Morphometric analysis of Capitella capitata ( Polychaeta , Capitellidae )

Accurate size measurements are fundamental in characterizing the population structure and secondary production of a species. The purpose of this study was to determine the best morphometric parameter to estimate the size of individuals of Capitella capitata (Fabricius, 1780). The morphometric analysis was applied to individuals collected in the intertidal zones of two beaches on the northern coast of the state of São Paulo, Brazil: São Francisco and Araçá. The following measurements were taken: the width and length (height) of the 4th, 5th and 7th setigers, and the length of the thoracic region (first nine setigers). The area and volume of these setigers were calculated and a linear regression analysis was applied to the data. The data were log-transformed to fit the allometric equation y = ax into a straight line (logy = loga + b * logx). The measurements which best correlated with the thoracic length in individuals from both beaches were the length of setiger 5 (r = 0.722; p<0.05 in São Francisco and r = 0.795; p<0.05 in Araçá) and the area of setiger 7 (r = 0.705; p<0.05 in São Francisco and r = 0.634; p<0.05 in Araçá). According to these analyses, the length of setiger 5 and/or the area of setiger 7 are the best parameters to evaluate the growth of individuals of C. capitata.

The polychaete Capitella capitata (Fabricius, 1780) is a non-selective deposit-feeding burrowing worm (FAUCHALD & JUMARS, 1979) found worldwide, from the intertidal to the deepest regions (HUTCHINGS, 2000).Electrophoretic analysis and morphological studies conducted by GRASSLE & GRASSLE (1974, 1976) and AMARAL (1980), respectively, revealed that C. capitata consists of a complex of as many as six sibling species that have similar adult reproductive modes, larval development and enzyme patterns for some genetic loci.Although some authors recognize that several different species are present under the concept of C. capitata (BRIDGES et al., 1994;BRIDGES, 1996;HOLMER et al., 1997;COHEN & PECHENIK, 1999;PECHENIK et al., 2000), sometimes species-level distinctions are not possible, as noted by MÉNDEZ et al. (1997).
Body-size relationships are essential in understanding population processes such as recruitment, growth and secondary productivity (TSUTSUMI, 1987(TSUTSUMI, , 1990;;QIAN et al., 1990;STEIMLE et al., 1990;MÉNDEZ et al., 1997;OMENA & AMARAL, 2000).However, with soft-bodied taxa such as marine polychaetes, body length can vary with the fixation process, and the field sampling procedures typically result in a large proportion of incomplete individuals for which body length cannot be determined directly (WARWICK & PRICE, 1975;DESROSIERS et al., 1988).One of the problems in studying the population structure of polychaetes is to establish an appropriate methodology to measure the animals.Most previous studies on the growth of C. capitata have been based on the total length of the thoracic region, and/or the length of a single thoracic setiger.The purpose of our study was to determine the best morphometric parameter to estimate the size of individuals of C. capitata from two sandy beaches before studying its population dynamics, secondary production and its role as pollution indicator.

MATERIAL AND METHODS
Specimens of C. capitata were collected from two beaches located on the mainland side of São Sebastião Channel (northern coast of state of São Paulo, Brazil): São Francisco (23°53'S, 45°24'W) and Araçá (23°49'S, 45°25'W).São Francisco Beach has a narrow intertidal zone and a gentle slope, and the sediment is composed of a mixture of sand and stones of different sizes.The site is located close to urban areas and is exposed to domestic sewage.Araçá Beach is a low-hydrodynamics, less exposed region, with some mangrove vegetation; the sediment is predominantly composed of fine sand and very fine sand.Because of the proximity to the city of São Sebastião, this beach is exposed to both domestic sewage and frequent oil spills from a local oil terminal (Ductos e Terminais Centro Sul -DTCS).In addition, since 1996 there has been a constant discharge of sewage from an underwater outlet (AMARAL et al., 1998).
Three core samples were taken every two weeks from November 2002 to March 2003, during spring tides, with a 0.01 m 2 cylindrical sampler inserted to a depth of 20 cm.Each sample was partitioned into upper and lower 10-cm fractions.The two fractions were processed differently.The upper 10 cm of sediment was sieved through 0.500, 0.250 and 0.125 mm mesh sieves.The sediment that remained on the last two sieves was processed further, using a flotation technique to collect larvae and juveniles.The flotation technique (modified after ANDERSON, 1959) consisted of saturating and agitating the sediment with seawater hyper-saturated with sugar (2.5 l seawater: 1 kg sugar) and then passing the supernatant through a 0.125 mm mesh sieve.The lower 10 cm of sediment was sieved through 1.000 and 0.500 mm mesh sieves.The collected individuals of C. capitata were relaxed in 5% MgCl 2 , fixed in 6% formalin and preserved in 70% ethanol.
All the examined individuals of C. capitata belong to a single species, according to morphological observations during the sorting and measuring processes, using a stereomicroscope and a microscope, when necessary, fitted with a graduated eyepiece.Because the morphological features of this species are the same as those described for C. capitata, we will continue to use this species name until a worldwide revision of the genus is accomplished.The width (W) and length (L) were measured for setigers 4, 5 and 7 (Fig. 1).In addition, the length of the thoracic region (first nine setigers) was measured (as recommended by WARREN (1976) and MÉNDEZ et al. (1997).The thoracic length is a precise measure as a reference parameter of the worm's size, because these animals are rarely collected entire.The area and volume of setigers 4, 5 and 7 were calculated using the formulae: external area = 2π(W/2)L and volume = (π(W/2) 2 )L, considering each setiger as a cylinder.The relationship between l/w of setigers 4, 5 and 7 was also examined.The sizes of the individuals from the two beaches were compared through a Z-test applied to the values of length of setiger 5 (L5) and length of the thoracic region.The size classes of L5 were established for both beaches (Tab.I).
Regression analysis was used to evaluate the relationship between each parameter and the length of the thoracic region.Data were log-transformed to fit the allometric equation y=ax b into a straight line (log y =log a + b*log x ), where b is the slope; its values will show positive allometry, negative allometry or isometry (TEISSIER, 1960; Figure 1.Capitella capitata (Fabricius, 1780), dorsal view of the worm (L, length of the setiger; TRL, thoracic region length; S4, setiger 4; S5, setiger 5; S7, setiger 7; W, width of the setiger).GOULD, 1966).Student's t-test was used to compare b critical values of allometry: 1 (X and Y have the same measure); 0.5 (X is a measure of area and Y is a linear measure); 0.33 (X is a measure of volume and Y is a linear measure).

RESULTS
For São Francisco Beach, the relationship between thoracic length and the size of setigers 4 and 5 was examined using 1,581 individuals; and that between thoracic length and the size of setiger 7 was examined using 1,578 individuals (three specimens did not have the complete thoracic portion).For Araçá Beach, the relationship between thoracic length and the size of setigers 4, 5 and 7 was examined using 511 individuals.Individuals collected in São Francisco Beach were significantly larger than those collected in Araçá for both length of setiger 5 (L5) and length of the thoracic region (Z = 3.15; p<0.05 and Z = 5.80; p<0.05).The length of setiger 5 (L5) ranged from 0.049 to 0.80 mm (mean 0.25 mm, sd = 0.067 mm) in individuals from São Francisco Beach (Fig. 2), and from 0.05 to 0.460 mm (mean 0.242 mm, sd = 0.059 mm) in individuals from Araçá Beach (Fig. 3).The length of the thoracic region ranged from 1.02 to 19.42 mm (mean 2.22 mm, sd = 0.58 mm) in individuals from São Francisco Beach, and from 0.80 to 3.28 mm (mean 2.10 mm, sd = 0.32 mm) in individuals from Araçá Beach.The most frequent size class interval for both São Francisco and Araçá beaches (38.4% and 41.2%) was in size class L5, ranging from 0.20 to 0.25 mm (Figs. 2, 3).
The regression analyses revealed strong relationships between almost all the parameters and the thoracic region (Tabs.II, III).The strongest correlations in individuals for both collecting sites were found for the length of setiger 5 (L5), r 2 = 0.722 (São Francisco) and r 2 = 0.795 (Araçá), and for the area of setiger 7 (A7), r 2 =0.705 (São Francisco) and r 2 = 0.634 (Araçá) (Tab.II, III; Fig. 4-7).The results from Student's t-test comparing b with critical values of allometry revealed significant negative allometric growth for all parameters (Tab.II, III).SP,Brazil. 6, linear regression between length of the thoracic region and length of setiger 5; 7, length of the thoracic region and the area of setiger 7.
Figures 4, 5. Capitella capitata (Fabricius, 1780) from São Francisco Beach, SP, Brazil.4, linear regression between length of the thoracic region and length of setiger 5; 5, length of the thoracic region and the area of setiger 7.

DISCUSSION
In general, the study of size-dependent processes in polychaetes requires identifying parameters other than total body length or body weight, because of the high percentage of individuals that are damaged during sample collection and processing.When entire individuals are available, total body length has been shown to be a good growth estimator in diverse polychaete taxa such as eunicids (FAUCHALD, 1991) and sabellids (KEMP, 1988), while other taxa such as spionids may show significant distortion in body dimensions during relaxation or preservation, causing concern about the reliability of body length as a growth estimator (YOKOYAMA, 1988).Sclerotized body parts, such as pharyngeal jaws, can be reliable body-size estimators for those taxa that possess such structures (OLIVE & GARWOOD, 1981;GLASBY, 1986).Other structures used as growth estimators in polychaetes include peristomial width in the nereidid Neanthes succinea (Frey & Leuckart, 1847) (CAMMEN, , 1991, 1994), total setiger number in the spionid Dipolydora armata (Langerhans, 1880) (LEWIS, 1998), and parapodial parameters in nereidid species (BEN-ELIAHU, 1987).
In the present study, we examined various setiger characteristics compared to thoracic length as estimates of body size in order to measure size-dependent processes in future studies of C. capitata, based on the results from WARREN (1976) andMÉNDEZ et al. (1997).After sample processing, most individuals of this important species are frequently broken into pieces or missing their posterior ends; however, the thoracic region is rarely damaged.Previous studies of the population dynamics and secondary production of C. capitata have estimated body size using parameters such as the width of various thoracic setigers (TSUTSUMI, 1987(TSUTSUMI, , 1990) ) and thoracic length (WARREN, 1976;MÉNDEZ et al., 1997).A review of the literature concerning estimating growth in C. capitata indicates that most authors used thoracic parameters such as thoracic length or the length or width of a single thoracic setiger.For example, TSUTSUMI & KIKUCHI (1984) found a strong correlation (r 2 = 0.962) between body weight and maximum thoracic width for C. capitata in Japan, and also that few specimens of this species were damaged during sample processing (TSUTSUMI, 1987(TSUTSUMI, , 1990)).When laboratory growth studies are performed, there are sufficient numbers of entire individuals of C. capitata to allow the use of total body length (TENORE, 1982;CHESNEY, 1985;TENORE & CHESNEY, 1985), wet weight (TENORE, 1983;PHILLIPS & TENORE, 1984;CHESNEY & TENORE, 1985) or body volume (BRIDGES et al., 1994;BRIDGES, 1996).Calculation of body volume typically uses thoracic setiger measurements.QIAN (1994) cautioned about the use of wet weight, because it is greatly affected by the weight of the gut contents.Although individuals of C. capitata were significantly different in size between the two beaches studied, our results show that the length of setiger 5 and the area of setiger 7 are the most appropriate parameters to estimate body size for these populations.We recommend the use of setiger 5 length, because the thoracic region is almost always intact and this parameter is easily accessible for measurement.

Table II .
Relationship between the length of the thoracic setigers (nine setigers) and each allometric parameter of Capitella capitata (Fabricius, 1780) from São Francisco Beach, SP, Brazil (Par, Parameters; N, number of individuals; a, Y-intercept; b, regression coefficient; r 2 , coefficient of determination; p, significance of t; *, p ≤ 0.05; t, Student t-test comparing b with critical values of allometry;