EVIDENCE OF ITEROPARITY IN JUMBO SQUID DOSIDICUS GIG AS IN THE GULF OF CALIFORNIA, MEXICO.
The jumbo squid Dosidicus gigas (d'Orbigny, 1835) is the largest ommastrephid squid, reaching up to 1.2 m mantle length (ML), 65 kg in weight, and life span of 1.4 y (Rosa et al. 2013, Zepeda-Benitez et al. 2014). This species spawn in the oceanic environment, the eggs masses are spherical in shape with diameters ranged from 69 to 141 cm (Rosa et al. 2013, Birk et al. 2016). Hatching occurs between 6 and 9 days after fertilization at 18[degrees]C, the size at hatching varied between 0.9 and 1.3 mm ML (Yatsu et al. 1999). The jumbo squid passes through a post-hatching paralarval stage called the rhynchoteuthion. During this stage, the two tentacles are fused into a well-developed proboscis. No eye or intestinal photophores are present in squid paralarvae, until becoming juveniles between 12 and 140 mm ML (Rosa et al. 2013). The jumbo squid is an endemic species in the eastern Pacific Ocean (EPO) and is the most important cephalopod caught in Latin America, from Chile to Mexico, with Peruvian waters being the main fishing grounds (Tafur et al. 2010, Ibanez et al. 2015, Morales-Bojorquez & Pacheco-Bedoya 2016a). The population dynamics of D. gigas show variations in the age and ML structure, number of cohorts, and individual growth; thus, squids along a latitudinal gradient have different features. A general pattern for intraspecific structure based on ML distribution in the EPO was proposed by Nigmatullin et al. (2001), who classified the squids in three ML groups: small, medium, and large. However, findings of studies estimating the number of cohorts along the distribution range of D. gigas differ from the previously suggested ML structure, with evidence reported in Ecuadorian, Peruvian, and Mexican waters and results varying between one and up to six cohorts, with different ML structures in each region (Keyl et al. 2011, Velazquez-Abunader et al. 2012, Morales-Bojorquez & Pacheco-Bedoya 2016b). According to the latitudinal gradient, the age structure is also different, with older individuals having been found in Peru (354 days) and Mexico (450 days); in contrast, younger squid are commonly located in Costa Rica (289 days) (Arguelles et al. 2001. Chen et al. 2013, Zepeda-Benitez et al. 2014).
In general, several cephalopods have been considered as semelparous species (Arnold 1984, Rodhouse 1998), meaning that individuals in a population spawn only once during their lifetime (Boyle & Rodhouse 2005); this reproductive strategy has been improperly attributed to Dosidicus gigas without a rigorous histological analysis (Nigmatullin et al. 2001, Nigmatullin & Markaida 2009, Hoving et al. 2013). Evidence of spawning more than one batch of eggs in a lifetime has been found for several cephalopod species, such as Nautilus spp., Idiosepius pygmaeus. Sthenoteuthis oualaniensis, and Lolliguncula panamensis, indicating iteroparity as their reproductive strategy (Saunders 1984, Harman et al. 1989, Lewis & Choat 1993, Arizmendi-Rodriguez et al. 2012). By contrast, the semelparous genotype will have a greater fecundity in its single reproductive episode, devoting all of its physiological resources to reproduction and then dying (Cole 1954, Young 1981). Therefore, the semelparous species are genetically programmed with an irreversible degeneration subsequent to breeding (Crespi & Teo 2002). Other semelparous species, such as capelin and spider crabs, are capable of facultative iteroparity, and many cephalopods, although considered semelparous, exhibit lengthy postreproductive senescence, with some capable of a second bout of reproduction (Hughes & Simons 2014). Conversely, the iteroparous genotype must divide its physiological resources between reproduction, somatic growth, and maintenance; hence, this genotype has some probability of surviving to reproduce again (Young 1981). After the first reproduction has occurred in iteroparous organisms, it may be repeated at various time intervals (e.g., seasonal, semiannual, annual) (Cole 1954); although the females of iteroparous species may breed more than once, several factors, such as unfavorable environmental conditions affecting the physiological condition of the individuals, marine pollution, reduced food supply, and high population density, could influence oocyte reabsorption (Rideout & Tomkiewicz 2011, Sieiro et al. 2016).
Iteroparity has been also documented in Vampyroteuthis infernalis, the evidence of which was based on adult specimens where postovulatory follicles were present, indicating previous spawning (Hoving et al. 2015). For Graneledone boreopacifica and Loligo vulgaris reynaudii, the evidence was supported by histological and oocyte diameter-frequency analysis (Melo & Sauer 1999, Bello 2006). Similarly, a histological description showed prevalence of atresia in Octopus vulgaris, indicating asynchronic ovary development and synchronous ovulation during spawning (Sieiro et al. 2016). For Octopus chierchiae maintained in laboratory conditions, the iteroparity was evidenced when the females copulated three times while captive (Rodaniche 1984). Recent histological studies on Dosidicus gigas have documented the presence of postovulatory follicles and atresia at different developmental ovarian stages, indicating that females have multiple spawning events (Hernandez-Munoz et al. 2016). For Photololigo spp. and Idiosepius pygmaeus, histological analyses of ovaries and oocyte stage distribution showed that ovaries of mature females always included immature and mature oocytes, suggesting multiple spawning events (Lewis & Choat 1993, Moltschaniwskyj 1995). This situation illustrates the variability in reproductive strategies for cephalopods. Reproductive strategy may be considered as a set of tactics specifying how an organism responds to a particular environment to achieve reproductive success (McNamara & Houston 1996). Wootton (1984) defined these tactics as population variations in the typical reproductive pattern in response to environmental fluctuations; these tactics include regulation and compensation processes, for example, the number of times that an organism reproduces (Rochet 2000).
The apparent semelparity attributable to Dosidicus gigas has been based on indirect observations related to reproductive tactics; e.g., (1) length at first maturity occurs in old (approximately 1.5 y) and large (approximately 86 cm ML) squids, meaning that they are close to their maximum longevity; and (2) sudden increases in the gonadosomatic index (GSI) have been observed, which are caused by rapid growth of the ovaries relative to the growth of mantle tissue, as is observed in terminally spawning squid with synchronous ovulation (Rodhouse & Hatfield 1990, 1992, Nigmatullin & Markaida 2009). According to Rocha et al. (2001), there are five reproductive strategies in cephalopods; these depend on the environmental conditions and demographic pressures. They classified D. gigas as a species with a monocyclic spawning pattern, and egg-laying occurs in separate batches during the same spawning season, where ovary regeneration is not observed. By contrast, there are clues for partial spawning and asynchronous ovarian development, along with the presence of different cohorts in a population (Diaz-Uribe et al. 2006, Hernandez-Munoz et al. 2016). This evidence indicates that ovary regeneration may possibly occur in D. gigas; thus, the hypothesis in this article assumes that postovulatory follicles and previtellogenic oocytes coexist in the ovarian tissues of resting females. Consequently, egg-laying occurs in separate batches during different spawning seasons; thus, the reproductive strategy could be iteroparity. In consequence, the main goal of this study was to re-examine the reproductive strategy of D. gigas by analyzing the presence of postovulatory follicles associated with different ovarian stages and the variability in the ML structure.
MATERIALS AND METHODS
Collection of Samples and Histological Analysis
Fishery-dependent data on jumbo squid were collected fortnightly from March 2008 to November 2009 off the coast of Santa Rosalia, Baja California Sur, Mexico (Fig. 1). The biological samples were obtained from an artisanal fleet, which uses small boats with outboard motors locally known as "pangas"; the fishing gear used included hand-held lines baited with fluorescent jigs and a light system. Biological data and samples collected for each specimen included ML (cm), total weight (g), and ovary weight (g). The ovaries (n = 287) were fixed with Davidson solution (Howard & Smith 1983) and were analyzed using a standard histological analysis. Fixed tissue samples of approximately 1 [cm.sup.3] were obtained from each ovary and dehydrated in ethanol, cleared with Hemo-De, and embedded in paraffin (m.p. 56[degrees]C). Then, 4-[micro]m-thick tissue sections were obtained using a rotary microtome and were stained with hematoxylin and eosin. In addition, Masson's trichrome stain was used for visualizing the collagenous connective tissue fibers, which characterize the postovulatory follicles (Arizmendi-Rodriguez et al. 2012). The slides were digitized and analyzed with an Olympus BX50 optical microscope fitted with a Cool SNAP-Pro digital camera in conjunction with Image Pro Plus software (Media Cybernetics ver. 4.5). The developmental stage of the ovaries was based on criteria described by Diaz-Uribe et al. (2006) and Hernandez-Munoz et al. (2016).
Sex Ratio and Gonadosomatic Index
Sex ratios were estimated considering the temporal scale (months) and the size structure for each 2-cm ML class and compared using the Chi-square test ([chi square];[alpha] < 0.05, df = 1) (Zar 2010, Cerdenares-Ladron de Guevara et al. 2013). The monthly GSI was used for identifying the spawning season of jumbo squid as follows: GSI = [GW/TW] X 100, where GW is the weight of the ovary (g) and TW is total body weight (g). According to Arizmendi-Rodriguez et al. (2012), the monthly GSI was estimated per individual and compared with a monthly average for the study period.
The areas of at least 30 oocytes per female were measured to determine their theoretical diameter (TD). To account for changes in oocyte shape, which is common during development, the TD was standardized using the following equation proposed by Saout et al. (1999):
TD = [square root of 4A/[pi]], (1)
where A is the area of the oocytes and [pi] is a constant. One-way ANOVA ([alpha] = 0.05) was used to reveal statistically significant differences in oocyte TD for each ovarian stage. If the ANOVA showed statistically significant differences, Tukey's post hoc test was then used to identify differences among means values of TD for each oocyte substage ([alpha] = 0.05) (Zar 2010).
Distribution of Oocyte Diameters
Melo and Sauer (1999) suggested that to provide stronger conclusions about the reproductive strategy of cephalopods, it is necessary to analyze two sources of information: (1) data from histological examinations and (2) data on the size distribution of the oocytes. Thus, the diameter frequency distributions of whole oocytes (FDWO) at each ovarian stage were graphically represented as a frequency histogram. According to Cerdenares-Ladron de Guevara et al. (2013), to statistically determine the expected diameter at each ovarian stage, a multinomial distribution was used, and its probability density function (PDF) was expressed as described by Haddon (2001):
[mathematical expression not reproducible] (2)
where [x.sub.i] is the number of times an event of type i occurs in n samples, n is the sample size, and [p.sub.i] is the separate probability of each one of the type k events possible. To estimate the model parameters, it was necessary to transform the previous equation into the following negative log-likelihood (- ln [L.sub.1]) expression:
-ln [L.sub.1] ([x.sub.i]|n, [p.sub.1], [p.sub.2],..., [p.sub.k]) = [n.summation over i=1][[x.sub.i] ln([p.sub.i])] (3)
The main assumption for the parameter estimation is that the FDWO for each expected diameter can be analyzed with a normal distribution, assuming that each estimated mean represents a different oocyte substage. Under this condition, the relative expected proportions of each mean oocyte diameter were estimated using the following PDF:
[mathematical expression not reproducible] (4)
where [O.sub.i] is the observed oocyte diameter, [[mu].sub.i] is the estimated oocyte diameter, [[sigma].sub.i] represents the SD of oocyte diameter, and [[lambda].sub.i] is a penalty function to force the predicted number of observations of each mean oocyte diameter, thus stabilizing the solution during the optimization process. To estimate the expected frequencies and the model parameters, the estimated and observed values were compared with a multinomial distribution expressed as a negative log-likelihood (-ln [L.sub.2]) (Haddon 2001):
-ln [L.sub.2]([O.sub.i]|[[mu].sub.i], [[sigma].sub.i], [[lambda].sub.i]) = [n.summation over i=1][[rho].sub.i] ln([[[??].sub.i]/[SIGMA][[??].sub.i]]) + [n.summation over m=1][([N.sub.m] - [[??].sub.m]).sup.2], (5)
where [[rho].sub.i] is the observed frequency of oocytes i, [[??].sub.i] is the expected frequency of oocytes i, [N.sub.m] is the total frequency of oocytes observed for all ovarian stages, and [[??].sub.m] represents the total frequency of oocytes estimated for all ovarian stages (Aguirre-Villasenor et al. 2006). The model parameters were estimated when the negative log-likelihood function was minimized with a nonlinear fit using the generalized reduced gradient method (Lasdon et al. 1973). The confidence interval for each estimated diameter was calculated using a Student t distribution. To examine how many mean oocyte diameters were statistically significant, Akaike's information criterion (AIC) was used as follows: AIC = [2K + 2(-ln [L.sub.2])], where -ln [L.sub.2] is the negative log-likelihood function and K is the number of parameters. Thus, the number of expected diameters for each ovarian stage was calculated from the PDF, assuming that different numbers of oocyte diameters represent specific oocyte substages. Consequently, the number of expected diameters was defined when AIC reached its lowest value (Burnham & Anderson 2002, Montgomery et al. 2010).
Mantle Length Frequency Distributions
To estimate the number of ML groups, the ML distributions were analyzed annually using a class interval of 2 cm. The statistical procedure was similar to that used for analyzing the diameter FDWO, and consequently, the same structural equation of the normal PDF was used (Eq. 4). Thus, the PDF used an additive residual expressed as [([ML.sub.i] - [[mu].sub.ML]).sup.2], whereas the objective function was based on the residual sum of squares expressed as RSS = [n.summation over i=1][([ML.sub.i] - [[mu].sub.ML]).sup.2], where [ML.sub.i] is the mantle length (cm) observed for the individual i, and [[mu].sub.ML] is the mantle length (cm) estimated for the individual i. Finally, the number of ML groups was also selected based on AIC using the estimator modified for RSS (Burnham & Anderson 2002). This criterion was applied by increasing the number of ML groups from one to four for the annual ML (cm) frequency data (Morales-Bojorquez & Pacheco-Bedoya 2016b).
The histological analysis showed five ovarian stages and nine oocyte substages, as described in the following paragraphs. Stage I (previtellogenesis) included germinal cells identified as oogonia (Og) and oocytes identified as early previtellogenic oocytes (Pv1), intermediate previtellogenic oocytes (Pv2), and late previtellogenic oocytes (Pv3). Stage II (vitellogenesis) was characterized by the presence of early vitellogenic oocytes (Vol). Stage III (postvitellogenesis) included late vitellogenic oocytes (Vo2) and postvitellogenic oocytes (Pos). These stages were defined based on the most advanced group of oocytes observed in the biological sample. However, atretic oocytes (a) and postovulatory follicles (Pf) were also observed, denoting the presence of ovarian stages related to spawning (stage IV) and postspawning (stage V). Figure 2 shows ovary sections of jumbo squid representing the different stages of gonad development, whereas descriptions for each oocyte substage, including TDs, are shown in Table 1.
Statistical comparison of oocyte diameters showed that there were significant differences among the oocyte substages [ANOVA, F(6, 17,445) = 7,199, P = 0.001]. Tukey's post hoc test showed significant differences among all TDs within the ovarian stages (P < 0.001). In ovarian stage III, the Pos had relatively deep invaginations of follicular cells (fcs), and consequently, a high variability in their TDs was observed. Nonetheless, the differences in the TDs from oogonia to Pos represent the growth of oocytes throughout the maturation process (Fig. 3).
The quantitative analysis using the multinomial distribution applied to the TDs for each ovarian stage revealed an asynchrony in the development of oocytes. For ovarian stage I, only one mode was estimated, with a value of 95.9 [micro]m; for ovarian stage II, three modes were estimated, indicating the presence of different expected diameters that varied from 50.9 to 230.7 [micro]m; ovarian stage III showed three modes, with estimated diameters from 95.8 to 365.8 [micro]m. Ovarian stage IV was characterized by two modes, with estimated diameters of 95.9 and 186.2 [micro]m. Finally, ovarian stage V also had two modes, with diameters of 140.8 and 276.0 [micro]m. The multinomial analysis showed evidence of a progressive increase in oocyte diameter from ovarian stages I to II and from II to III. Although there was an observed increase in oocyte diameter for ovarian stage IV, the quantitative estimation did not show a modal value given the low frequency of oocytes, with diameters varying between 366 and 456 [micro]m. A similar pattern was observed for ovarian stage V, where larger oocyte diameters between 456 and 636 [micro]m were measured.
Using the most advanced group of oocytes observed in the gonad as the criterion, the next ovarian stages were defined: ovarian stage I showed only one kind of oocyte, Og. Ovarian stage II showed four oocyte substages, including Og, Pv1, Pv2, and Pv3. Oocyte development from ovarian stage I to II showed the permanence of Og at both ovarian stages, but Og had a low frequency in ovarian stage II. Ovarian stage III showed oocyte development in early vitellogenesis, indicating maturation from previtellogenesis to vitellogenesis, with the presence of Pv3 and Vol. Ovarian stage IV simultaneously showed oocytes associated with vitellogenesis, including Og and Pvl. From ovarian stage IV to V, the presence of new oocyte substages were observed, mainly Pos, denoting the postvitellogenic stage; however, Og and Pv3 were also observed (Tables 2 and 3).
The proportions of ovarian stages along the time series showed that approximately 91% of the jumbo squid females were undergoing previtellogenesis (stage I). Vitellogenesis (stage II) was observed over two periods, during May-July, with the highest proportion of 4.5% in May, and during November, peaking at 20%. Postvitellogenesis (stage III) was observed over a single period, from June to August, with the highest proportion in June (7.7%). Squid showing signs of spawning (stage IV) were observed during May-June, with the highest proportion during June (3.8%). Finally, the post-spawning (stage V) was identified during April, June, and August-October, with the highest proportion in September (20%). By contrast, during March, only immature females were observed. The GSI estimates showed two peaks in the time series, the first during June (0.4) and the second during September (0.5), indicating that the GSI has higher values and a more extended period during summer months than during other times of the year (Fig. 4). Table 4 shows the predominance of females in the jumbo squid population as determined from the sex ratio. Significant differences were found during May ([chi square] = 5.4, df = 1, P = 0.02), June ([chi square] = 4.1, df = 1, P = 0.04), July ([chi square] = 7.3, df = 1, P = 0.01), and August ([chi square] = 11.1, df = 1, P = 0.001). During March-April and September-November, the sex ratio was 1:1.
The histological analysis showed the presence of postovulatory follicles in jumbo squid throughout the ML structure, including females with MLs from 32 to 82 cm. The highest frequencies of postovulatory follicles were found in ML classes at 38-40, 44-48, and 60 cm; conversely, the lowest frequencies of postovulatory follicles were found in ML classes at 32, 52, 76, and 78 cm. This indicates that younger (32 cm ML) and older (82 cm ML) jumbo squid females spawned simultaneously (Fig. 5). The presence of postovulatory follicles at ovarian stage I was also observed in jumbo squid throughout the ML structure for females with MLs ranging from 32 to 70 cm (Fig. 5). Postovulatory follicles were also found at ovarian stage II, although they had low frequency in ML classes at 38, 42, and 76 cm (Fig. 5). Similar results were observed for ovarian stage III, where postovulatory follicles were found at MLs ranging from 38 to 44 cm (Fig. 5). A low frequency of postovulatory follicles was observed for ovarian stages IV and V, including females with MLs ranging from 36 to 82 cm (Fig. 5). The presence of postovulatory follicles in a resting female jumbo squid was demonstrated using a Masson's trichrome stain. The histological evidence showed that the ovaries simultaneously have postovulatory follicles and previtellogenic oocytes, mainly characterized by early and intermediate previtellogenic oocytes (Fig. 6A-F), meaning that ovary regeneration occurs in Dosidicus gigas. This finding was consistent for females from 30 to 80 cm ML, as shown in Figure 5.
Assuming that the number of cohorts in the jumbo squid population can be represented by the quantity of ML groups, the multinomial density function applied to the ML frequency data showed that during 2008, four cohorts were estimated (with mean MLs of 41.2, 48.5, 55.2, and 68.5 cm); in this year, the larger individuals (ML between 66 and 70 cm) showed a very well-defined mode, suggesting the existence of older individuals in the population, although individuals with an ML between 48 and 58 cm were the most frequent (Fig. 7A). Conversely, during 2009, only three ML groups were observed (mean MLs 39.9, 47.8, and 53.9 cm). In this year, smaller individuals were more abundant, and the larger squid varied in ML from 60 to 82 cm, with low frequencies of individuals grouped into this range of ML classes; therefore, this cohort was not very well defined, and its variance was the highest estimated for both years (Fig. 7B, Tables 5 and 6).
The histological analysis showed the presence of postovulatory follicles, which were found at all ovarian stages, with stage I showing the greatest frequency. This occurs because these females have previously spawned, and as a consequence, they have begun oocyte regeneration for the next spawning season, highlighting the presence of both oocyte substages, mainly early previtellogenic oocytes and postovulatory follicles (Brown-Peterson et al. 2011). This pattern was also reported by Hernandez-Munoz et al. (2016); however, they found the highest presence of postovulatory follicles in stage V. The results showed that postovulatory follicles were also found in individuals throughout the ML structure, including small (32 cm) and large females (82 cm). In this study, atretic oocytes were also identified; however, these are not necessarily characteristic of a spawning event because this substage can be influenced by environmental factors (e.g., temperature, photo-period) or simply represent the reabsorption of a batch of oocytes to maintain energy reserves (Melo & Sauer 1999, Valdebenito et al. 2011). According to Tyler and Sumpter (1996), Lowerre-Barbieri et al. (2011), and Sieiro et al. (2016), the presence of atretic oocytes is not suggestive of the end of the spawning season unless extensive atresia can be quantified, which was not observed in Dosidicus gigas. By contrast, postovulatory follicles are better indicators of spawning events.
The cellular dynamics of oocyte growth described previously show the asynchronous development among groups of oocytes, characterized by the simultaneous presence of seven types of oocytes at different substages of growth. In addition, the multinomial analysis applied to the TD of the oocytes identified synchronous development within groups; thus, ovarian stage I showed a predominance of oogonia, which were recruited to the next ovarian stage depending on how quickly they increased their diameter. As a consequence, ovarian stage II showed the presence of oocytes characteristic of early and intermediate previtellogenesis, along with oogonia remaining in the ovary. This maturation process of oocytes continued in ovarian stage III, with observations of late previtellogenic oocytes, including oogonia and intermediate previtellogenic oocytes.
Based on oocyte diameter-frequency analysis, a similar reproductive strategy was reported for chokka squid (Loligo vulgaris reynaudii), where mature ovaries containing late vitellogenic oocytes were found along with a batch of previtellogenic oocytes (Melo & Sauer 1999). Jointly, the oocyte diameter-frequency analysis also showed asynchronous oocyte development, and histological evidence of spent ovaries containing postovulatory follicles alongside oocytes in various stages of vitellogenesis confirms that Dosidicus gigas spawns multiple times (Rocha et al. 2001), which was also reported for veined squid, Loligo forbesi (Boyle et al. 1995, Collins et al. 1995). If the reproductive strategy of D. gigas was based on semelparity, such as was reported by Nigmatullin et al. (2001), Nigmatullin and Markaida (2009), and Hoving et al. (2013), then the oocyte diameter-frequency analysis would have shown all oocytes growing as a single batch up to a maximum diameter, indicating synchronous oocyte development, as observed in cephalopods with this reproductive strategy, such as the greater hooked squid (Onykia ingens; previously known as Moroteuthis ingens), Antarctic squid (Gonatus antarcticus), and common octopus (Octopus vulgaris) (Rodriguez-Rua et al. 2005, Laptikhovsky et al. 2007). Usually, the oocyte diameter-frequency distribution analysis for semelparous species exhibits a single bell curve, showing only one modal value. By contrast, the multinomial analysis applied to the frequency distribution of oocyte diameter for D. gigas showed a multimodal distribution, demonstrating asynchronous oocyte development, which is common for iteroparous species (Murua & Saborido-Rey 2003).
In this study, four and three cohorts of Dosidicus gigas were found during 2008 and 2009, respectively. For both years, the cohorts were characterized by individuals with MLs less than 56 cm, although during 2008, a cohort with a mean ML of 68.8 cm was also identified. This variability in the number of cohorts of jumbo squid in the Gulf of California has changed between one and three, and the population is frequently composed of only two cohorts (Velazquez-Abunader et al. 2012); however, during 1980 to 1981, five cohorts were clearly distinguishable (Ehrhardt et al. 1983). Sudden interannual changes between the number of cohorts has also been observed in the Humboldt Current, where fluctuations between one and six cohorts have been reported (Keyl et al. 2011), and at least three cohorts have been documented along the EPO in different marine and coastal zones, mainly along Costa Rica, Ecuador, Peru, and Chile (Arguelles et al. 2001, Chen et al. 2013, Ibanez et al. 2015, Morales-Bojorquez & Pacheco-Bedoya, 2016b). In the Gulf of California, the presence of multiple cohorts in the squid population is a consequence of several spawning events. From 1995, the small females spawned have been observed every year in the fishery-dependent data (Hernandez-Herrera et al. 1998), indicating that each cohort begins its maturity at small sizes (approximately 20 cm ML and 5 mo) and continues spawning during the respective spawning season along its life span, reaching approximately 85 cm ML and 1.4 y (Zepeda-Benitez et al. 2014). In addition, the abundance of females in the population was greater than that of males, which is a populational feature commonly reported for D. gigas (Morales-Bojorquez & Pacheco-Bedoya 2016b). A similar sex ratio pattern was reported by Nigmatullin et al. (2001) in the Humboldt Current, suggesting that reproduction by D. gigas occurs year-round, with peaks in spring and summer. The results in this study indicated that spawned squid females are mainly observed during May-June, and squid females in postspawning stage were identified during April-June and August-October. The spawning period of April-June was previously reported by Hernandez-Herrera et al. (1998) through identification of ovarian stages; similarly, during 2006 to 2007 and 2015, egg masses of D. gigas in the Gulf of California were identified (Staaf et al. 2008, Birk et al. 2016). However, the second spawning period evidenced by the histological analysis has not been confirmed by egg masses found in the field. Thus, D. gigas in the Gulf of California spawns two times, generating different numbers of cohorts.
Traditionally, semelparity is defined by a single, highly fecund bout of reproduction and death of the individual; in contrast, iteroparity is defined by repeated bouts of reproduction throughout life (Cole 1954, Charnov & Schaffer 1973, Young 1981, Crespi & Teo 2002). However, given the extreme plasticity of parity in some organisms, this classification does not reflect biological reality. Hughes and Simons (2014) and Hughes (2017) explained that parity should be understood as the distribution of reproductive effort through time and should, therefore, be treated as a continuous trait rather than a discrete one; thus, semelparity and iteroparity are the endpoints of this continuum trait. Rocha et al. (2001) suggested changing the terms semelparity to spawning once and iteroparity to spawning more than once to avoid confusion. In this context, they classified Dosidicus gigas as a multiple spawner. Hernandez-Munoz et al. (2016) also reported that jumbo squid spawn more than once as suggested by the presence of postovulatory follicles in female D. gigas. However, they did not make specific conclusions about the reproductive strategy, as their findings were based on 73 ovaries collected over a broad regional scale (Gulf of California and the western region of the Baja California Peninsula) and over a limited temporal scale (samples obtained during October 2000, September 2003, February 2012, and August 2012). By contrast, in this study, the regional scale was limited to the main distribution area of D. gigas in the Gulf of California, and the biological samples were collected fortnightly from March 2008 to November 2009. Under these circumstances, the analysis showed evidence of the asynchronous development of oocytes in the ovaries of D. gigas, with oocyte diameters ranging from 4.5 [micro]m for oogonia to 619 [micro]m for Pos, representing smaller oocytes than those reported by Hernandez-Munoz et al. (2016), which ranged from 11.7 [micro]m for oogonia to 656.7 [micro]m for Pos. The pattern of ovarian development was characterized by a mixture of oocyte sub-stages, which was highlighted by the presence of dominant oocyte substages, as indicated by the presence of multiple cohorts of oocytes where they were well recognized according to their average diameters and histological features. These histological and statistical procedures have been commonly used to redefine the reproductive strategies for several erroneously classified cephalopod species (Table 7). Nesis (1996), Nigmatullin et al. (2001), Rocha et al. (2001), Nigmatullin and Markaida (2009), and Hoving et al. (2013) attributed semelparity to D. gigas without histological evidence, taking as the criterion the general reproductive pattern observed for members of the Ommastrephidae family. The first histological description of oocyte development for D. gigas was realized by Diaz-Uribe et al. (2006). Later, Hernandez-Munoz et al. (2016) found postovulatory follicles as evidence of previous spawning events. The postovulatory follicles occur in ovaries after spawning and represent remaining fcs. In resting females of D. gigas, postovulatory follicles were simultaneously found with previtellogenic oocytes, denoting ovarian recovery. The hypothesis proposed in this study was confirmed with strong evidence of postovulatory follicles at all ovarian stages, highlighting their presence in resting females, and in individuals throughout the ML structure. Moreover, oocyte diameter-frequency analysis and the presence of multiple cohorts in the population studied here indicate that this species has an iteroparous reproductive strategy.
We thank the Consejo Nacional de Ciencia y Tecnologia Mexico (CONACYT) for the financial support (project contract number CB-2012-01-179322). X. A. P.-P. was a recipient of a CONACYT postgraduate fellowship (contract number 290124). A. H.-H. thanks the Instituto Politecnico Nacional fellowships (EDI and COFAA), and Eulalia Meza Chavez for her support with histological analysis.
Aguirre-Villasenor, H., E. Morales-Bojorquez, R. E. Moran-Angulo, J. Madrid-Vera & M. C. Valdez-Pineda. 2006. Biological indicators for the Pacific Sierra (Scomberomorus sierra) fishery in the southern Gulf of California, Mexico. Cienc. Mar. 32:471-484.
Arguelles, J., P. G. Rodhouse, P. Villegas & G. Castillo. 2001. Age, growth and population structure of the jumbo flying squid Dosidicus gigas in Peruvian waters. Fish. Res. 54:51-61.
Arizmendi-Rodriguez, D. I., C. Rodriguez-Jaramillo, C. Quinonez-Velazquez & C. A. Salinas-Zavala. 2012. Reproductive indicators and gonad development of the Panama brief squid Lolliguncula panamensis (Berry, 1911) in the Gulf of California. Mexico. J. Shellfish Res. 31:817-826.
Arnold, J. M. 1984. Cephalopods. In: Tompa, A. S., N. H. Verdonk & J. A. M. van den Biggelar, editors. The mollusca, vol. 7. Reproduction. New York, NY: Academic Press, pp. 419-454.
Bello, G. 2006. Signs of multiple spawning in Graneledone pacifica (Cephalopoda: Octopodidae). J. Mar. Biol. Assoc. U.K. 86:1183-1186.
Birk, M. A., C. Paight & B. A. Seibel. 2016. Observations of multiple pelagic egg masses from small-sized jumbo squid (Dosidicus gigas) in the Gulf of California. J. Nat. Hist. 51:2569-2584.
Boyle, P., G. J. Pierce & L. C. Hastie. 1995. Flexible reproductive strategies in the squid Loligo forbesi. Mar. Biol. 121:501-508.
Boyle, P. & P. Rodhouse. 2005. Cephalopods: ecology and fisheries. Oxford, UK: Wiley-Blackwell. 452 pp.
Brown-Peterson, N. J., D. M. Wyanski, F. Saborido-Rey, B. J. Macewicz & S. K. Lowerre-Barbieri. 2011. A standardized terminology for describing reproductive development in fishes. Mar. Coast. Fish. 3:52-70.
Burnham, K. P. & D. R. Anderson. 2002. Model selection and multi-model inference: a practical information theoretic approach. New York, NY: Springer. 490 pp.
Cerdenares-Ladron de Guevara, G., E. Morales-Bojorquez, C. Rodriguez-Jaramillo, A. Hernandez-Herrera & A. Abitia-Cardenas. 2013. Seasonal reproduction of sailfish Istiophorusplatypterus from the southeast Mexican Pacific. Mar. Biol. Res. 9:407-420.
Charnov, E. L. & W. M. Schaffer. 1973. Life-history consequences of natural selection: Cole's result revisited. Am. Nat. 107:791-793.
Chen, X., J. Li. B. Liu, Y. Chen, G. Li, Z. Fang & S. Tian. 2013. Age, growth and population structure of jumbo flying squid, Dosidicus gigas, off the Costa Rica Dome. J. Mar. Biol. Assoc. U.K. 93: 567-573.
Cole, L. C. 1954. The population consequences of life history phenomena. Q. Rev. Biol. 29:103-137.
Collins, M. A., G. M. Burnell & P. G. Rodhouse. 1995. Reproductive strategies of male and female Loligo forbesi (Cephalopoda: Loliginidae). J. Mar. Biol. Assoc. U.K. 75:621-634.
Crespi, B. J. & R. Teo. 2002. Comparative phylogenetic analysis of the evolution of semelparity and life history in salmonid fishes. Evolution 56:1008-1020.
Diaz-Uribe, J. G, A. Hernandez-Herrera, E. Morales-Bojorquez, S. Martinez-Aguilar, M. Suarez-Higuera & A. Hernandez-Lopez. 2006. Histological validation of the gonadal maturation stages of female jumbo squid (Dosidicus gigas) in the Gulf of California, Mexico. Cienc. Mar. 32:23-31.
Ehrhardt, N. M., P. S. Jacquemin, F. Garcia, G. Gonzalez, J. M. Lopez, J. Ortiz & A. Solis. 1983. On the fishery and biology of the giant squid Dosidicus gigas in the Gulf of California, Mexico. In: J. F. Caddy, editor. Advances in assessment of world cephalopod resources. Roma, Italy: FAO Fish. Tech. Paper. 231. pp. 306-340.
Haddon, M. 2001. Modelling and quantitative methods in fisheries. Boca Raton, FL: Chapman and Hall. 449 pp.
Harman, R. F., R. E. Young, S. B. Reid, K. M. Mangold, T. Suzuki & R. F. Hixon. 1989. Evidence for multiple spawning in the tropical oceanic squid Stenoteuthis oualaniensis (Teuthoidea: Ommastrephidae). Mar. Biol. 101:513-519.
Hernandez-Herrera, A., E. Morales-Bojorquez, M. A. Cisneros-Mata, M. O. Nevarez-Martinez & G. I. Rivera-Parra. 1998. Management strategy for the giant squid (Dosidicus gigas) fishery in the Gulf of California, Mexico. Calif. Coop. Ocean. Fish. Inveslig. Rep. 39:212-218.
Hernandez-Munoz, A. T., C. Rodriguez-Jaramillo, A. Mejia-Rebollo & C. A. Salinas-Zavala. 2016. Reproductive strategy in jumbo squid Dosidicus gigas (D'Orbigny, 1835): a new perspective. Fish. Res. 173:145-150.
Hoving. H. J. T., W. F. Gilly, U. Markaida, K. J. Benoit-Bird, Z. W. Brown, P. Daniel, J. C. Field, L. Parassenti, B. Liu & B. Campos. 2013. Extreme plasticity in life-history strategy allows a migratory predator (jumbo squid) to cope with a changing climate. Glob. Chang. Biol. 19:2089-2103.
Hoving, H. J. T.. V. V. Laptikhovsky & B. H. Robison. 2015. Vampire squid reproductive strategy is unique among coleoid cephalopods. Curr. Biol. 25:R322-R323.
Howard, D. W. & C. S. Smith. 1983. Histological techniques for marine bivalve mollusks. In: Natl. Ocean. Atmos. Adm. Tech. Memo. NMFS-F/NEC-25. pp. 35-38.
Hughes, P. W. 2017. Between semelparity and iteroparity: empirical evidence for a continuum of modes of parity. Ecol. Evol. 7:8232-8261.
Hughes, P. W. & A. M. Simons. 2014. The continuum between semelparity and iteroparity: plastic expression of parity in response to season length manipulation in Lobelia inflata. BMC Evol. Biol. 14:1-11.
Ibanez, C. M., R. D. Sepulveda, P. Ulloa, F. Keyl & M. C. Pardo-Gandarillas. 2015. The biology and ecology of the jumbo squid Dosidicus gigas (Cephalopoda) in Chilean waters: a review. Lat. Am. J. Aquat. Res. 43:402-414.
Keyl, F., J. Arguelles & R. Tafur. 2011. Interannual variability in size structure, age, and growth of jumbo squid (Dosidicus gigas) assessed by modal progression analysis. ICES J. Mar. Sci. 68:507-518.
Laptikhovsky, V. V., A. 1. Arkhipkin & H. J. T. Hoving. 2007. Reproductive biology in two species of deep-sea squids. Mar. Biol. 152:981-990.
Laptikhovsky, V., M. A. Collins & A. Arkhipkin. 2013. First case of possible iteroparity among coleoid cephalopods: the giant warty squid Kondakovia longimana. J. Motluscan Stud. 79:270-272.
Lasdon, L. S., R. L. Fox& M. W. Ratner. 1973. Nonlinear optimization using the generalized reduced gradient method. Rev. Fr. Autom. Inf. Rech. Oper. 8:73-103.
Lewis, A. R. & H. Choat. 1993. Reproductive output of the tropical cephalopod ldiosepiuspygmaeus. Can. J. Fish. Aquat. Sci. 50:20-28.
Lowerre-Barbieri, S. K., N. J. Brown-Peterson, H. Murua, J. Tomkiewicz, D. M. Wyanski & F. Saborido-Rey. 2011. Emerging issues and methodological advances in fisheries reproductive biology. Mar. Coast. Fish. 3:32-51.
McNamara, J. M. & A. I. Houston. 1996. State-dependent life histories. Nature 380:215.
Melo, Y. C. & W. H. H. Sauer. 1999. Confirmation of serial spawning in the chokka squid Loligo vulgaris reynaudii off coast of South Africa. Mar. Biol. 135:307-313.
Moltschaniwskyj, N. A. 1995. Multiple spawning in the tropical squid Photololigo sp.: what is the cost in somatic growth? Mar. Biol. 124:127-135.
Montgomery, S. S., C. T. Walsh, M. Haddon, C. L. Kesby & D. D. Johnson. 2010. Using length data in the Schnute Model to describe growth in a metapenaeid from waters off Australia. Mar. Freshw. Res. 61:1435-1445.
Morales-Bojorquez, E. & J. L. Pacheco-Bedoya. 2016a. Jumbo squid Dosidicus gigas: a new fishery in Ecuador. Rev. Fish. Sci. Aquae. 24:98-110.
Morales-Bojorquez, E. & J. L. Pacheco-Bedoya. 2016b. Population dynamics of jumbo squid Dosidicus gigas in Pacific Ecuadorian waters. J Shellfish Res. 35:211-224.
Murua, H. & F. Saborido-Rey. 2003. Female reproductive strategies of marine fish species of the North Atlantic. J. Northwest All. Fish. Sci. 33:23-31.
Nesis, K. N. 1996. Mating, spawning, and death in oceanic cephalopods: a review. Ruthenica 6:23-64.
Nigmatullin, C. M. & U. Markaida. 2009. Oocyte development, fecundity and spawning strategy of large sized jumbo squid Dosidicus gigas (Oegopsida: Ommastrephinae). J. Mar. Biol. Assoc. U.K. 89:789.
Nigmatullin, C. M., K. N. Nesis & A. I. Arkhipkin. 2001. A review of the biology of the jumbo squid Dosidicus gigas (Cephalopoda: Ommastrephidae). Fish. Res. 54:9-19.
Rideout. R. M. & J. Tomkiewicz. 2011. Skipped spawning in fishes: more common than you might think. Mar. Coast. Fish. 3:176-189.
Rocha, F., A. Guerra & A. F. Gonzalez. 2001. A review of reproductive strategies in cephalopods. Biol. Rev. 76:291-304.
Rochet, M. J. 2000. A comparative approach to life-history strategies and tactics among four orders of teleost fish. ICES J. Mar. Sci. 57:228-239.
Rodaniche. A. F. 1984. Iteroparity in the lesser Pacific striped octopus Octopus chierchiae (Jatta, 1889). Bull. Mar. Sci. 35:99-104.
Rodhouse, P. G. 1998. Physiological progenesis in cephalpod molluscs. Biol. Bull. 195:17-20.
Rodhouse, P. G. & E. M. C. Hatfield. 1990. Dynamics of growth and maturation in the cephalopod Illex argentinus de Castellanos, 1960 (Teuthoidea: Ommastrephidae). Philos. Trans. R. Soc. B Biol. Sci. 329:229-241.
Rodhouse, P. G. & E. M. C. Hatfield. 1992. Production of soma and gonad in maturing male Illex argentinus (Mollusca: Cephalopoda). J. Mar. Biol. Assoc. U.K. 72:293-300.
Rodriguez-Rua, A., I. Pozuelo, M. A. Prado, M. J. Gomez & M. A. Bruzon. 2005. The gametogenic cycle of Octopus vulgaris (Mollusca: Cephalopoda) as observed on the Atlantic coast of Andalusia (south of Spain). Mar. Biol. 147:927-933.
Rosa, R., C. Yamashiro, U. Markaida, P. G. K. Rodhouse, C. M. Waluda, C. A. Salinas-Zavala, F. Keyl, R. O'Dor, J. S. Stewart & W. F. Gilly. 2013. Dosidicus gigas, Humboldt squid. In: Rosa, R., G. J. Pierce & R. O'Dor, editors. Advances in squid biology, ecology and fisheries. Part II-Oegopsid squids. New York, NY: Nova Science Publishers, Inc. pp. 169-206.
Saout. C, C. Quere, A. Donval, Y. M. Paulet & J. F. Samain. 1999. An experimental study of the combined effects of temperature and photoperiod on reproductive physiology of Pecten maximus from the Bay of Brest (France). Aquaculture 172:301-314.
Saunders, W. B. 1984. Nautilus growth and longevity: evidence from marked and recaptured animals. Science 224:990-992.
Sieiro, P., J. Otero & A. Guerra. 2016. Histomorphological study of ovarian atresia over the reproductive cycle of Octopus vulgaris from Galician waters (NW Spain). Zoomorphology 135:419-431.
Staaf, D. J., S. Camarillo-Coop, S. H. D. Haddock, A. C. Nyack, J. Payne, C. A. Salinas-Zavala, B. A. Seibel, L. Trueblood, C. Widmer & W. F. Gilly. 2008. Natural egg mass deposition by the Humboldt squid (Dosidicus gigas) in the Gulf of California and characteristics of hatchlings and paralarvae. J. Mar. Biol. Assoc. U.K. 88:759-770.
Tafur, R., F. Keyl & J. Arguelles. 2010. Reproductive biology of jumbo squid Dosidicus gigas in relation to environmental variability of the northern Humboldt Current System. Mar. Ecol. Prog. Ser. 400:127-141.
Tyler, C. R. & J. P. Sumpter. 1996. Oocyte growth and development in teleosts. Rev. Fish Biol. Fish. 6:287-318.
Valdebemto, I., L. Paiva & M. Berland. 2011. Follicular atresia in teleost fish: a review. Arch. Med. Vet. 43:11-25.
Velazquez-Abunader, J. I., A. Hernandez-Herrera, S. Martinez-Aguilar. J. G. Diaz-Uribe & E. Morales-Bojorquez. 2012. Interannual variability in mantle length structure, recruitment, and sex ratio of jumbo squid, Dosidicus gigas, in the central Gulf of California, Mexico. J. Shellfish Res. 31:125-134.
Wootton, R. J. 1984. Introduction: strategies and tactics in fish reproduction. In: Potts, G. W. & R. J. Wooton, editors. Fish reproduction strategies and tactics. Orlando, FL: Academic Press, pp. 1-12.
Yatsu, A., R. Tafur & C. Maravi. 1999. Embryos and rhynchoteuthion paralarvae of the jumbo flying squid Dosidicus gigas (Cephalopoda) obtained through artificial fertilization from Peruvian waters. Fish. Sci. 65:904-908.
Young, T. P. 1981. A general model of comparative fecundity for semelparous and iteroparous life histories. Am. Nat. 118:27-36.
Zar, J. 2010. Biostatistical analysis. Upper Saddle River, NJ: Pearson Prentice-Hall. 944 pp.
Zepeda-Benitez, V. Y., E. Morales-Bojorquez, J. Lopez-Martinez & A. Hernandez-Herrera. 2014. Growth model selection for the jumbo squid Dosidicus gigas from the Gulf of California, Mexico. Aquat. Biol. 21:231-247.
XCHEL AURORA PEREZ-PALAFOX, (1) ENRIQUE MORALES-BOJORQUEZ, (2*) MARIA DEL CARMEN RODRIGUEZ-JARAMILLO, (2) JUAN GABRIEL DIAZ-URIBE, (3) AGUSTIN HERNANDEZ-HERRERA, (1) OSWALDO URIEL RODRIGUEZ-GARCIA (2) AND DANA ISELA ARIZMENDI-RODRIGUEZ (4)
(1) Centre Interdisciplinario de Ciencias Marinas, Instituto Politecnico Nacional, Av. Institute Politecnico Nacional 195, Col. Playa Palo de Santa Rita Sur, CP 23096, La Paz, Mexico; (2) Centre de Investigaciones Biologicas del Noroeste S.C., Av. Instituto Politecnico Nacional 195, Col. Playa Palo de Santa Rita Sur, CP 23096, La Paz, Mexico; (3) Centre Regional de Investigation Acuicola y Pesquera La Paz, Instituto Nacional de Pesca y Acuacultura, Carretera a Pichilingue s/n km 1CP 23020, La Paz, Mexico; (4) Cetttro Regional de Investigacion Acuicola y Pesquera Guaymas, Instituto Nacional de Pesca y Acuacultura, Calle 20 Sur 605, Colonia Cantera, CP 85400, Guaymas, Mexico
(*) Corresponding author. E-mail: email@example.com
TABLE 1. Description of the stages of ovarian development in Dosidicus gigas. Ovarian stage Oocyte Description substages Previtellogenesis I Oogonia (Og) Rounded cells with a big nucleus that fills almost all the cytoplasm, which is basophilic. These cells form groups or nest in the germinal epithelium. Early Oocytes with irregular previtellogenic shape and a big (Pv1) nucleus, with the presence of nucleoli and basophilic cytoplasm. The fcs are flat and form a thin layer. Intermediate Rounded and irregular previtellogenic oocytes with many (Pv2) nucleoli around the nucleus. The cytoplasm is abundant and basophilic. The layer of fcs ranges from flat to cubic. Late Oval oocytes with a previtellogenic well-defined nucleus (Pv3) and peripherals nucleoli. The granulosa or fcs are cubic and beginning to penetrate inside the cell, forming invaginations. Vilellogenesis II Early Oval oocytes, with the vitellogenic cytoplasm basophilic but (Vo1) starting to turn acidophilic. Cubic fcs with prominent invaginations inside the oocyte. The nucleus is displaced to the animal pole. Postvitellogenesis Late Oval oocytes with III vitellogenic acidophilic cytoplasm (Vo2) because of the accumulation of vitellum. Nucleus completely displaced to the animal pole. Very deep invaginations of the fcs. Post Oval to rounded oocytes, vitellogenic with acidophilic cytoplasm (Pos) completely filled with vitellum. The invaginations are less noticeable. Spawning IV Atresia (a) Oocytes without a clear shape. Follicular cells deteriorating. Postspawning V Postovulatory Layer of fcs with follicles (pf) pyknotic nuclei forming a central cavity. Ovarian stage Diameter, [micro]m (IC) Min-Max Previtellogenesis I 60.13 (59.85-60.42) 4.4-100.2 121.54 (120.95-122.13) 51.9-209.8 199.51 (192.50-206.53) 70.1-268.5 254.81 (241.89-267.73) 91-350.6 Vilellogenesis II 309.91 (285.47-334.36) 216.4-499.5 Postvitellogenesis 411.27 (384.24-438.29) 286.3-597.4 III 358.10 (257.30-458.91) 161.9-618.9 Spawning IV - - Postspawning V - - TABLE 2. Number of size groups estimated using a multinomial density function applied to TDs of oocytes ([micro]m) for each ovarian stage in Dosidicus gigas. Ovarian stage -lnL K AIC I One size group 3.02E-9 3 6.0 (*) II One size group 11,488.8 3 22,983.6 Two size groups 6,213.9 6 12,439.7 Three size groups 73.1 9 164.2 (*) III One size group 17,305.6 3 34,617.2 Two size groups 12,834.2 6 25,680.5 Three size groups 41.2 9 100.5 (*) IV One size group 2,513.7 3 5,033 Two size groups 18.2 6 48 (*) V One size group 14,856.1 3 29,718 Two size groups 50.2 6 112 (*) Negative log-likelihood (-ln L) is the estimated value of the objective function, and K is the number of parameters estimated. (*) The smaller AIC. TABLE 3. Number of size groups estimated using multinomial analysis applied to TDs of oocytes ([micro]m) for each ovarian stage in Dosidicus gigas. Ovarian Mean [[sigma].sub.i] [[lambda].sub.i] stage ([micro]m) I One size group 95.92 0.33 5,035.19 II One size group 50.97 0.50 26.32 Two size groups 140.99 0.79 181.09 Three size 230.70 0.81 82.08 groups III One size group 95.89 0.37 85.90 Two size groups 231.00 0.37 21.64 Three size 365.82 0.95 20.93 groups IV One size group 95.93 1.02 98.38 Two size groups 186.22 0.50 29.67 V One size group 140.89 0.41 205.09 Two size groups 276.02 0.47 64.43 Ovarian Minimum (cm) Maximum (cm) stage I One size group 95.91 95.93 II One size group 50.90 51.05 Two size groups 140.86 141.11 Three size groups 230.57 230.83 III One size group 95.83 95.96 Two size groups 230.93 231.06 Three size groups 365.65 365.99 IV One size group 95.67 96.19 Two size groups 186.09 186.34 V One size group 140.84 140.94 Two size groups 275.97 276.08 Values of the mean ([micro]m), SD ([[sigma].sub.i]), proportion ([[lambda].sub.i]), and CI (P < 0.05) for each size group are shown. TABLE 4. Monthly [chi square] and P values for Dosidicus gigas. Month Female Male [chi square] P value March 5 4 0.06 0.81 April 25 21 0.17 0.68 May 44 18 5.45 0.02 (*) June 26 9 4.13 0.04 (*) July 43 14 7.38 0.01 (*) August 43 9 11.12 0.00 (*) September 5 6 0.05 0.83 October 26 12 2.58 0.11 November 5 6 0.05 0.83 (*) Significant differences. TABLE 5. Number of annual size groups estimated using a multinomial density function applied to the ML (cm) frequency distribution of jumbo squid Dosidicus gigas. OF K AIC 2008 One size group 99.4 3 215.0 Two size groups 68.2 6 169.0 Three size groups 33.4 9 76.6 Four size groups 24.1 12 38.1 (*) 2009 One size group 565.4 3 747.6 Two size groups 242.1 6 499.3 Three size groups 160.9 9 383.0 (*) OF is the estimated value of the objective function and K is the number of parameters estimated. (*) The smaller AIC. TABLE 6. Number of size groups estimated using multinomial analysis applied to annual ML (cm) frequency distributions of Dosidicus gigas. Mean (em) [[sigma].sub.i] [[lambda].sub.i] 2008 One size group 41.2 3.7 66.9 Two size groups 48.6 2.1 38.0 Three size groups 55.2 3.8 145.7 Four size groups 68.6 1.6 22.3 2009 One size group 40.0 3.8 408.1 Two size groups 47.9 1.5 98.5 Three size groups 53.9 8.0 92.6 Minimum (cm) Maximum (cm) 2008 One size group 40.6 41.9 Two size groups 48.2 48.9 Three size groups 54.6 55.9 Four size groups 68.3 68.8 2009 One size group 39.6 40.4 Two size groups 47.7 48.1 Three size groups 53.0 54.8 Values of the mean ML (cm), SD ([[sigma].sub.i]), proportion ([[lambda].sub.i]), and CI (P < 0.05) for each size group are shown. TABLE 7. Reproductive strategy for cephalopod species reclassified from semelparous to multiple span tiers or iteroparous. Species Study period Region Sample size Grane/edone December Northeast 5 boreopacifica 1961 and Pacific March 1970 Ocean Kondakovia 15 February Ross Sea 1 longimana 2012 Octopus February 1975 Pacific 7 chierchiae coast of Vampyroteuthis Individuals Panama 43 infernalis collected Southern between California 1962 and 1972 Octopus vulgaris 2004 to 2007 Galicia waters 359 Loligo vulgaris 1992 to 1993 Southern coast of 250 reynaudii South Africa Lolliguncula 2003 to 2006 Gulf of 2,460 panamensis and 2008 California Dosidicus gigas 2008 to 2009 Gulf of 287 California Species Reproductive strategy New classification of previously reported reproductive strategy Grane/edone Unknown Multiple spawner boreopacifica Kondakovia Unknown Possibly, longimana iteroparity Octopus Semelparous Iteroparity chierchiae Vampyroteuthis Semelparous Iteroparity infernalis Octopus vulgaris Simultaneous terminal Asynchronic ovary spawner with a development and a synchronous ovulation synchronous ovulation during spawning. Loligo vulgaris Unknown Multiple spawner reynaudii Lolliguncula Unknown Synchronous ovarian panamensis development with multiple spawning Dosidicus gigas Multiple spawning Iteroparity Species Criteria for defining Source reproductive strategy Grane/edone Ovary and oviducts Bello (2006) boreopacifica were examined and measured macroscopically to determine oocyte size-frequency distributions. Kondakovia Ripe eggs in the Laptikhovsky et al. longimana oviduct in a (2013) specimen that appears to be immature with flabby oviducts do not prove this species to be iteroparous. However, they are likely to be the remnants of a previous spawning. Octopus Females copulated three Rodaniche (1984) chierchiae times while captive. Vampyroteuthis Postovulatory follicles Hoving et al. (2015) infernalis were present as evidence of previous spawning. Octopus vulgaris Histological description Sieiro et al. (2016) and prevalence of atresia. Loligo vulgaris Histological and oocyte Melo and Sauer(1999) reynaudii size-frequency analysis. Lolliguncula Histology for different Arizmendi-Rodriguez panamensis stages of development. et al. (2012) Dosidicus gigas Histology for different This study stages of development, presence of postovulatory follicles in resting females, oocyte size-frequency analysis.
|Printer friendly Cite/link Email Feedback|
|Author:||Perez-Palafox, Xchel Aurora; Morales-Bojorquez, Enrique; Del Carmen Rodriguez-Jaramillo, Maria; Diaz|
|Publication:||Journal of Shellfish Research|
|Date:||Apr 1, 2019|
|Previous Article:||HISTOLOGY ATLAS AND SYSTEMATIC APPROACH TO POSTMORTEM EXAMINATION OF THE QUEEN CONCH LOBATUS GIGAS.|
|Next Article:||POPULATION GENETIC STRUCTURE AND DIVERSITY OF THE INVASIVE ISLAND APPLE SNAIL POMACEA MACULATA (PERRY, 1810) IN SOUTH CAROLINA AND GEORGIA.|