fig. 1), apparently provoked by mechanical disturbances (noise or motion). The contrasted three-way HSD Tukey indicated that pearl oysters had high Rr and dispersion at lower doses for both exposure times. The experiment of 2.5 mg-60 min presented highest slope and intercept value with dispersion toward high Rr. Individuals placed into 5 mg for 30 and 60 min had less variation in their recovery profile, also coinciding with lowest intercept and "average" -0.003 slope. Longer exposure time in doses of 5.0 and 7.5 mg apparently contributed to decrease individual differences; however, this effect was only observed when comparing early recovery stages of both exposure times (Fig. 1, Tables 2 and 4). A related analysis with elimination of outliers in T30 and T50 for 7.5 mg-60 min dataset did not alter the similarity of trials with 5.0 mg for both exposure times. The two-way ANOVA-HSD Tukey test applied at T10 and T50 of recovery supports the above. In the first time-log, the effect of dose was highly significant (F = 14.95, *** P < 0.001), so was its interaction with exposure time (F = 11.19, *** P < 0.001), while this last apparently had no influence (F = 1.19, P > 0.05); in the second time-log, only the effect of dose was significant (F = 23.97, *** P < 0.001). At T10, individuals of 7.5 mg-30 min and 2.5 mg-60 min had significantly higher Rr (also higher intercept and slope inclination) than those treated with 5 mg at both exposure times (lower intercept and "average" slope) (Tables 2 and 4). At T50, pearl oysters placed into 2.5 mg during 30 and 60 min of exposure had higher Rr, while the rest were relatively similar (Table 4). Elimination of outliers 7.5 mg-60 min at T50 made no impact in this pattern. Survival was 100%.

Quinaldine

All treatments had significant positive correlation of Rr over time (** P < 0.01). R coefficient varied between 0.50 and 0.88, intercept from 0.001 and 0.03 [micro]L [L.sup.-1] [g.sup.-1] [min.sup.-1] and slope from 0.0003 to 0.0007 (Table 2). Overall range of R values suggests slightly better reliability of this predictors than in PP. Besides, the absolute variation of Rr within QN was narrow and had lower values than in other drugs and CTRL, almost throughout the experiment (Fig. 1). The covariance analysis indicated similarity of slopes (Fcalc = 1.84 < Ftab [approximately] 2.27, P > 0.1 < 0.25), whereas elevations were highly different (Ftotal = 58.54 > Ftab [approximately equal to] 2.81, *** P < 0.001). The effect of recovery time was very important in this pattern at P level of [10.sup.-24]. Similar to other drugs and CTRL, Rr of pearl oysters under QN followed a main background profile related with time, although in positive direction. Dose also had strong influence but at relatively less extent than in PP or MS, whereas exposure time and its interaction with the former had similar P level as in other drugs. All-effects interaction was not significant (Table 3). Overlap of time-logs occurred in all datasets for T60 and T80 min of recovery (Fig. 1), but in some trials this started earlier coinciding with larger individual differences of Rr (i.e., from T20 in 15 mL-60 min, and from T40 in 45 mL-30 min). Cases of T20 tended to group in last place (low Rr) except those of 15 mL (Fig. 1). The three-way HSD Tukey with recovery time as contrast suggests significant differences between lowest Rr in 45 mL-60 min and highest in 15 mL-30 min, coinciding respectively with low and high intercept and similar slope. Differences of recovery profiles in treatments with 30 and 45 mL at both exposure times were not substantial. Response homogeneity among individuals seemed to increase with dose, while variation of Rr increased with longer exposure times. R value of each batch also indicate individual differences in 15 mL at both exposure times, while moderate to low dispersion was detected for other treatments (Table 2). However, the general trace of Rr values over recovery time did not tend to dispersion declining as in other drugs or CTRL (Fig. 1). In some treatments, individual differences remained rather constant, either in low Rr such as those of 45 mL in both exposure times or high Rr like 15 mL in both exposure times. Dispersion increase seemed more evident in dose of 30 mL at both exposure times (Fig. 1, Table 2). The analysis at T20 and T80 of recovery stands for this pattern (Table 4). At T20, the effect of dose was significant (F = 9.12, *** P < 0.001), while not the effect of exposure time (F = 2.39, P > 0.1), neither their interaction (F = 0.37, P > 0.5). Batches of 15 mL-30 min and 15 mL-60 min had higher Rr but dispersion weighed toward low values in the first and high values in the second (Fig. 1); difference between these was not significant. Significantly low Rr oscillated between 0.011 and 0.014 [micro]L [g.sup.-1] [min.sup.-1] at T20, the first--and lowest of all drug sets--corresponding to 45 mL-60 min. At T80, the effect of dose was significant (F = 6.44, ** P < 0.01), whereas exposure time had no effect (F = 0.003, P > 0.5). Their interaction was not significant (F = 0.82, P > 0.1). Differences of Rr were evident only between 15 and 45 mL in the same exposure time (30 min). The first started (T20) and finished (T80) with significantly higher Rr; it also presented the highest intercept value and similar slopes for each exposure time but also the lowest R value. The second had lower Rr, coinciding with low intercept and the most horizontal slope of each exposure time; R values were higher than in doses of 15 mL. Apparently, individuals treated with 45 mL still had low Rr at T80 (Fig. 1, Tables 2 and 4). There was no mortality during the experiment.

Mortality in Field Culture

The generation of P. mazatlanica corresponding to this study accumulated 4% of average mortality in six months, with frequent monthly zeroes in many containers. This behavior is normal for the species (Monteforte et al. 2000). In general, the current experiment did not diverge from this pattern. Mortality after 6 mo was: 1 individual for CTRL and QN, respectively, and 2 for PP and MS, respectively. We noted that all incidence of mortality occurred in the first month of field culture. Survival was 100% until the sixth month.

DISCUSSION

Evaluation of presurgery therapies for P. mazatlanica was based on the effect of dose, exposure time, and interactions within and between drugs, and their comparison with the model pattern of CTRL. These therapies were explained as index of recovery or "stress level" by the position of intercept and Rr value at starting T, by the pattern of dispersion and range of individual differences during recovery, and by the attributes of slope, including time needed for apparent stabilization toward a "common" Rr value (Fig. 1, Tables 2 and 4). Stress has been interpreted as physiologic response involving particular meaning when related to anesthetics. Records of corticosteroids have been used in studies regarding stress indexes for several species of fish. In most cases, drugs themselves appeared as stressors and did not reveal special advantages to prevent physiologic stress responses (Wagner et al. 2002). In our work, the monitoring of Rr was considered a reliable index to interpret the response of P. mazatlanica under different therapies. The model pattern in CTRL started at initial T with differencial activity, most individuals into conditions of high intercept and Rr value, followed by rapid decrease of dispersion and Rr value until apparent stabilization at about 0.05 mL [L.sup.-1] in T50 (Fig. 1, Tables 2 and 4). Significance of recovery time in this model was on [10.sup.-18] P level (Table 3). Pearl oysters under therapy also were strongly influenced by recovery time but with wide differences of P level from [10.sup.-10] to < [10.sup.-24] between drugs (Fig. 1, Table 3). This suggest that P. mazatlanica bad a dynamic overall response during recovery acting as main background variation and strongly affected by time, while dose, exposure time, and their interactions contributed to arrange particular patterns for each therapy regarding the CTRL model. Dose had equivalent P level in PP and MS, and relatively less for QN; the effect of exposure time and interaction with dose were also grouped (Table 3). Differences in significance P level of recovery time may reside on the nature of drug components and their specific influence on mechanical and physiologic functions of also specific body parts (Iversen et al. 2003). In these terms, the appraisal of relationships among treatments and therapy profiles was supported with a cluster analysis of the multiple covariance matrix using complete linkage rule and Euclidian distances (Sokal and Rholf 1981, Dillon & Goldstein 1984, Zar 1999, CSS-Statistica 1999). As expected, the effect of drug appeared as important discriminant between treated and untreated individuals (Fig. 2). Main clusters distinguished recovery profiles based on activity level, from higher in CTRL to lower in QN (Fig. 1, Tables 2 and 4). CTRL stated at 60% of definition. Trials into PP, MS seemed to share a common model near to CTRL: high activity and/or dispersion of Rr in early phase, dispersion declining toward low Rr as recovery progressed, and apparent stabilization at about 50 min. Aside their different slope attributes, pearl oysters exposed to QN had no evident dispersion declining of Rr through time as in the previous drugs and CTRL; however, overall absolute variation remained narrow and values were much lower in QN. Stabilization was considered present at 80 min (Fig. 1). The "stress" cluster was formed by all MS datasets and PP 1 mL-30 min and it remained to 44%. Pearl oysters in these trials presented lower Rr dispersion than CTRL and also a tendency to decrease over time, however, their recovery profile appeared similar in several elements that suggest unadequate relaxation (Fig. 1, Tables 2 and 4). Notably, trial of PP 1 mL-30 min joined this cluster at 16% apparently because it had the highest intercept, slope, and covariance values within this drug. Its attachment to "MS" cluster was on trial MS 2.5 mg-30 min, as confirmed by their similar profiles (Figs. 1 and 2, Tables 2 and 4). Within "MS," clusters were apparently defined by the dispersion pattern of Rr and its position in the plot (intercept and Rr value at initial T). The cluster formed by trials with 5 mL at both exposure times and trial 7.5 mg-60 min indicate homogeneity of Rr toward low-activity profiles (i.e., low intercept, moderate slope, dispersion decrease, and low Rr value at final T) (Fig. 1, Tables 2 and 4). Components of cluster "PP" suggest that higher dose promoted low disturbances of Rr. Treatments of 5 mL in both exposure times joined at 4% and remained as separate cluster up to 20%. Lower dose of this substance and/or lesser exposure time provoked a similar response pattern as in the "stress" cluster, this was notable in the mentioned trial 1 mL-30 min of PP (Figs. 1 and 2; Tables 2 and 4). With this exception, both trials of 3 mL and the one of 1 mL-60 min stayed as cluster of higher activity within PP to 12% (Fig. 2). Cluster "PP" joined the "QN" nearly at 40% with the trials of low-profile PP 5 mL of both exposure times and stayed separated from the previous "CTRL" + "MS" through 100%, however, cluster "QN" remained defined only at 10%. The recovery profile of pearl oysters treated with QN had very low intercept position and Rr value at starting T, and lower range during recovery in comparison with the other experiments. Slope was nearly horizontal but stabilization was considered present later than in CTRL and other treatments. Components of cluster "QN" were apparently defined by the dispersion pattern of Rr and intercept. Low-profile response (lowest of all trials) could be considered for clusters of 45 mL, where dispersion of Rr during recovery was narrow but it remained rather constant. The QN trial of 15 mL-60 min was apparently attached to the former for its tendency to low intercept and constant dispersion, although this last was larger than in doses of 45 mL. Lower dose or exposure time under QN promoted tendencies to increase Rr dispersion; these trials compose the high-activity area within cluster "QN," although their range value remained lower than in CTRL, MS, and PP (Figs. 1 and 2). The overview suggests that resemblance or difference of recovery profiles in CTRL pearl oysters and those under treatment were principally defined during early phases. Trials with MS and PP coincided in having a negative correlation over time indicating recovery from stressed state, but relatively less than in CTRL. Specimens placed into QN had slow recovery from depressed activity toward higher Rr (Fig. 1). This suggest that drugs were no stressors for P. mazatlanica. It was also evident that the overall projection of recovery profiles moves toward a Rr of 0.05 [micro]L [g.sup.-1] [min.sup.-1], assumed from measures in controls at T50 (Fig. 1, Table 4). Specimens under PP and MS had rather similar end Rr as in CTRL at T50, but it took longer time for those exposed to QN. Even at T80, a good proportion of pearl oysters still were less active than CTRL and some other drug trials. However, it would be inexact to consider these cad values as probable standard rate for the species because particular studies should be previously applied. This information is primary when the physiologic response of an organism to environmental changes is measured or interpreted through some index of stress over time. In marine bivalves, stress has been explained by functions such as clearance and filtration rate, heart beat, components of excretion, oxygen consumption, and so forth. (Bayne et al. 1976a, Bayne et al. 1976b). This last function is implicit in the recovery index used in this work. Bayne et al. (1976a) stated that Mytilus edulis under starvation rapidly arrives to standard respiration rate when gametogenesis is completed and energy reserves of the body are at the minimum. In this context, we assumed that all individuals entered the current experiment at a similar level of activity. Nevertheless, the comparative information of oxygen consumption in bivalves is controversial due to the variety of measurement methods, units used, and experimental designs (Shumway 1982). Concerning pearl oysters, Yukihira et al. (1998) reports Rr (R in the cited work) of 0.857 and 1.039 mL [O.sub.2] [oyster.sup.-1] [h.sup.-1] in P. maxima and P. margaritifera, respectively, as intercept values from a significantly inclined and positive regression of Rr, or R, over dry weight. The authors considered these species as exceptional in having the highest indexes of energy flux recorded for bivalves. Saucedo et al. (2003) defined respiration rate of 0.4 to 0.8 [micro]mol [g.sup.-1] [h.sup.-1] in fed and acclimatized P. mazatlanica between 18[degrees]C and 33[degrees]C. With these considerations, presurgery treatments in P, mazatlanica are difficult to explain in terms of physiologic response, and this lack of information calls for such studies. Aside from the intrinsic effect of chemicals even into complex body functions (Wagner et al. 2000; Eversen et al. 2003), the induction surgery would initiate hardly known immunologic processes in pearl oysters (i.e., Ortuno et al. 2002 in Sparus aurata). Norton et al. (2000) associated the use of PP with total failures in round pearl induction on P. margaritifera (mortality + nucleus rejection).

[FIGURE 2 OMITTED]

CONCLUSIONS

The best therapies in P. mazatlanica should sustain a nearly horizontal slope and low Rr-intercept during recovery; homogeneity of individual response is also important. These observations may comply with other species of pearl oysters. In real practice, the use of QN at doses of 45 mL eased the surgery for round pearl due to adequate and longer relaxation of specimens, but protection equipment is recommended. Treatments with 5 mL of PP were also acceptable, which suggest that higher dose and/or exposure time under this drug would provoke a "QN"-like response profile in this species, probably with shorter recovery time. Doses of MS-222 between 5 and 7.5 mL at 60 min of exposure could be used in P. mazatlanica for Mabe implants only. All specimens treated with this therapy had reasonably relaxed their abductor muscle, and mantle retraction was none or very slow; however, we noted about 70% of them arriving with contracted gonadal tissue and ligaments and/or foot muscle, which introduced a problematical approach to the insertion site. Based on our results, there is no evidence that longer exposure and/or higher dose of MS-222 might lead to low-activity profiles. Aside from the need for studies regarding the physiologic response profiles and relationship with functions of body parts, tolerance tests to these three drugs are important. In this experiment, mortality could not be attributed to any drug into the effective ranges of therapies. Incidence of mortality in field culture suggest handling effects.

TABLE 1.
Experimental design of anesthesic therapies with three drugs.

                                Exposure Time        Recovery Time

        Drug Dose                   (min)                (min)

Propylene phenoxetol (PP)                                 10
  1 mL, 3 mL. 5 mL                                        20
Finquel MS-222 (MS)                   30                  30
  2.5 mg, 5 mg. 7.5 mg                                    40
                                      60                  50
Quinaldine (QN)                                           60
  15 mL, 30 mL, 45 mL                                     80
Control (CRTL) (n = 18)     Untreated                  10 to 50
                              (n = 9 sims/Exp. T.)

Trials under treatment (dose, exposure time) consist of n = 18
individuals each. CTRL specimens were divided in groups of
n = 9 for each "exposure" time. Experimental sessions were not
simultaneous: CTRL was first, PP second, MS third, and QN last.

TABLE 2.
Statistical calculations to infer regression attributes and covariance
comparisons in recovery profiles of the Calafia pearl oyster, Pinctada
mazatlanica, under different relaxation therapies with propylene
phenoxetol (PP), MS-222 (MS) and quinaldine (QN).

                            Rr Intercept
                              ([micro]L
                             [g.sup.-1]               ANOVA F
Group   Exposure   Dose     [min.sup.-1])    Slope    Regression

CTRL       --        --         0.20        -0.004    121.22 ***
PP         30        1 mL       0.21        -0.003     15.05 ***
                     3 mL       0.12        -0.001     14.99 ***
                     5 mL       0.08        -0.000      0.32
           60        1 mL       0.15        -0.002     11.93 ***
                     3 mL       0.13        -0.001      8.23 **
                     5 mL       0.09        -0.001      4.17
MS         30      2.5 mg       0.21        -0.002     22.66 ***
                   5.0 mg       0.19        -0.003     48.72 ***
                   7.5 mg       0.26        -0.004     92.55 ***
           60      2.5 mg       0.30        -0.004     45.28 ***
                   5.0 mg       0.18        -0.003     83.86 ***
                   7.5 mg       0.22        -0.003     37.01 ***
QN         30       15 mL       0.03         0.0004    11.65 ***
                    30 mL       0.01         0.0006    36.68 ***
                    45 mL       0.01         0.0003    22.55 ***
           60       15 mL       0.01         0.0005     8.60 **
                    30 mL      -0.001        0.0007    91.82 ***
                    45 mL       0.002        0.0004    47.42 ***

                            R. Regression                Post hoc
Group   Exposure   Dose      Coefficient    Covariance   HSD Tukey

CTRL       --        --         0.85          -0.66         --
PP         30        1 mL       0.64          -0.26         d
                     3 mL       0.64          -0.17         abc
                     5 mL       0.12          -0.06         ab
           60        1 mL       0.59          -0.23         c
                     3 mL       0.52          -0.14         bc
                     5 mL       0.40          -0.04         a
MS         30      2.5 mg       0.71          -0.67         c
                   5.0 mg       0.83          -0.75         ab
                   7.5 mg       0.90          -1.17         bc
           60      2.5 mg       0.82          -1.03         d
                   5.0 mg       0.89          -0.74         a
                   7.5 mg       0.79          -0.91         abc
QN         30       15 mL       0.56           0.17         c
                    30 mL       0.76           0.21         ab
                    45 mL       0.68           0.11         ab
           60       15 mL       0.50           0.12         bc
                    30 mL       0.88           0.23         ab
                    45 mL       0.80           0.22         a

Significance of the regression ANOVA F is * P < 0.05, ** P < 0.01,
and *** P < 0.001. The post hoc HSD Tukey considered recovery time
as contrast factor. Different indexes denote significant differences
at * P < 0.05.

TABLE 3.
Results of three-way ANOVA to determine influence and P levels of
factors and interactions in the recovery profiles of treated and
untreated Pinctada mazatlanica.

     Effect                CTRL                   PP

Recovery time       2.20 x [10.sup.-18]   2.11 x [10.sup.-10]
Dose                        --            2.97 x [10.sup.-14]
Exposure time               --            1.24 x [10.sup.-2]
Recovery time               --            7.81 x [10.sup.-3]
  vs. dose
Recovery time vs.           --                   0.34
  exposure time
Dose vs.                    --            1.63 x [10.sup.-4]
  exposure time
All effects                 --                   0.81

     Effect                 MS                    QN

Recovery time          <[10.sup.-24]      1.99 x [10.sup.-24]
Dose                4.56 x [10.sup.-15]   4.91 x [10.sup.-11]
Exposure time       4.84 x [10.sup.-2]    1.84 x [10.sup.-3]
Recovery time              0.86                  0.22
  vs. dose
Recovery time vs.          0.91                  0.15
  exposure time
Dose vs.            1.52 x [10.sup.-5]    9.75 x [10.sup.-3]
  exposure time
All effects         2.76 x [10.sup.-2]           0.96

CTRL, control; PP, propylene phenoxetol; MS, MS-222; QN, quinaldine.

TABLE 4.
Results of two-way ANOVA and HSD'fukey at first and final Rr
for each drug therapy.

                                    ANOVA F
        Exposure                   First Rr
Group    (min)      Dose      (Mean [+ or -] SD)

CTRL       --        --      0.25 [+ or -] 0.07
PP         30        1 mL    0.19 [+ or -] 0.09 b
                     3 mL    0.12 [+ or -] 0.03 a
                     5 mL    0.09 [+ or -] 0.01 a
           60        1 mL    0.13 [+ or -] 0.07 ab
                     3 mL    0.12 [+ or -] 0.03 a
                     5 mL    0.09 [+ or -] 0.01 a
MS         30      2.5 g     0.20 [+ or -] 0.05 ab
                   5.0 g     0.17 [+ or -] 0.04 a
                   7.5 g     0.23 [+ or -] 0.03 bc
           60      2.5 g     0.28 [+ or -] 0.05 c
                   5.0 g     0.17 [+ or -] 0.02 a
                   7.5 g     0.20 [+ or -] 0.02 ab
QN         30       15 mL   0.035 [+ or -] 0.019 b
                    30 mL   0.014 [+ or -] 0.011 a
                    45 mL   0.017 [+ or -] 0.004 ab
           60       15 mL   0.026 [+ or -] 0.019 ab
                    30 mL   0.013 [+ or -] 0.006 a
                    45 mL   0.011 [+ or -] 0.007 a

                                    ANOVA F
        Exposure                   Final Rr
Group    (min)      Dose      (Mean [+ or -] SD)

CTRL       --        --      0.05 [+ or -] 0.01
PP         30        1 mL    0.11 [+ or -] 0.03 b
                     3 mL    0.08 [+ or -] 0.01 ab
                     5 mL    0.08 [+ or -] 0.01 ab
           60        1 mL    0.08 [+ or -] 0.02 ab
                     3 mL    0.08 [+ or -] 0.02 ab
                     5 mL    0.07 [+ or -] 0.02 a
MS         30      2.5 g     0.10 [+ or -] 0.03 bc
                   5.0 g     0.07 [+ or -] 0.02 a
                   7.5 g     0.07 [+ or -] 0.01 a
           60      2.5 g     0.13 [+ or -] 0.03 c
                   5.0 g     0.06 [+ or -] 0.01 a
                   7.5 g     0.07 [+ or -] 0.04 ab
QN         30       15 mL   0.058 [+ or -] 0.009 b
                    30 mL   0.048 [+ or -] 0.010 ab
                    45 mL   0.040 [+ or -] 0.008 a
           60       15 mL   0.053 [+ or -] 0.017 ab
                    30 mL   0.054 [+ or -] 0.015 ab
                    45 mL   0.040 [+ or -] 0.011 a

CTRL, control; PP, propylene phenoxetol; MS, MS-222; QN, quinaldine.

LITERATURE CITED

Alagarswami, K. 1987. Technology of cultured pearl production. In: K. Alagarswami, editor. Pearl culture. Bull. Cent. Mar. Fish Res. Inst. No. 39.

Aquilina, B. & R. Roberts. 2000, A method for inducing muscle relaxation in the abalone, Haliotis iris. Aquaculture 190(3-4):403-408.

Bayne, B. L., R. J. Thompson & J. Widdows, 1976a. Physiology I. In: B. L. Bayne. editor. Marine mussels: their ecology and physiology. International Biological Program 10. Cambridge University Press. pp. 121206.

Bayne, B. L., J. Widdows & R. J. Thompson. 1976b. Physiology II. In: B. L. Bayne, editor. Marine mussels: their ecology and physiology. International Biological Program. Cambridge University Press. pp. 207-260.

CSS-Statistica. 1999. Statistica software. StatSoft Inc. Tulsa, OK.

Culloty, S. C. & M. F. Mulcahy. 1992. An evaluation of anaesthetics for Ostrea edulis (L.). Aquaculture 107(2-3):249-252.

Dillon, W. R. & M. Goldstein. 1984. Multivariate analysis: methods and applications. New York: John Wiley & Sons. Inc.

Ehteshami, F. 1993. Anaesthetizing Pinctada radiata with MS 222. Iran Fish. Bull. 3:1.

Fassler. R. 1994. Pearls '94 International Conference and Exposition. Honolulu, Hawaii. May 1994. Collection of Abstracts. J. Shellfish Res. 13:40 pp.

Iversen, M., B. Finstad, R. S. McKinley, et al. 2003. The efficacy of metomidate, clove oil. Aqui-S and Benzoak (R) as anaesthetics in Atlantic salmon (Salmo salar L.) smolts, and their potential stress-reducing capacity. Aquaculture 221(1-4):549-566.

Gervis, M. H. 1992. A bibliography of the pearl oysters (Bivalvia: Pteriidae). ICLARM Contribution No. 765. Bibliogr. 11:99.

Haws, M. 1998. Nucleus implantation for cultivation of black pearls: review of current technology and constraints for development of the industry. Book of abstracts. WAS 1998, 15-19 February, Las Vegas, NV.

Haws, M., J. Tapu, A. O. Bailey. et al. 1999. An educational module for training grafting technicians for the black pearl culture industry. Book of abstracts. WAS 1999, 26 April-2 May. Sidney, Australia.

Heasman, M. P., W. A. O'Connor & A. W. J. Frazer. 1995. Induction of anaesthesia in the commercial scallop, Pecten fumatus Reeve. Aquaculture 131:231-238.

Kumlu, M. & M. Yanar. 1999. Effects of the anesthetic quinaldine sulphate and muscle relaxant diazepam on sea bream juveniles (Sparus aurata). Israeli Journal of Aquaculture/Bamidgeh 51(4):143-147.

Lellis, W. A. & T. A. Plerhoples. 1997. Development of anesthetics for the mussel Elliptio complanata. J. Shellfish Res. 16(1):355-356.

Lellis, W. A., T. A. Plerhoples & K. A. Lellis. 2000. Evaluation of potential anesthetics for the freshwater mussel Elliptio complanata. J. Shellfish Res. 19(2):983-990.

Lehninger, A. L. 1982. Biochemistry. New York: Worth Publishers.

Massee, K. C., M. B. Rust, R. W. Hardy, et al. 1995. The effectiveness of tricaine, quinaldine sulfate and metomidate as anesthetics for larval fish. Aquaculture 134(3-4):351-359.

Mills, D., A. Tlili & J. Norton. 1997. Large-scale anesthesia of the silver-lip pearl oyster, Pinctada maxima Jameson. J. Shellfish Res. 16(2):573-574.

Monteforte, M. 1996. Cultivo de Ostras Perleras y Perlicultura. In: G. Ponce & M. Casas, editors. Diagnostico Pesquero y Acuicola del Estado de Baja California Sur, Vol. II. Mexico D.F.: CIBNOR/ CICIMAR/UABCS/CET-MAR/SEMARNAP/FAO. pp. 571-613.

Monteforte, M., H. Bervera, S. Morales, et al. 1994. Results on the production of cultured pearls in Pinctada mazatlanica and Pteria sterna from Bahia de La Paz, South Baja California, Mexico. Pearls '94 International Conference and Exposition. Honolulu, Hawaii. May 1994. Abstracts. J. Shellfish Res. 13:344-345.

Nava, M., E. Arizmendi, S. Farell, et al. 2000. Evaluation of success in the seeding of round nuclei in Pteria sterna (Gould 1851), a new species in pearl culture. SPC Pearl Oyster Information Bulletin 14 (Dec): 12-16.

Norton, J. H., J. S. Lucas, I. Turner, et al. 2000. Approaches to improve cultured pearl formation in Pinctada margaritifera through use of relaxation, antiseptic application and incision closure during bead insertion. Aquaculture 184(1-2):1-17.

Ortuno, J., M. A. Esteban & J. Meseguer. 2002. Effects of four anaesthetics on the innate immune response of gilthead seabrearn (Sparus aurata L.). Fish & Shellfish Immunology 12:49-59.

Saucedo, P. & M. Monteforte. 1997. Breeding cycle of pearl oysters Pinctada mazatlanica and Pteria sterna (Bivalvia: Pteriidae) at Bahia De La Paz, Baja California Sur. Mexico. J. Shellfish Res. 16(1): 103-110.

Saucedo, P., M. Monteforte & F. Blanc. 1998. Changes in shell dimensions of pearl oysters Pinctada mazatlanica (Hanley 1856) and Pteria sterna (Gould 1852), during growth as criteria for Mabe pearl implants. Aquaculture Research 29:801-814.

Saucedo, P., L. Ocampo. M. Monteforte, et al. 2003. Effect of temperature on oxygen consumption and amonia excretion in the Calafia mother-of-pearl oyster, Pinctada mazatlanica (Hanley, 1856). Aquaculture.

Shirai, S. 1994. An overview of pearl oyster culture in the world. Master Conference. Pearls '94 International Conference and Exposition. Honolulu, Hawaii. May 1994.

Shumway, S. E. 1982. Oxygen consumption in oysters: an overview. Mar. Biol. Lett. 3:1-23.

Sokal, R. R. & F. J. Rholf. 1981. Biometry: the principles and practice of statistics in biological research, 2nd ed. New York: W.H. Freeman and Co.

SPC-POIB. Secretariat of the Pacific Community. Pearl Oyster Information Bulletin. N. Sims, editor. Marine Resources Division, Noumea. Nos. 1 to 14.

Wagner, E., R. Arndt & B. Hilton. 2002, Physiological stress responses, egg survival and sperm motility for rainbow trout broodstock anesthetized with clove oil, tricaine methanesulfonate or carbon dioxide. Aquaculture 211:353-366.

Yukihira, H., D. W. Klumpp & J. S. Lucas. 1998. Effects of body size on suspension feeding and energy budgets of the pearl oysters Pinctada margaritifera and P. maxima. Mar. Ecol. Prog. Ser. 170: 119-130.

Zar, J. H. 1999. Biostatistical Analysis, 3rd ed. Prentice-Hall International, Inc.

MARIO MONTEFORTE, * HORACIO BERVERA AND PEDRO SAUCEDO Centro de Investigaciones Biologicas del Noroeste, S.C. (CIBNOR), La Paz, Baja California Sur, CP 23090, Mexico

* Corresponding author. Fax: [phone]; E-mail: montefor04@ cibnor.mx