Effect of temperature, sexual maturity and sex on growth, food intake and gross growth efficiency in the "Pulpito" Octopus tehuelchus (d'Orbigny, 1834).

ABSTRACT The effects of temperature (10[degrees]C and 15[degrees]C), sexual maturity (immature and mature) and sex on instantaneous growth rates, daily feeding rates and gross growth efficiency were studied in the "Pulpito," Octopus tehuelchus (d'Orbigny, 1834), under laboratory conditions. At 15[degrees]C, immature octopus showed higher instantaneous growth and daily feeding rates than individuals maintained at 10[degrees]C (P < 0.01). At t0[degrees]C, immature individuals showed higher daily feeding rates than mature specimens (P < 0.01). Aside from a lack of measurable growth observed in mature males, sex appears to have had no effect on either instantaneous growth or daily feeding rates in either maturity stage or temperature treatment (P > 0.05). An alternative method to compare gross growth efficiencies (GGE) among 6 treatments, based on the relationship between growth and food intake, is proposed. Immature octopuses at 10[degrees]C showed a higher GGE value than at 15[degrees]C (P < 0.05). Mature individuals at 10[degrees]C were less efficient in using food in growth than immature ones. GGE value (52%) found for immature octopus at 10[degrees]C is among the highest reported in the literature. This could have important implications for a possible rearing activity of this species. Mature males and females at 10[degrees]C showed a lower GGE (35.5%) than immature individuals held at 10[degrees]C (GGE = 52%) and 15[degrees]C (GGE = 44%). Food intake, growth and food conversion are linked processes, which in O. tehuelchus are influenced by temperature and sexual maturity. In addition, whereas classic statistical tests failed to detect any of these effects on GGE, the alternative method here proposed based on the relationship of growth versus food intake was successful.

KEY WORDS: Octopus tehuelchus, gross growth efficiency, food intake, temperature, sexual maturity, Southwestern Atlantic

INTRODUCTION

The "Pulpito," Octopus tehuelchus (d'Orbigny, 1834) has an important role in octopus artisanal fishery in Argentina. The main capture period takes place from December to March-April with an official annual mean capture of 20 tons (R6 1998 a). O. tehuelchus is distributed from Porto Seguro (17[degrees]S), Brazil to San Jorge Gulf (45[degrees]S), Argentina and has an intertidal distribution from San Blas Bay (40[degrees]S) to Rawson (43[degrees]S), Argentina (Re, 1998b). This species has a marked seasonal growth, increasing its food intake activity and growth rates as temperature increases and down these rates when temperature decreases with means of 10 [+ or -] 1[degrees]C for the colder months and 15 [+ or -] 1[degrees]C for the warmer months. O. tehuelchus has a life-span of 18 mo of duration with dorsal mantle length and body weight of first maturation for males and females of 38mm-18g and 52mm-40g respectively (Re, 1989). Despite its economical importance to rural communities, there is no information about the effects of temperature, sexual maturity and sex on growth, food intake and gross growth efficiency under laboratory conditions in terms of a possible complementary rearing activity. Such basic studies are necessary to Understand the relationship between environment conditions on food intake, growth and life-span of a species in both natural and laboratory conditions.

In cephalopods, growth and food intake are strongly dependent on a suite of biotic and abiotic variables. Water temperature and ration levels are implicated as factors most obviously responsible for variations in growth (Mangold 1983, Forsythe & Toll 1991). Final size of male and female cephalopods may vary greatly among individuals of the same species, depending on factors such as food availability and quality, and temperature (Van Heukelem 1979). Such differences in the final size typically become dramatic only in mid to later life with the approach of sexual maturity (Forsythe & Van Heukelem 1987). In addition, there also appears to be much individual variation in growth rate and final size even within groups or siblings reared under identical conditions (Semmens et al. 2004).

The cephalopod literature contains few studies reporting relationships between growth and food intake. In octopus, the functional relationship between growth and food intake appear to be linear above maintenance requirement (Van Heukelem 1976, Joll 1977), and it depends mainly on the organism's efficiency at converting food into tissue growth and on physiological constraints. This relationship has been studied in laboratory conditions for only three tropical-subtropical species, Octopus maya (Voss/Solis Ramirez 1966), O. cyanea (Gray, 1849) (Van Heukelem 1976) and O. tetricus (Gould, 1852) (Joll 1977); no information is available for temperate species.

For many purposes, one of the most useful energetic measures is the gross growth efficiency (GGE), which expresses the amount of food intake required to fuel a unit amount of growth. It should be recognized, however, that any value calculated for GGE will vary with circumstances (temperature, locomotor activity, and similar) and is also dependent on food intake itself (specific dynamic action, maintenance ration) (Wells & Clarke 1996). The simplest way to solve this discrepancy is to compare the growth of animals at different ration levels and estimate the maintenance ration in terms of the weight of food is required to maintain a stable weight (O'Dor & Wells 1987). With this information we would be able to calculate the partial growth efficiency (PGE), which is the efficiency of the organism at converting food into growth, after the maintenance requirements have been satisfied (Kleiber 1961). In young stages of cephalopods very high relative growth rates may be achieved and although feeding rates increase with temperature, there is no empirical evidence that GGE change detectably with temperature (Mangold & Boletzky 1973, Van Heukelem 1987). Aguado Giminez and Garcia Garcia (2002) by mean of a multiple regression analysis observed in O. vulgaris (Cuvier 1797) a temperature dependence of GGE. Considering that GGE is a ratio of two variables (growth/food intake); are the classics statistics tests adequate to detect such differences between two or more GGE samples measured on two or more conditions? Moreover, could the great individual variability reported in cephalopods reduce the probability to detect differences between GGE under different conditions?

The aims of this work are to study the effect of temperature, sexual maturity and sex on the growth, food intake and gross growth efficiency of the octopus, Octopus tehuelchus, under laboratory conditions. Specifically, we wanted to investigate the effectiveness of the relationship between growth and food intake as an alternative method to identify the response of gross growth efficiency to temperature, sexual maturity and sex. Finally, to compare the results of this alternative method with those obtained using the classics statistics test.

MATERIALS AND METHODS

Experimental Conditions

Octopuses were collected from tidepools in the intertidal zone of Puerto Lobos (42[degrees]S, 46[degrees]06'W, San Matias gulf, Argentina). Immature specimens were collected in November (late spring) 2001 and mature specimens were collected in March (early autumn) 2001. In the laboratory, octopuses were acclimatized in glass aquaria for 14 days. Octopuses were classified as immature when their body weights were lower than 18 g for males and lower than 34 g for females. For each group of immature and mature individuals, 4 experimental conditions were established: (1) males and (2) females at a 10[degrees]C water temperature, (3) males and (4) females at a 15[degrees]C temperature. Every 10 days, food intake and growth were measured over a period of 50 days for mature octopuses and 30 days for immature ones. Note the high mortality and escapement resulted in two entire missing treatments (i.e., mature males and females held at 15[degrees]C). Therefore, 6 treatments were used instead of the 8 originally proposed. For each of the 6 treatments, 5 isolated octopuses per aquaria were used. Before the end of the immature period an unbalanced design was the result of 2 immature males dead kept at 15[degrees]C. Because of this set back, only 11 records of food intake and growth were available to compare with the other 5 treatments. In consequence, to perform the statistic tests that require a balanced design, 11 measurements pairs food intake-growth were randomly chosen from each of the others 5 treatments.

Over the whole trial the light regimen was of 8 h light and 16 h darkness, and octopuses were fed daily on a monospecific diet with live crab (Cyrtograpsus altimanus), which has a good acceptability in laboratory conditions (Re & Gomez Simes 1992). Each octopus was fed with one crab, which was chosen randomly with a minimum of 3% and a maximum of 17% of octopus's body weight. The effective percentage offered to each octopus was defined as a daily ration for that day. Any predated crabs were replaced with crabs of known wet weight. The weights of live crabs remaining and the unconsumed remains of predated crabs were recorded to establish the weight of tissue ingested.

Food Intake

The amount of food intake by each octopus at intervals of 10 days was calculated for immature and mature individuals over periods of 30 and 50 days respectively. The individual daily feeding rates were calculated following Choe (1966):

DFR = FI/(t * W) * 100

where FI is the total food intake (in grams) over time period (t = 10 days); W is the average of the weight measured at the beginning and end of each 10-day period. ANOVA was used to compare DFR between treatments. Because the two missing treatments (i.e., mature males and females held at 15[degrees]C) two analysis of variance tests were performed to evaluate the effects of temperature and sex on DFR in immature individuals and the effects of sexual maturity and sex on DFR in individuals kept at 10[degrees]C. In all cases the number of observations was 11 per treatment.

Growth Measurements

To monitor individual change in wet weight, all specimens were weighed at intervals of 10 days throughout the trial. Specimens were removed from their aquaria and weighed in a plastic beaker on a digital balance (Mettler, model PC 440) to the nearest centigram. The octopuses were not anaesthetized, and the period of handling during weighing was kept brief to minimize stress to the animal. Least-squares linear regression was used to calculate instantaneous growth rates (G), expressed as the percent increase in body mass per day (Forsythe & Van Heukelem 1987). G values were compared among treatments by a parallelism test (Zar 1996).

Growth Versus Food Intake

Growth and food intake (as a percentage of body weight) were calculated at 10 days intervals to study the relationship between growth and food intake under the experimental conditions. For each octopus, these values were calculated as follows:

FI(%) = (FI/[BW.sub.i]), 100,

where [BW.sub.i] is the octopus wet weight (in grams) at the beginning of each time interval, and:

[DELTA]BW(%) = ([DELTA]WB/[BW.sub.i]) * 100,

where [DELTA]BW is the difference between the weight at the beginning and end of each time period (t = 10 days).

Following Van Heukelem (1976) and Joll (1977), the relationship between [DELTA]BW(%) versus FI(%) was investigated using the linear model:

[DELTA]BW(%) = b * FI(%) + a (1)

where b is the slope and a is the intercept, which estimates the weight lost at starvation condition (i.e., when food intake is equal to zero). Note that [a.sub.i] will always be negative because this is an estimate of the weight lost at starvation condition. The slopes and intercepts were compared among treatments using parallelism and ANCOVA tests (Zar 1996) respectively.

For each treatment, the maintenance level (ML) was calculated from Eq. l (as a percentage of body weight) as:

ML(%) = -a/b (2)

This follows the recommendation of Wells and Clarke (1996). From ML(%) definition, Eq. (1) can be rewritten using ML(%) as a new parameter in the function:

[DELTA]BW(%) = b * [FI(%) - ML(%)]

Following the definition of Kleiber (1961), b is the partial growth efficiency (PGE) of conversion of the food remaining after the maintenance requirements have been satisfied, then:

PGE = [DELTA]BW(%)/[FI(%) - ML(%)]

Gross Growth Efficiency

The gross growth efficiency (GGE) was calculated following Choe (1966):

GGE = [[DELTA]BW/FI] * 100 (4)

where [DELTA]BW is the change in wet weight (in grams) and FI is the food intake (in grams) over each period of 10 days. The variance of GGE estimates were obtained using the delta method (Seber 1982), to include the variability caused by [DELTA]BW and FI estimation.

Using the relationship between [DELTA]BW(%) and FI(%), an alternative criterion can be defined that could allow comparing gross growth efficiencies calculated under two or more treatments. Let [T.sub.1] and [T.sub.2] be two treatments under the relationship between food intake and growth is estimated, then from Eq.(1):

[[DELTA]BW(%).sub.1] = [b.sub.1] * [FI(%).sub.1] - [ and [[DELTA]BW.sub.2](%) = b2 * [FI(%).sub.2] - [a.sub.2]

If [b.sub.1] = [b.sub.2] then, for an equal food intake value of FI(%) the change in wet weight [DELTA]BW(%) values under treatments [T.sub.1] and [T.sub.2] will depend on intercepts [a.sub.1] and [a.sub.2] because:

If

-[a.sub.1] > - [a.sub.2] then, [[DELTA]BW(%).sub.1] > [[DELTA]BW(%).sub.2],

therefore:

[GGE.sub.1] = [[DELTA]BW(%).sub.1]/FI(%)] * 100 > [GGE.sub.2] = [[DELTA]BW(%).sub.2]/FI(%)] * 100

Thus, a difference between [a.sub.1] and [a.sub.2] will indicate a difference between gross growth efficiencies calculated under treatments [T.sub.1] and [T.sub.2]. The gross growth efficiencies were compared among treatments by means of the criterion established earlier. The results obtained with this alternative method, were compared with those obtained using the classical statistical approach: (a) a direct comparison between intervals of confidence at 95% of GGE for immature octopuses kept at 10[degrees]C and 15[degrees]C, and (b) a t' Welch approximation test for disparate variances to compare GGE of immature and mature octopuses kept at 10[degrees]C.

Because a marked difference exits between the process of gain and loss weight, only positive values were used to investigate the effects of temperature, sex and sexual maturity on GGE. In fact, the effect of temperature, sex and sexual maturity on weight loss will be studied using the ANCOVA analysis.

RESULTS

Food Intake

Means of daily feeding rates (DFR) obtained for males and females, temperature levels and sexual maturity stages are shown in Table 1. Immature octopus of both sexes showed significantly higher DFR when held at a higher temperature value (15[degrees]C) than those held at a lower temperature value (10[degrees]C) (P = 0.0001), whereas no effect of sex was found (P = 0.1301) on DFR. Mature individuals at 10[degrees]C had a significantly lower daily feeding rate than immature octopuses held at the same temperature (P = 0.0312). There was no sex-based difference between DFR of mature and immature animals of the same maturation stages when kept at 10[degrees]C (P = 0.2457).

Growth

By means of adjusted R square criterion, for both immature individuals and mature females, the best fit to the observed growth over time was an exponential curve, whereas mature males showed a linear growth as a best fit (Fig. 1). The slopes of the regressions for all treatments were positive. When each slope, as the instantaneous growth rate (G), was tested for significance of difference from zero, the slopes for immature individuals under 10[degrees]C and 15[degrees]C, and mature females held at 10[degrees]C, were all significantly different from zero (P < 0.05), whereas the slope for mature males was not (P = 0.7780). Immature octopuses held at 15[degrees]C grew faster than those at 10[degrees]C (P = 0.0061). No differences were found between G values of immature individuals at 10[degrees]C and mature females kept at the same temperature (P = 0.1561). The values of instantaneous growth rates for immature individuals at 10[degrees]C and 15[degrees]C and mature octopuses at 10[degrees]C are showed in Table 2.

[FIGURE 1 OMITTED]

Growth Versus Food Intake

For all treatments the best fit for the relationship between growth and food intake was linear (Fig. 2 and Fig. 3) and Table 3. The slope b, as the partial growth efficiency (PGE) for each experimental condition was positive and significantly different from zero (Table 3). Parallelism tests indicated no difference between PGE of immature octopuses at 10[degrees]C and 15[degrees]C (P = 0.5972). For mature individuals, the parallelism test showed no difference between PGE of males and females kept at 10[degrees]C (P = 0.7618).

[FIGURES 2-3 OMITTED]

These results indicate that when maintenance requirements were satisfied, an equal portion from food remaining was invested in growth by immature octopus at 10[degrees]C and 15[degrees]C. The same situation was observed between mature males and females kept at 10[degrees]C. For immature individuals, weight lost (WL) as the percentage of body weight for an interval of 10 days was significantly higher in octopus kept at 15[degrees]C than those at 10[degrees]C (P = 0.0352). No difference was found between WL for males and females at mature stage kept at 10[degrees]C (P = 0.3075). The maintenance level (ML) in immature octopuses at 15[degrees]C was greater than those at 10[degrees]C and no effect of sex was observed on ML. There was no difference between ML of males and females at mature stage kept at 10[degrees]C. Results of simple regression analyses are showed in Table 3.

Gross Growth Efficiency

Gross growth efficiency (GGE) mean values indicated that immature octopuses of both sexes at 15[degrees]C invested 44% from total ingested food in growth, whereas 52% of ingested food was invested in growth at 10[degrees]C. When negative growth data were excluded from the analysis, GGE mean value for mature individuals of both sexes kept at 10[degrees]C was 35.5% of total food intake. A significant individual variation in GGE was observed for all analyzed treatments (Table 4). No difference between the variance of GGE for immature individuals at 10[degrees]C and 15[degrees]C was found (P > 0.05). However, GGE variance of mature octopuses kept at 10[degrees]C was significantly different from that of immature individuals calculated under the same temperature treatment (P < 0.001).

Comparison of GGE by Classical Statistical Methods

For immature octopus kept at 10[degrees]C and 15[degrees]C, direct comparison between confidence intervals at 95% indicated no significant difference in GGE mean values at different temperatures (Fig. 4). The t'Welch approximation test for disparate variances showed no differences in GGE mean values between immature and mature individuals kept both at 10[degrees]C (P = 0.0939).

[FIGURE 4 OMITTED]

Comparison of GGE Based on Growth Versus Food

Intake Relationship

These are the results obtained using the relationship between growth and food intake as an alternative method to investigate differences in GGE under different treatment. From intercepts analysis (ANCOVA) for immature octopuses kept at 10[degrees]C and 15[degrees]C, difference between intercepts of two temperature levels was found (P = 0.0326), whereas no significant effect of sex was observed (P = 0.3695). These results allow inferring in the first place that immature individuals at 10[degrees]C showed a higher GGE than those at 15[degrees]C. Secondly, under a fixed temperature males and females invested in growth the same portion of ingested food. For mature octopuses kept at 10[degrees]C, either males or females invest in growth-equivalent portions of food intake. From the intercept analysis a significant difference was detected between GGE of mature and immature octopuses kept at 10[degrees]C (P = 0.0020). This result indicates that at 10[degrees]C immature specimens were more effective in converting food in body growth than mature ones.

DISCUSSION

This study reported the effects of temperature, sexual maturity and sex on growth and food intake under laboratory conditions in the "Pulpito" Octopus tehuelchus. In addition, an alternative method to investigate the influence of different treatments on gross growth efficiency was developed.

Within the normal range temperature adaptation of a cephalopod species, higher temperatures lead to greater food intake, with exceptions such as sepiolids (v Boletzky 1975). In addition, if food is not limiting, temperature is a key factor influencing cephalopod growth rates, particularly during the early growth phase (Forsythe 1993). In this study, immature specimens of O. tehuelchus displayed higher daily feeding and instantaneous growth rates at 15[degrees]C than 10[degrees]C. These results are in accordance with previous experimental cephalopod studies (e.g., O. briareus; Robson 1929, Borer 1971, O. vulgaris; Mangold & Boletzky 1973 and Aguado & Garcia 2002, Eledone moschata; Lamarck 1798, Mangold & Boucher-Rodoni 1973, O. tetricus; Joll 1977, O. ocellatus Gray 1849, Segawa & Nomoto 2002.

A gradual daily feeding rate reduction appears to be the general rule for an octopus nearing sexual maturity (Borer 1971, Van Heukelem 1976, Wodinsky 1978). This is in agreement with the present findings for O. tehuelchus, where mature males and females showed lower daily feeding rates than immature specimens kept at the same temperature.

In immature specimens of O. tehuelchus the sex appears to have had no effect on feeding and growth rates; in neither 10[degrees]C nor 15[degrees]C temperature water. Mature males kept at 10[degrees]C showed a G value equal to zero whereas mature females showed a significant positive G value. This finding implies that in mature octopuses sex would has an effect on growth. Re (1989) found that in natural conditions, O. tehuelchus males reach sexual maturity three months earlier than females. Therefore, by the time those specimens were collected, males were already into an advanced sexual maturity state than females, who still were in their sexual maturation process. Moreover, although mature males and females showed no differences between daily feeding rates, only females had measurable positive growth. Likely, the lack of measurable growth in mature males of O. tehuelchus could be because of an effect of advanced age on growth process, which would allow a normal feeding activity with either low or none growth-tissue production. Similar results have been reported by Cortez et al. (1995) for O. mimus (Gould, 1852) collected from the Pacific waters of Chile.

For the present work, no statistically significant, sex-based differences were found among immature animals in terms of instantaneous growth rates or daily feeding rates. However, a slight tendency of instantaneous growth rates was observed to show slightly higher values in males at 10[degrees]C but higher in females at 15[degrees]C. Similarly, Forsythe and Hanlon (1988) observed higher growth in O. bimaculoides (Pickford/McConnaughey, 1949) males at 18[degrees]C but higher in females at 23[degrees]C. Considering that males of O. tehuelchus reach the sexual maturity approximately 3 mo earlier than females, this tendency in data could indicate an interaction between sex and temperature on instantaneous growth rates and daily feeding rates for immature specimens. Thus, in colder periods, immature males should show significantly higher daily feeding rates, instantaneous growth rates and gross growth efficiencies than females.

For all treatments performed in the present study, the relationship between growth and food intake was linear. Estimated values of body weight lost and maintenance level in immature octopuses at 15[degrees]C were higher than those at 10[degrees]C. In mature individuals kept at 10[degrees]C these values showed no differences between sexes. Immature individuals showed constant partial growth efficiency over sexes and temperature treatments. Because of the difference, experimental and biological conditions (e.g., temperature level and body weight) and comparisons among species become an arduous task. For individuals of O. tetricus kept at a mean of 20[degrees]C, a graphic estimation from the linear relationship between growth and food intake a maintenance level of 1% BW [day.sup.-1] and a daily body weight lost of 0.7% were obtained. The value of partial growth efficiency was estimated as 82.8% after the maintenance requirements have been satisfied (Joll 1977). For O. cyanea and O. maya reared at 25[degrees]C, the best fit to explain the relationship between growth and food intake was linear (Van Heukelem 1976, 1987). For both species the least square method indicated a maintenance level of 1.8% BW [day.sup.-1] and a body weight lost of 1.01% BW per day with a partial growth efficiency of 70% (Van Heukelem 1976, 1987). From present data and literature cited, an increase in temperature yielded higher maintenance level and daily body weight lost, whereas the linear form of the relationship between growth and food intake remained unalterable.

Gross growth efficiency (GGE) estimated for immature O. tehuelchus held at 10[degrees]C (GGE = 52%) are among the highest reported in the literature. Nixon (1966) observed a GGE value of 46% for O. vulgaris. Research on O. maya and O. cyanea held at 25[degrees]C produced GGE means values of 39.5% and 38.3% respectively, although for the estimated values of GGE mature octopuses data were not included (Van Heukelem 1976). Joll (1977) found that O. tetricus kept between 16.5[degrees]C and 24.2[degrees]C possess a GGE mean value of 46.8%. A lower GGE value of 37% was found for Eledone moschata held at 15[degrees]C (Mangold 1987). Daly and Peck (2000); recorded for three individuals of the Antarctic octopus Pareledone charcoti (Joubin, 1905) at 0[degrees]C under laboratory conditions a GGE mean value of 32%.

Gross growth efficiency declines when old males and maturing females of O. maya and O. cyanea decrease food intake (Van Heukelem 1976). Moreover, O'Dor and Wells (1978) pointed out that sexual maturation in females involves different needs for nutrients and energy, because females have higher reproductive costs for egg formation. When the t'Welch approximation test for disparate variances was used to compare GGE means between mature and immature specimens kept at 10[degrees]C, no significant difference was detected. Contrariwise, the alternative method based on the comparison between intercepts was able to detect a significant difference between GGE of immature and mature octopuses. Agreeing to Wells et al. (1983) and O'Dor et al. (1983) findings for O. vulgaris, O. tehuelchus showed in growth and food intake a high individual variability. Likely, this variation in GGE, which produces great variance value, can explain why t'Welch approximation test failed to detect differences between GGE. Furthermore, the low daily feeding rate and high maintenance level observed in mature specimens of O. tehuelchus could yield a lower GGE than immature octopuses kept at 10[degrees]C.

The confidence intervals at 95% comparison indicated no difference between GGE of immature specimens kept at 10[degrees]C and 15[degrees]C. Contrariwise, the alternative method based on the comparison between intercepts indicated that immature specimens held at a 10[degrees]C showed higher GGE than individuals kept at 15[degrees]C. Mangold and v Boletzky (1973) for individuals of O. vulgaris feed ad libitum and held at 10[degrees]C, 15[degrees]C and 20[degrees]C recorded gross growth efficiencies mean values of 56%, 55% and 48% respectively. Despite this, no statistics differences in GGE of O. vulgaris between temperature levels were reported. As it was observed in O. tehuelchus, O. vulgaris showed the highest GGE value at the lowest temperature treatment. Aguado Gimenez and Garcia Garcia (2002) by mean of a multiple regression analysis observed in O. vulgaris that GGE increased from 13[degrees]C to an optimal temperature value at 16.5[degrees]C, after this point GGE decreased to zero at 23[degrees]C. As it was observed in O. tehuelchus and O. vulgaris, for the normal range of temperature for a species, gross growth efficiency is affected by temperature. In addition, in O. tehuelchus the sexual maturity has an effect on GGE. The results obtained in this work indicate that, whereas classic statistical tests failed to detect the effects of temperature and sexual maturity on GGE, the alternative method based on the growth versus food intake relationship was able to do it. The information obtained about the effect of temperature and sexual maturity on growth, food intake and gross growth efficiency of O. tehuelchus under laboratory conditions could be used to develop potential rearing techniques.

ACKNOWLEDGMENTS

The authors thank Hernan Marani and Martin Serran for their help in the octopus sampling and P. Baron, F. Cremonte and R. Williams for their valuable comments on the earlier version of this manuscript.

LITERATURE CITED

Aguado Gimenez, F. & B. Garcia Garcia. 2002. Growth and food intake models in Octopus vulgaris Cuvier (1797): influence of body weight, temperature, sex and diet. Aquac. Int. 10:361-377.

Boletzky, S.v. 1975. Le developpment D' Eledone moschata (Mollusca, Cephalopoda) elevee au laboratoire. Bull. Soc. Zool. Fr. 100:361-367.

Borer, K. T. 1971. Control of food intake in Octopus briareus Robson. J. Comp. and Physiol. Psych. 75:171-185.

Choe, S. 1966. On the growth, feeding rates and the efficiency of food conversion for cuttlefishes and squids. Korean J. Zool. 9:12-20.

Cortez, T., B. G. Castro & A. Guerra. 1995. Feeding dynamics of Octopus mimus (Mollusca: Cephalopoda) in northern Chile waters. Marine Biology 123:497-503.

Daly, H. I. & L. S. Peck. 2000. Energy balance and cold adaptation in the octopus Pareledone charcoti. Exp. Mar. Biol. Ecol. 245:197-214.

Forsythe, J. W. 1993. A working hypothesis of how seasonal temperature change may impact the field growth of young cephalopods. In: T. Okutami, R. K. O'Dor, & T. Kubodera, editors. Recent advances in cephalopod fisheries biology. Tokyo, Japan: Tokai University Press. pp. 133-143.

Forsythe, J. W. & W. F. Van Heukelem. 1987. Growth. In: P. R. Boyle, editor. Cephalopod life cycles. Vol. II Comparative reviews. London, UK: Academic Press. pp. 135-156.

Forsythe, J. W. & R. T. Hanlon. 1988. Effect of temperature on laboratory growth, reproduction and life span of Octopus bimaculoides. Mar. Biol. 98:369-379.

Forsythe, J. W. & R. B. Toll. 1991. Clarification of the Western Atlantic Ocean pygmy octopus complex: the identity and life history of Octopus joubini (Cephalopoda: Octopodinae). Bull Mar. Sci. 49:88-97.

Joll, L. M. 1977. Growth and food intake of Octopus tetricus (Mollusca: Cephalopoda) in aquaria. Aust. J. Mar. Freshwater Res. 28:45-56.

Kleiber, M. 1961. The fire of life, an introduction to animal energetics, New York, USA.

Mangold, K. 1983. Food, feeding and growth in cephalopods. Mem. Nat. Mus. Victoria 4:81-93.

Mangold, K. 1987. Eledone moschata. In: P.R. Boyle, editor. Cephalopod life cycles. Vol. I Species account. London, UK: Academic Press. pp. 387-400.

Mangold, K. & S. v Boletzky. 1973. New data on reproductive biology and growth of Octopus vulgaris. Mar. Biol. 19:7-12.

Mangold, K. & R. Boucher-Rodoni. 1973. Nutrition et croissance de trois Octopodides mediterraneens, Etude preliminaire. Rapp. Comm. int. Mer Medit. 21:789-791.

Nixon, M. 1966. Changes in body weight and intake of food by Octopus vulgaris. J. Zool. 150:1-9.

O'Dor, R. K. & M. J. Wells. 1978. Reproduction versus somatic growth: hormonal control in Octopus vulgaris. J. Exp. Biol. 77:529-540.

O'Dor, R. K., K. Mangold, R. Boucher-Rodoni, M. J. Wells & J. Wells. 1983. Nutrient absorption, storage and remobilization in Octopus vulgaris. Mar. Behav. Physiol. 11:239-258.

O'Dor, R. K. & M. J. Wells. 1987. Energy and nutrient flow. In: P.R. Boyle, editor. Cephalopod life cycles. Vol. II Comparative reviews. London, UK: Academic Press. pp. 109-133.

Re, M. E. 1989. Estudios ecologicos sobre el crecimiento y la alimentacion de Octopus tehuelchus d'Orbigny en Puerto Lobos, golfo San Matias. Tesis Doctoral, Facultad de Ciencias Naturales y Museo de la Universidad Nacional de La Plata.

Re, M. E. 1998a. Pulpos octopodidos (Cephalopoda, Octopodidae). In: E. E. Boschi, editor. El Mar Argentino y sus Recursos Pesqueros. Tomo 2: Los moluscos de interes pesquero. Cultivos y estrategias reproductivas de bivalvos y equinoideos, INIDEP, pp. 69-98.

Re, M. E. 1998 b. Pesquerias de pulpos. In: Boschi, E.E. (Ed.) E1 Mar Argentino y sus Recursos Pesqueros. Tomo 2: Los moluscos de interds pesquero. Cultivos y estrategias reproductivas de bivalvos y equinoideos, INIDEP, pp. 99-114.

Re, M. E. & E. Gomez Simes, 1992. Habitos alimentarios del pulpo Octopus tehuelchus. I. Analisis cuali-cuantitativo de la dieta en el intermareal de Puerto Lobos, golfo San Matias (Argentina). Frente Maritimo 11, Sec. A: 119-128.

Seber, G. A. F. 1982. The estimation of animal abundance and related parameters, 2nd ed. Charles Griffin & Company Ltd. Segawa, S. & A. Nomoto. 2002. Laboratory growth, feeding, oxygen consumption and amonia excretion of Octopus ocellatus. Bull. Mar. Sci. 71:801-813.

Semmens, J. M., G. T. Pecl, R. Villanueva, D. Jouffre, I. Sobrino, J. B. Wood & P. R. Rigby. 2004. Understanding octopus growth: patterns, variability and physiology. Mar. Freshwater Res. 55:367-377.

Van Heukelem, W. F. 1976. Growth, bioenergetics and life-span of Octopus cyanea and Octopus maya. PhD. Thesis, University of Hawaii.

Van Heukelem, W. F. 1979. Environmental control of reproduction and life span in Octopus: An hypothesis. In: S. E. Stancyk, editor. Reproductive ecology of marine invertebrates. The Belle W. Baruch Library in Marine Science, No. 9. Columbia: University of South Carolina Press. pp. 123-133.

Van Heukelem, W. F. 1987. Octopus cyanea. In: P. R. Boyle, editor, Cephalopod life cycles. Vol. I Species account. London, UK: Academic Press. pp. 267-276.

Wells, M. J., R. K. O'Dor, K. Marigold & J. Wells. 1983. Diurnal changes in activity and metabolic rate in Octopus vulgaris. Mar. Behav. Physiol. 9:275-287.

Wells, M. J. & A. Clarke. 1996. The costs of living and reproducing for an individual cephalopod. Philos. Trans. R. Soc. Lond. B Biol. Sci. 351: 1083-1104.

Wodinsky, J. 1978. Feeding behaviour of broody female Octopus vulgaris. Anita. Behav. 26:803-813.

Zar, J. H. 1996. Biostatistical analysis, 3rd edition. New Jersey: Prentice Hall.

M. JAVIER KLAICH (1,2) *, MARIA E. RE (2) AND SUSANA N. PEDRAZA (1,2)

(1) Universidad Nacional de la Patagonia San Juan Bosco, Boulevard Brown 3700, Sede Puerto Madryn, Chubut, Argentina; (2) Centro Nacional Patagonico, CONICET, Boulevard Brown s/n, 9120 Puerto Madryn, Chubut Argentina

* Corresponding author. E-mail: [e-mail]

TABLE 1.

Daily feeding rate mean values obtained for mature Octopus
tehuelchus held at 10[degrees]C and immature specimens held at
10[degrees]C and 15[degrees]C.

Temperature
([degrees]C)         Immature              Mature

                               Both                    Both
              Males  Females   sexes  Males   Females  sexes
10            1.42   1.46      1.44   0.93    1.25     1.09
15            1.96   2.47      2.21   --      --
10 and 15     1.69   1.96      1.82*

For all treatments N = 11.

(*) = Overall mean; (--) = No value is presented because of both
scaping and high mortality in these treatments.

TABLE 2.
Instantaneous growth rates (G) estimated for mature O. tehuelchus
held at 10[degrees]C and immature specimens held at 10[degrees]C
and 15[degrees]C.

Temperature
([degrees]C)         Immature                Mature

                               Both                      Both
              Males  Females   sexes     Males  Females  sexes
10            0.78   0.53      0.67      0.05   0.60     0.32
15            1.05   1.28      1.19      --     --
10 and 15     0.97   0.89      0.93 (*)

(*) = Overall mean; (--) = No value is presented because of both
scaping and high mortality in these treatments.

TABLE 3.

Results of regression analyses for the relationship between growth
and food intake for each treatment (*) and pooled data by sex and
temperature (**) in O. tehuelchus.

                                                     ANOVA
                                                     (signifi-
                 b      a      a/b        [R.sup.2.  cance
Treatments       (PGE)  (WL)   (ML)   n   sub.adj]   level)
Immature
  M at 10
  [degrees]C (*)  0.65   1.28   1.95   14    0.788    p < 0.0001
  F at
  10[degrees]
  C (*)           0.79   6.13   7.74   17    0.436    p < 0.05
M and F at 10
    [degrees]C    0.75   4.28   5.70   31    0.513    p < 0.0001
    (**)
  M at
  15[degrees]C
  (*)             0.76   7.11   9.23   11    0.687    p < 0.01
  F at
  15[degrees]C
  (*)             1.05   14.70  14.01  12    0.650    p < 0.01
  M and F at 15
    [degrees]C    0.88   9.79   11.13  23    0.668    p < 0.01
    (**)
Mature
  M at
  10[degrees]C
  (*)             0.56   4.35   7.80   26    0.377    p < 0.001
  F at
  10[degrees]C
  (*)             0.60   4.43   7.42   14    0.755    p < 0.0001
  M and F at 10
    (**)
    [degrees]C   0.59   4.55   7.67   40    0.615    p < 0.0001

(M) = males; (F) = females; (PGE) = partial growth efficiency; (WL) =
weight lost and (ML) = maintenance level (as the percentages of body
weight) per 10 days.