1 Introduction

In the marine ecosystem, cephalopods are the predators of crustaceans, mollusks and various fishes (Arkhipkin and Bizikov, 1998; Nigmatullin et al., 2001). Meanwhile,they are an important food source for whales (Sekiguchi et al., 1996), sharks (Stevens, 1973; Tricas, 1979), tuna(Perrin et al., 1973), swordfish (Bello, 1991), porpoise(Wilke and Nicholson, 1958) and seabirds (Imber, 1975).Being both a predator and a prey, cephalopods play a crucial role in the marine ecosystem (Navarro et al., 2013).

The cephalopod beak, the primary feeding organ, has a stable shape and is primarily composed of proteins and chitin fibers (Broomell et al., 2007; Miserez et al., 2007).The hard beaks resist digestion and are usually found in the stomachs of several marine predators (Lu and Ickeringill, 2002). This durability offers an opportunity to estimate the size and biomass of squid post-predation(Jackson and McKinnon, 1996; Lalas, 2009; Lu and Ickeringill, 2002). The development of beak pigmentation is the most important variation in the growth process of the beak (Mangold and Fioroni, 1966). Hernańdez-García et al. (1998) found that darkening indicates that the beak has become harder and stronger, which can affect the feeding habits and behavior of the squids. Growth increments were found in both the rostrum sagittal section and the lateral wall surface, with increments being validated in the daily deposition of beak (Hernández-López et al.,2001; Perales-Raya et al., 2014; Perales-Raya et al., 2010).To this point, beaks have been used to estimate the specimen age of several species (Castanhari and Tomás,2012; Liu et al., 2015a; Barcenas et al., 2014), with other studies using the beaks to investigate the feeding ecology(Cherel and Hobson, 2005) and for species identification(Pierce et al., 1994; Xavier et al., 2007).

The cephalopod beaks can be conveniently measured due to their stable configuration. Traditional morphomet-rics of beaks are commonly used for interspecies and intraspecies identification (Chen et al., 2012; Martínez et al.,2002; Vega et al., 2002). Ogden et al. (1998) measured the beak shapes of nine species and suggested that beak morphometrics could be used to distinguish between genera. Martínez et al. (2002) have classified the I. coindetii,I. illecebrosus, and I. argentinus based on standardized body and beak variables and even identified different populations of each species using this method. Vega et al.(2002) analyzed the morphological differences among three geographic populations of the Patagonian squid Loligo gahi and identified different populations based on 23 morphometric characters of body, beaks, gladius and statoliths. They found that hard structures were more accurate than soft body features in identifying different populations. Fang et al. (2014) have identified different geographic populations of Ommastrephes bartramii using corrected morphometric characteristics of beaks and statoliths and suggested that more appropriate hard structure variables would improve the rate of correct population identification. Liu et al. (2015b) found that the beak variables were more effective than body variables in identifying different geographic populations of Dosidicus gigas,and that standardized beak variables were more effective than original variables in population identification. In addition, geometric morphometrics have also been used for beak identification (Crespi-Abril et al., 2010; Neige and Dommergues, 2002). Crespi-Abril et al. (2010) analyzed the body and beak shape of I. argentinus using geometric morphometrics and found that beak shape did not vary between the maturity stages of the individuals,but the shape of the body varied with the growth of squids.However, although traditional beak morphology has been widely used for classifying species in previous studies, no recognized standardization method has yet been commonly accepted. In this study, we attempted to compare the methods and determine the best one.

The ommastrephid species are widely distributed in the Pacific Ocean, Atlantic Ocean, and Indian Ocean, accounting for more than one-half of the word’s total cephalopod catch (Wang et al., 2005). Therefore, it is the most important target in the cephalopod fishery. I. argentinus, O. bartramii, and D. gigas are the primary commercial species of Ommastrephidae (Chen et al., 2009)with a similar beak shapes. using these three ommastrephid species as experimental examples, this study was conducted to analyze the data of the upper and lower beak morphological variables based on three standardization methods to remove the effects of size. The three species were identified by stepwise discriminant analysis (SDA).Then the most effective method was selected to provide an effective standardization method to identify cephalopods through their beaks.

2 Materials and Methods

Samples of the three ommastrephid species, I. argentinus, O. bartramii and D. gigas were randomly collected from the catch of Chinese commercial jigging vessels in 2007, 2009, 2010 and 2012 (Table 1). A total of 1163 squids were sampled and immediately frozen aboard the vessels. The frozen squids were then defrosted in the laboratory and the mantle length (ML) was measured to the nearest 1 mm. Sex was identified, and the maturity stage was determined following the approach defined by Lipiński and Underhill (1995). Beaks were extracted from the buccal mass and preserved in 70% alcohol. A total of 12 morphometric beak variables were measured using digital calipers to the nearest 0.01 mm. These variables included upper hood length (UHL), upper crest length(UCL), upper rostrum length (URL), upper rostrum width(URW), upper lateral wall length (ULWL), upper winglength (UWL), lower hood length (LHL), lower crest length (LCL), lower rostrum length (LRL), lower rostrum width (LRW), lower lateral wall length (LLWL), and lower wing length (LWL) (Fig.1).

Table 1 Sample information of the three ommastrephid species

Species Sampling date Fishing area ML (mm) Number of samples Sex (F, M)I. argentinus Feb. to Sep. 2007 57˚50´–60˚50´W, 40˚00´–47˚00´S 174–335 397 216, 181 O. bartramii May to Oct. 2012 153˚20´–180˚00´E, 39˚00´–45˚30´N 202–483 382 275, 107 D. gigas Sep. 2009 to Jul. 2010 82˚00´–83˚50´W, 10˚00´–18˚00´S 220–545 384 305, 79

Fig.1 Schematic diagram of beak morphometric variables (modified from Fang et al., 2014). A is upper hood length(UHL), B is upper crest length (UCL), C is upper rostrum length (URL), D is upper rostrum width (URW), E is upper lateral wall length (ULWL), F is upper wing length (UWL), G is lower hood length (LHL), H is lower crest length (LCL), I is lower rostrum length (LRL), J is lower rostrum width (LRW), K is lower lateral wall length (LLWL), and L is lower wing length (LWL).

随着医院快速发展，CT、MRI、DSA、PET、直线加速器等大型医疗设备已成为现代医院常规配置。它们在诊疗过程中发挥越来越重要作用的同时，安全运行、维护保养等系列售后问题，直接牵涉的持续不断的资金支出，也越来越令管理者头痛。

1) The first method:

The beak morphometric variables should be standardized before analyzing the data to account for the effect of size (Moltschaniwskyj, 1995; O’Dor and Hoar, 2000). In this study, the following three methods that were used to standardize the beak morphometric variables were compared:

These ratios are used as morphometric indices of beaks in the following stepwise discriminant analyses (SDA)(Pineda et al., 2002; Ogden et al., 1998), where CL is the crest length of the beak. The upper beak indices are the ratios of the upper beak morphometric variables to UCL,while the lower beak indices are the ratios of the upper beak morphometric variables to LCL.

For the upper beaks, five standardized morphological variables, namely, URWs, ULWLs, UWLs, URLs, and UHLs, were selected in the discriminant analysis for species identification (Table 5). The overall successful classification rate was 65.11% with 85.39% for I. argentinus,57.33% for O. bartramii and 52.60% for D. gigas (Table 5, Fig.3A). For the lower beaks, the SDA indicated that only four standardized morphological variables, i.e., LRLs,LRWs, LWLs, and LHLs, could be included for classifying the three species (Table 6). The correct classification rates of I. argentines, O. bartramii and D. gigas, were 66.50%,59.16 and 79.95%, respectively. The overall cross-validation classification rate was 68.54% (Table 6, Fig.3B).

浙江人民美术出版社从即日起面向社会各界公开征集陆抑非先生作品及著述相关资料，包括绘画、书法作品，画稿、手稿、著作、书信、讲稿、照片、讲课笔记等。

Y is a beak morphometric variable, is the standardized value for the individual i, and Yi and CLi are the observed values of Y and CL for the individual i, respectively. CL0 is the arithmetic mean of CL. b can be predicted from the following formula:

where a and b are the parameters to be estimated and σ2 is the variance for the normally distributed unexplained error ε (Lleonart et al., 2000). The upper beak and lower beak morphometric variables are standardized by UHL and LHL, respectively. The standardized morphometric variables are represented by adding a lower case letter ‘s’after each variable (i.e., UCLs, URLs, URWs, ULWL,UWLs, LCLs, LRLs, LRWs, LLWLs and LWLs).

2.3.1 临床疗效 26个RCTs报告了治疗后CP的临床疗效，根据干预措施的5个亚组分析结果如图3所示：针刺治疗的疗效显著优于中成药、西药、中药坐浴和微波的疗效，与中药汤剂相比不具有显著差异。

3) The third method:

where Y is one of the morphometric variables for beaks, is the standardized value for the individual i, and Yi and HLi are the observed values of Y and HL for the individual i, respectively. HL0 is the arithmetic mean of HL. b can be predicted from the following formula:

where a and b are the parameters to be estimated, and σ2 is the variance for the normally distributed random error ε(Lleonart et al., 2000). The upper beak and lower beak morphometric variables are standardized by UCL and LCL, respectively. The standardized morphometric variables are denoted by a lower case letter ‘s’ after each variable (i.e., UHLs, URLs, URWs, ULWL, UWLs, LHLs,LRLs, LRWs, LLWLs, and LWLs).

有效运用智能化技术，可以使电气设备在运作的过程当中得到更好的监管，确保电气设备能够始终保持稳定运行的状态。在使用相应的电气设备前，工作人员一定要对其进行认真的调试，这样做主要是为了在最大限度上避免电气设备在后续运作的过程中出现严重错误，降低安全事故出现的频率。通过应用智能化技术，相关的工作人员就能够进行远程的调试，且调试的方向也会更加的明确，以此确保电气设备可以一直正常的运行。

如2017高考作文“中国不仅拥有古老文明的传统，还拥有现代化的高科技”，把传统的“国粹”“京剧”和“高铁”“移动支付”关联起来；“中国虽然美食林立，但食品安全令人担忧”，把“中华美食”和“食品安全”关联起来。2018云浮高二期末作文里，“成功人生不仅需要理想，而且需要为理想去拼搏”，把“理想”和“拼搏”关联起来；“成功人生不仅表现在学业上，还要表现在家庭上”，把“大学”“事业”和“家庭”关联起来；“即使没有财富，但能付出，能知足也是成功人生”，把“财富”“奉献”“知足”关联起来。

家长都是成年人，每个人都应该有能力根据当时具体的情况来确认合理和适当的处理办法，合理地维护自己孩子的权益。当然，原则还有一条：“不能伤害对方孩子的心理和身体。”

An analysis of variance (ANOVA) was conducted to evaluate the differences between beak ratios among the three species. An SDA was used to identify the beaks of the three species based on the standardized beak morphometric variables. Mean scores on the first two functions in the SDA were calculated, and a leave-one-out cross-validation method was used to determine the rates of correct classification of the three species using the upper and lower beaks.

3 Results

3.1 Discriminant Analysis Using Beak Ratios

For the upper beaks, the average ratios of the upper beak morphological variables to UCL of O. bartramii and D. gigas were greater than those of I. argentinus (Table 2).For the lower beaks, the average values of LRL/LCL,LRW/LCL and LLWL/LCL of O. bartramii and D. gigas were greater than those of I. argentinus. However, the average values of LWL/LCL of O. bartramii and D. gigas were lower than those of I. argentinus (Table 2). All the ratios of the beak morphological variables and CL were significantly different among the three species (ANOVA,P ＜ 0.001) (Table 2).

Table 2 The ratios between beak morphometric variables and crest length for I. argentinus, O. bartramii, and D. gigas

Note: Data are listed as mean ± SD.

Ratio I. argentinusO. bartramii D. gigas P UHL/UCL 0.781±0.0270.829±0.033 0.818±0.028 ＜0.001 URL/UCL 0.224±0.0210.283±0.027 0.295±0.025 ＜0.001 URW/UCL 0.172±0.0140.254±0.026 0.262±0.023 ＜0.001 ULWL/UCL 0.814±0.0250.882±0.026 0.868±0.028 ＜0.001 UWL/UCL 0.237±0.0180.260±0.020 0.247±0.021 ＜0.001 LHL/LCL 0.511±0.0390.513±0.043 0.493±0.049 ＜0.001 LRL/LCL 0.438±0.0360.491±0.045 0.581±0.050 ＜0.001 LRW/LCL 0.413±0.0370.502±0.045 0.548±0.046 ＜0.001 LLWL/LCL 1.449±0.0761.514±0.095 1.540±0.093 ＜0.001 LWL/LCL 0.923±0.0620.869±0.065 0.804±0.070 ＜0.001

For the upper beaks, five morphological indices, i.e.,URW/UCL, ULWL/UCL, UWL/UCL, URL/UCL, and UHL/UCL, were included in the discriminant analysis (Table 3).The cross-validation classification rates were 98.99% for I. argentinus, 67.28% for O. bartramii, and 69.53% for D.gigas, and the overall successful classification rate was 78.60% (Table 3, Fig.2A). For the lower beaks, the SDA revealed that five morphological indices (LRL/LCL,LWL/LCL, LRW/LCL, LHL/LCL, and LLWL/LCL) could be chosen for species discrimination (Table 4). The overall cross-validation classification rate was 88.26%, with 91.69% for I. argentinus, 80.63% for O. bartramii, and 92.45% for D. gigas (Table 4, Fig.2B).

Table 3 The results estimated by stepwise discriminant analysis based on upper beak ratios for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

?

Table 4 The results estimated by stepwise discriminant analysis based on lower beak ratios for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

?

Fig.2 Canonical discriminant plots of upper beak (A) and lower beak (B) ratios of samples from the three ommastrephid species (C1 stands for I. argentinus, C2 stands for O. bartramii, and C3 stands for D. gigas).

3.2 Discriminant Analysis Using Morphometric Beak Variables Standardized by CL

2) The second method:

Table 5 The results estimated by stepwise discriminant analysis based on upper beak variables standardized by UCL for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

Step Variable Wilks’ λ P Species I. argentinus O. bartramii D. gigas Correct (%)Number of specimens classified 1 URWs 0.670 ＜ 0.001 I. argentinus 339 27 31 85.39 2 ULWLs 0.606 ＜ 0.001 O. bartramii 92 219 71 57.33 3 UWLs 0.575 ＜ 0.001 D. gigas 88 94 202 52.60 4 URLs 0.560 ＜ 0.001 5 UHLs 0.550 ＜ 0.001 Total 519 340 304 65.11

Table 6 The results estimated by stepwise discriminant analysis based on lower beak variables standardized by LCL for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

Number of specimens classified Step Variable Wilks’ λ P Species Correct (%)I. argentinus O. bartramii D. gigas 1 LRLs 0.650 ＜ 0.001 I. argentinus 264 100 33 66.50 2 LRWs 0.513 ＜ 0.001 O. bartramii 100 226 56 59.16 3 LWLs 0.465 ＜ 0.001 D. gigas 51 26 307 79.95 4 LHLs 0.450 ＜ 0.001 Total 415 352 396 68.54

Fig.3 Canonical discriminant plots of upper beak variables (A) standardized by UCL and lower beak variables (B) standardized by LCL of samples from the three ommastrephid species (C1 stands for I. argentinus, C2 stands for O. bartramii,and C3 stands for D. gigas).

3.3 Discriminant Analysis Using Beak Morphometric Variables Standardized by HL

For the upper beaks, five standardized variables,namely, URWs, UCLs, UWLs, ULWLs, and URLs, were selected in the discriminant analysis to identify the species (Table 7). The highest correct classification rate was 79.60% for I. argentinus. The successful classificationrates of O. bartramii and D. gigas were 56.81% and 50.78%, respectively. The overall cross-validation classification rate was 62.40% (Table 7, Fig.4A). For the lower beaks, only LRLs, LRWs and LWLs were chosen among the five standardized variables for the discriminant analysis of species identification (Table 8). The successful classification rates of I. argentinus, O. bartramii, and D.gigas were 61.46%, 57.59%, and 81.77%, respectively.The overall successful classification rate was 66.94%(Table 8, Fig.4B).

在式(7)和式(8)中：w(j,k)为待处理系数，为公式处理之后的系数.sign(·)为符号函数，λ为阈值规则计算出来的阈值.

Table 7 The results estimated by stepwise discriminant analysis based on upper beak variables standardized by UHL for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

Step Variable Wilks’ λ P Species Number of specimens classified Correct (%)I. argentinus O. bartramii D. gigas

Table 8 The results estimated by stepwise discriminant analysis based on lower beak variables standardized by LHL for the three ommastrephid species and a classification matrix with percentages of correctly classified individuals

Step Variable Wilks’ λ P Species I. argentinus O. bartramii D. gigas Correct (%)Number of specimens classified 1 LRLs 0.601 ＜ 0.001 I. argentinus 244 120 33 61.46 2 LRWs 0.485 ＜ 0.001 O. bartramii 108 220 54 57.59 3 LWLs 0.425 ＜ 0.001 D. gigas 48 22 314 81.77 Total 400 362 401 66.94

Fig.4 Canonical discriminant plots of upper beak variables (A) standardized by UHL and lower beak variables (B) standardized by LHL of samples from the three ommastrephid species (C1 stands for I. argentinus, C2 stands for O. bartramii,and C3 stands for D. gigas).

4 Discussion

Species identification is fundamental for research on the feeding ecology and the structure of marine ecosystems (Chen et al., 2012). Since the cephalopod beak can be easily measured, this traditional morphological method has been widely used for species identification (Martínez et al., 2002; Vega et al., 2002). It is very important to remove the allometric effects of body size from traditional morphometrics (Liu et al., 2015b; Lleonart et al.,2000).

In this study, the ratios of beak morphometrics to crest length were found to be effective in classifying different species, and the overall accurate rates of discriminant analysis were 78.60% and 88.26% using the upper beak and lower beak variables, respectively (Tables 3 and 4).Meanwhile, the ratios of beak morphological variables to crest length showed significant differences among the three species (Table 2), suggesting that the beak ratios could be defined as important indices to classify species in future studies. In previous studies, the beak ratios were also determined as important taxonomic variables in distinguishing species because of their stability with the growth of individuals (Brunetti and Ivanovic, 1997; Ikica et al., 2014; Lu and Ickeringill, 2002; Yang et al., 2012).Yang et al. (2012) analyzed the morphometric beak variables of Todarodes pacificus and found that the ratios of rostrum length and hood length to crest length were consistent with mantle length. Lu et al. (2013) found that the ratios of beak morphometric variables to crest length were nearly constant with the growth of I. argentinus, and showed no significant differences among the groups with different mantle lengths. Meanwhile, Lu and Ickeringill(2002) determined that the ratios of beak morphometric variables to rostrum length and crest length were important morphomeric indices for classifying the cephalopod beaks of 75 species. The ratios of beak morphometric variables to rostrum length and crest length have also been used as improtant indices for beak identification of 18 cephalopod species from the Pacific Ocean (Wolff, 1984). It is also believed that character/character of beaks can be used as a taxonomic variable in discriminant analysis to identify different species (Ogden et al., 1998). However, the rostrum length could not be accurately determined due to the erosion of the beak rostrum (Clarke, 1962; Lefkaditou and Bekas, 2004).Therefore, in this study, the ratios of beak morphomeric variables to crest length were determined to be used in the discriminant analysis to classify the three ommastrephid species.

To remove the allometric effects of body size, the measured morphometric variables need to be standardized(Lleonart et al., 2000). Previous studies have used standardized morphometric variables in the discriminant analysis for species identification (Chen et al., 2012; Vega et al., 2002). Chen et al. (2012) successfully identified four species using beak morphological variables standardized using an allometric model. Two geographic groups were also identified by standardized statolith and beak morphological variables (Fang et al., 2014). Therefore, in this study, not only the character/CL of beaks, but also the beak variables standardized by crest length by an allometric model were used in the discriminant analysis.When these two methods were compared, the successful classification rate was higher when using the ratios of beak morphological variables and crest length (Tables 3–6).

In previous studies, the beak morphometric variables were standardized by hood length, which were then used in the discriminant analysis (Chen et al., 2012; Fang et al.,2014; Liu et al., 2015b). However, the hood length is smaller than the crest length, which could be due to the effect of the erosion of the beak rostrum (Fig.2A).Therefore, the beak morphometric variables standardized by crest length may be a better method than being standardized by hood length. To validate this hypothesis,the beak variables standardized by hood length were used for identifying the species, which resulted in a successful classification rate that was slightly lower than that when using beak variables standardized by crest length (Tables 5–8). Therefore, the ratios of beak variables to crest length are the most effective for identifying the three ommastrephid species among the three standardization methods. The results of the discriminant analysis using beak ratios showed that all the beak variables could be included in the discriminant analysis for species identification. For the lower beaks, each species has a high successful classification rate, which was greater than 80%.However, for the upper beaks, the correct classification rate of I. argentinus (98.99%) was much greater than those of O. bartramii (67.28%) and D. gigas (69.53%),and there was a considerable overlap between the classification rates of O. bartramii and D. gigas (Fig.2A). This might be caused due to the different degrees of genetic differentiation among these three species. Yokawa (1994)also found that the position of I. argentinus is considerably distant from those of O. bartramii and D. gigas in the dendrogram, and the genetic distance between O.bartramii and D. gigas is close. Some outliers were far from the species in the discriminant scatter plots (Figs.2–4), which was probably induced by the serious erosion of beaks.

In previous studies, the upper and lower beak morphometric variables were usually combined for intraspecies and interspecies identification (Pineda et al., 2002; Vega et al., 2002). However, if the beaks in the stomachs of predators do not appear in pairs, or one of them is injured,the species can not be identified using both the upper beak and the lower beak simultaneously. Therefore, in this study, the upper beaks and the lower beaks were used discretely in the discriminant analysis and the successful classification rates of the three standardization methods were compared. The lower beaks were found to be more effective than the upper beaks for each standardization method. This was probably caused due to the differences in lower beaks that were more significant than those of the upper beaks among the three species. After identifying the beaks in the stomachs of predators, the body size and biomass of preys can be estimated using beak morphometric variables (GroÈger et al., 2000; Jackson, 1995).Meanwhile, stable isotope analysis of the beaks can improve the understanding of the trophic structure and the life history of cephalopods (Cherel et al., 2009; Hobson and Cherel, 2006; Ruiz-Cooley et al., 2006).

In summary, among the three standardization methods,the ratios of beak morphometric variables to crest length were the most effective in classifying the species.Compared with hood length, the beak variables standardized by crest length using an allometric model were more effective for species identification. Compared with upper beaks, the lower beaks were more effective in the discriminant analysis for each standardization method.Therefore, the ratios of beak morphological variables and crest length should be used in future studies of interspecies and intraspecies identification. This method is not only effective, but also convenient. Meanwhile, the measured beak morphometric variables can be used to estimate the body size and the biomass of cephalopods.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 41306127 and 41276156),the National Science Foundation of Shanghai (No. 13ZR 1419700), the Innovation Program of Shanghai Municipal Education Commission (No. 13YZ091), and the Shanghai Leading Academic Discipline Project (Fisheries Discipline). The involvement of Y. Chen was supported by Shanghai Ocean University (SHOU) International Center for Marine Studies and Shanghai Visiting 1000 Talent Program.

References

Arkhipkin, A., and Bizikov, V., 1998. Statoliths in accelerometers of squid and cuttlefish. Ruthenica, 8: 81-84.

Barcenas, G. V., Perales-Raya, C., Bartolome, A., Almansa, E.,and Rosas, C., 2014. Age validation in Octopus maya (Voss and Solis, 1966) by counting increments in the beak rostrum sagittal sections of known age individuals. Fisheries Research, 152: 93-97.

Bello, G., 1991. Role of cephalopods in the diet of the swordfish,Xiphias gladius, from the eastern Mediterranean Sea. Bulletin of Marine Science, 49: 312-324.

Broomell, C. C., Khan, R. K., Moses, D. N., Miserez, A., Pontin,M. G., Stucky, G. D., Zok, F. W., and Waite, J. H., 2007.Mineral minimization in nature’s alternative teeth. Journal ofthe Royal Society Interface, 4: 19-31.

Brunetti, N., and Ivanovic, M., 1997. Description of Illex argentinus beaks and rostral length relationships with size and weight of squids. Revista de Investigación y Desarrollo Pesquero INIDEP, 11: 135-144.

Castanhari, G., and Tomás, A. R. G., 2012. Beak increment counts as a tool for growth studies of the common octopus Octopus vulgaris in southern Brazil. Boletim Do Instituto De Pesca, 38:323-331.

Chen, X., Liu, B., and Wang, Y., 2009. The Cephalopod in the World. Ocean Press, Beijing, 89-91 (in Chinese).

Chen, X., Lu, H., Liu, B., Chen, Y., Li, S., and Jin, M., 2012.Species identification of Ommastrephes bartramii, Dosidicus gigas, Sthenoteuthis oualaniensis and Illex argentinus (Ommastrephidae) using beak morphological variables. Scientia Marina, 76: 473-481.

Cherel, Y., and Hobson, K. A., 2005. Stable isotopes, beaks and predators: A new tool to study the trophic ecology of cephalopods, including giant and colossal squids. Proceedings of the Royal Society of London B: Biological Sciences, 272:1601-1607.

Cherel, Y., Ridoux, V., Spitz, J., and Richard, P., 2009. Stable isotopes document the trophic structure of a deep-sea cephalopod assemblage including giant octopod and giant squid.Biology Letters, 5: 364.

Clarke, M. R., 1962. The identification of cephalopod ‘beaks’and the relationship between beak size and total body weight.Bulletin of the British Museum (Natural History), Zoology, 8(10): 419-480.

Crespi-Abril, A. C., Morsan, E. M., and Barón, P. J., 2010.Analysis of the ontogenetic variation in body and beak shape of the Illex argentinus inner shelf spawning groups by geometric morphometrics. Journal of the Marine Biological Association of the United Kingdom, 90: 547-553.

Fang, Z., Liu, B., Li, J., Su, H., and Chen, X., 2014. Stock identification of neon flying squid (Ommastrephes bartramii)in the North Pacific Ocean on the basis of beak and statolith morphology. Scientia Marina, 78: 239-248.

GroÈger, J., Piatkowski, U., and Heinemann, H., 2000. Beak length analysis of the Southern Ocean squid Psychroteuthis glacialis (Cephalopoda: Psychroteuthidae) and its use for size and biomass estimation. Polar Biology, 23: 70-74.

Hernańdez-García, V., Piatkowski, U., and Clarke, M., 1998.Development of the darkening of Todarodes sagittatus beaks and its relation to growth and reproduction. South African Journal of Marine Science, 20: 363-373.

Hernández-López, J. L., Castro-Hernández, J. J., and Hernández-García, V., 2001. Age determined from the daily deposition of concentric rings on common octopus (Octopus vulgaris) beaks. Fishery Bulletin-National Oceanic and Atmospheric Administration, 99: 679-684.

Hobson, K. A., and Cherel, Y., 2006. Isotopic reconstruction of marine food webs using cephalopod beaks: New insight from captively raised Sepia officinalis. Canadian Journal of Zoology, 84: 766-770.

Ikica, Z., Vuković, V., Đurović, M., Joksimović, A., and Šifner,S. K., 2014. Analysis of beak morphometry of the horned octopus Eledone cirrhosa, Lamarck 1798 (Cephalopoda: Octopoda), in the south-eastern Adriatic Sea. Acta Adriatica, 55:43-56.

Imber, M. J., 1975. Lycoteuthid squids as prey of petrels in New Zealand seas. New Zealand Journal of Marine and Freshwater Research, 9: 483-492.

Jackson, G. D., 1995. The use of beaks as tools for biomass estimation in the deepwater squid Moroteuthis ingens (Cephalopoda: Onychoteuthidae) in New Zealand waters. Polar Biology, 15: 9-14.

Jackson, G. D., and McKinnon, J. F., 1996. Beak length analysis of arrow squid Nototodarus sloanii (Cephalopoda: Ommastrephidae) in southern New Zealand waters. Polar Biology, 16:227-230.

Pineda, S. E., Hernandez, D. R., Brunetti, N. E., and Jerez, B.,2002. Morphological identification of two southwest Atlantic Loliginid squids: Loligo gahi and Loligo sanpaulensis. Revista de Investigacion y Desarrollo Pesquero, 15: 67-84.

Lalas, C., 2009. Estimates of size for the large octopus Macroctopus maorum from measures of beaks in prey remains. New Zealand Journal of Marine and Freshwater Research, 43:635-642.

Lefkaditou, E., and Bekas, P., 2004. Analysis of beak morphometry of the horned octopus Eledone cirrhosa (Cephalopoda: Octopoda) in the Thracian Sea (NE Mediterranean).Mediterranean Marine Science, 5: 143-150.

Lipiński, M., and Underhill, L., 1995. Sexual maturation in squid: Quantum or continuum? South African Journal of Marine Science, 15: 207-223.

Liu, B. L., Chen, X. J., Chen, Y., and Hu, G. Y., 2015a. Determination of squid age using upper beak rostrum sections:Technique improvement and comparison with the statolith.Marine Biology, 162: 1685-1693.

Liu, B. L., Fang, Z., Chen, X. J., and Chen, Y., 2015b. Spatial variations in beak structure to identify potentially geographic populations of Dosidicus gigas in the eastern Pacific Ocean.Fisheries Research, 164: 185-192.

Lleonart, J., Salat, J., and Torres, G. J., 2000. Removing allometric effects of body size in morphological analysis. Journal of Theoretical Biology, 205: 85-93.

Lu, C., and Ickeringill, R., 2002. Cephalopod beak identification and biomass estimation techniques: Tools for dietary studies of southern Australian finfishes. Museum Victoria Science Reports, 6: 1-32.

Lu, H., Chen, X., and Liu, B., 2013. Effects of individual size on beak morphology of Illex argentinus in the southwestern Atlantic Ocean. Journal of Fisheries of China, 35: 247-253(in Chinese with English abstract).

Mangold, K., and Fioroni, P., 1966. Morphologie et biométrie des mandibules de quelques céphalopodes méditerranéens.Vie et Milieu, 17: 1139-1196.

Martínez, P., Sanjuan, A., and Guerra, A., 2002. Identification of Illex coindetii, I. illecebrosus and I. argentinus (Cephalopoda:Ommastrephidae) throughout the Atlantic Ocean; by body and beak characters. Marine Biology, 141: 131-143.

Miserez, A., Li, Y., Waite, J. H., and Zok, F., 2007. Jumbo squid beaks: Inspiration for design of robust organic composites.Acta Biomaterialia, 3: 139-149.

Moltschaniwskyj, N., 1995. Changes in shape associated with growth in the loliginid squid Photololigo sp.: A morphometric approach. Canadian Journal of Zoology, 73: 1335-1343.

Navarro, J., Coll, M., Somes, C. J., and Olson, R. J., 2013.Trophic niche of squids: Insights from isotopic data in marine systems worldwide. Deep Sea Research Part II: Topical Studies in Oceanography, 95: 93-102.

Neige, P., and Dommergues, J. L., 2002. Disparity of beaks and statoliths of some coleoids: A morphometric approach to depict shape differentiation. Abhandlungen der Geologischen Bundesanstalt, 57: 393-399.

Nigmatullin, C. M., Nesis, K., and Arkhipkin, A., 2001. A review of the biology of the jumbo squid Dosidicus gigas (Cephalopoda: Ommastrephidae). Fisheries Research, 54: 9-19.

O’Dor, R., and Hoar, J., 2000. Does geometry limit squid growth?ICES Journal of Marine Science: Journal du Conseil, 57: 8-14.

Ogden, R. S., Allcock, A., Wats, P., and Thorpe, J., 1998. The role of beak shape in octopodid taxonomy. South African Journal of Marine Science, 20: 29-36.

Perales-Raya, C., Almansa, E., Bartolomé, A., Felipe, B. C.,Iglesias, J., Sánchez, F. J., Carrasco, J. F., and Rodríguez, C.,2014. Age validation in Octopus vulgaris beaks across the full ontogenetic range: Beaks as recorders of life events in octopuses. Journal of Shellfish Research, 33: 481-493.

Perales-Raya, C., Bartolomé, A., García-Santamaría, M. T., Pascual-Alayón, P., and Almansa, E., 2010. Age estimation obtained from analysis of octopus (Octopus vulgaris Cuvier, 1797)beaks: Improvements and comparisons. Fisheries Research,106: 171-176.

Perrin, W. F., Warner, R., Fiscus, C., and Holts, D., 1973.Stomach contents of porpoise, Stenella spp., and yellowfin tuna, Thunnus albacares, in mixed-species aggregations.Fishery Bulletin, 71: 1077-1092.

Pierce, G. J., Thorpe, R. S., Hastie, L. C., Brierley, A., Guerra,A., Boyle, P. R., Jamieson, R., and Avila, P., 1994. Geographic variation in Loligo forbesi in the northeast Atlantic Ocean: Analysis of morphometric data and tests of causal hypotheses. Marine Biology, 119: 541-547.

Ruiz-Cooley, R., Markaida, U., Gendron, D., and Aguiniga, S.,2006. Stable isotopes in jumbo squid (Dosidicus gigas) beaks to estimate its trophic position: Comparison between stomach contents and stable isotopes. Journal of the Marine Biological Association of the United Kingdom, 86: 437-445.

Sekiguchi, K., Klages, N., and Best, P., 1996. The diet of straptoothed whales (Mesoplodon layardii). Journal of Zoology,239: 453-463.

Stevens, J., 1973. Stomach contents of the blue shark (Prionace glauca L.) off south-west England. Journal of the Marine Biological Association of the United Kingdom, 53: 357-361.

Tricas, T. C., 1979. Relationships of the blue shark, Prionace glauca, and its prey species near Santa Catalina Island, California. Fishery Bulletin, 77: 175-182.

Vega, M. A., Rocha, F. J., Guerra, A., and Osorio, C., 2002.Morphological differences between the Patagonian squid Loligo gahi populations from the Pacific and Atlantic Oceans.Bulletin of Marine Science, 71: 903-913.

Wang, Y., Chen, X., and Liu, B., 2005. The Resource and Biology of Economic Oceanic Squid in the World. Ocean Press, Beijing, 124-137 (in Chinese).

Wilke, F., and Nicholson, A., 1958. Food of porpoises in waters off Japan. Journal of Mammalogy, 39: 441-443.

Wolff, G. A., 1984. Identification and estimation of size from the beaks of 18 species of cephalopods from the Pacific Ocean.NOAA Technical Report NMFS, 17: 1-50.

Xavier, J., Clarke, M. R., Magalhães, M. C., Stowasser, G.,Blanco, C., and Cherel, Y., 2007. Current status of using beaks to identify cephalopods: III international workshop and training course on cephalopod beaks, Faial Island, Azores,April 2007.

Yang, L., Jang, Y., Liu, Z., Lin, N., Li, S., and Cheng, J., 2012.Variation analysis on partial morphometric measurements of beaks of Todarodes pacificus inhabiting East China Sea.Periodical of Ocean University of China, 42: 51-57.

Yokawa, K., 1994. Allozyme differentiation of sixteen species of ommastrephid squid (Mollusca, Cephalopoda). Antarctic Science, 6: 201-204.