10
Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specic growth patterns in Conception Bay, Newfoundland Nami Choe , Don Deibel Ocean Sciences Centre, Memorial University of Newfoundland, St. John's, Newfoundland, Canada A1C 5S7 abstract article info Article history: Received 26 February 2009 Received in revised form 21 May 2009 Accepted 22 May 2009 Keywords: Age determination Larvacean Length-at-age Statolith We explored the use of statolith diameter as an age indicator for the larvacean Oikopleura vanhoeffeni. Laboratory studies indicated that variability in statolith diameter-at-age is substantially lower than that of trunk length-at-age, and that variability in statolith diameter-at-age remains constant as statolith diameter increases with age, while variability in trunk length-at-age increases with age. These results suggest that statolith diameter is the better indicator of age at all body sizes and ages. Using statolith diameter as a proxy for age, trunk length-at-age of O. vanhoeffeni was observed in Conception Bay, Newfoundland, for two years. Seasonal variation in trunk length-at-age was observed in older age groups, and this variability increased with age, suggesting that body size is not a reliable indicator of age in eld populations. Length-at-age increased in older individuals during the spring phytoplankton bloom, suggesting variation of age-specic growth in relation to food availability. This new method to determine age of animals in the eld provides opportunities to dene age structure and age-specic life history characters that are essential for understanding the population dynamics of larvaceans. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Larvaceans are one of the most abundant zooplankton groups and have several important roles in marine ecosystems. Using their complex ltering structure called the house, these gelatinous pelagic tunicates are able to feed on a wide size range of particles ranging from DOM to large diatoms (Deibel, 1986; Flood et al., 1992; Urban- Rich et al., 2006). Because larvaceans are prey for many invertebrates and sh (Purcell et al., 2005), they efciently transfer energy within food webs by short-circuiting intermediate trophic links (Gorsky and Fenaux, 1998). In favorable conditions larvaceans can grow and multiply quickly, forming dense blooms that consume up to 5066% of the standing crop of phytoplankton daily (Alldredge, 1981; Deibel, 1988; Maar et al., 2004). The sinking of larvacean fecal pellets and discarded houses may transport a substantial portion of primary production to the benthos (Alldredge, 2005; Robison et al., 2005; Dagg et al., 2008). Although the distribution and abundance of larvaceans are well known, population dynamics have not been studied in depth. To better understand the population dynamics of this organism, demographic parameters need to be determined, in which age determination is a crucial factor. Age determination is essential for the study of population dynamics, which requires demographic information regarding age- specic fecundity and mortality as well as generation time (Stearns, 1992; Vandermeer and Goldberg, 2003). Determination of age is a major, general problem in the study of soft-bodied, directly develop- ing marine invertebrates (Duchêne and Bhaud, 1988; Lalli and Gilmer, 1989; Gordon et al., 2004). For marine invertebrates which lack discrete life history stages and have continuous growth, age is conventionally estimated based on body size. However, body size can be a poor indicator of age because size-at-age may vary depending on the rate of growth, which can be affected by temperature or food availability. For example, under severe food limitation, soft-bodied animals such as gelatinous zooplankton may experience de-growth as a result of using internal energy reserves and the reabsorption of somatic tissue to maintain metabolism (Hamner and Jensen, 1974; Kremer, 1976; Deason and Smayda, 1982). For these reasons, various chemical and physical characters have been used for age estimation of aquatic organisms, including lipofuscin concentration in neural tissue, or age increments and dimensions of permanently calcied structures such as otoliths and statoliths. Lipofuscin is autouorescent material that accumulates over time in lysosomes of postmitotic cells, such as neurons and cardiac myocytes, in various vertebrates and invertebrates (Porta, 1991; Yin, 1996; Terman and Brunk, 1998). Lipofuscin concentration has been particularly useful for aging aquatic crustaceans that lack permanently calcied structures (Sheehy and Wickins, 1994; Ju et al., 1999; Bluhm and Brey, 2001). Daily and annual incremental rings that form in the otoliths, statoliths, and shells of sh, cephalopods, gastropods, and bivalves are commonly used for age and growth rate estimates (Campana and Neilson, 1985; Lipinski, 1993; Sejr et al., 2002; Barroso et al., 2005). Dimensions of otoliths in sh have been found to increase with age. Journal of Experimental Marine Biology and Ecology 375 (2009) 8998 Corresponding author. E-mail address: [email protected] (N. Choe). 0022-0981/$ see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jembe.2009.05.013 Contents lists available at ScienceDirect Journal of Experimental Marine Biology and Ecology journal homepage: www.elsevier.com/locate/jembe

Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Embed Size (px)

Citation preview

Page 1: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Contents lists available at ScienceDirect

Journal of Experimental Marine Biology and Ecology

j ourna l homepage: www.e lsev ie r.com/ locate / jembe

Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni:Variability in age-specific growth patterns in Conception Bay, Newfoundland

Nami Choe ⁎, Don DeibelOcean Sciences Centre, Memorial University of Newfoundland, St. John's, Newfoundland, Canada A1C 5S7

⁎ Corresponding author.E-mail address: [email protected] (N. Choe).

0022-0981/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.jembe.2009.05.013

a b s t r a c t

a r t i c l e i n f o

Article history:Received 26 February 2009Received in revised form 21 May 2009Accepted 22 May 2009

Keywords:Age determinationLarvaceanLength-at-ageStatolith

We explored the use of statolith diameter as an age indicator for the larvacean Oikopleura vanhoeffeni.Laboratory studies indicated that variability in statolith diameter-at-age is substantially lower than that oftrunk length-at-age, and that variability in statolith diameter-at-age remains constant as statolith diameterincreases with age, while variability in trunk length-at-age increases with age. These results suggest thatstatolith diameter is the better indicator of age at all body sizes and ages. Using statolith diameter as a proxyfor age, trunk length-at-age of O. vanhoeffeni was observed in Conception Bay, Newfoundland, for two years.Seasonal variation in trunk length-at-age was observed in older age groups, and this variability increasedwith age, suggesting that body size is not a reliable indicator of age in field populations. Length-at-ageincreased in older individuals during the spring phytoplankton bloom, suggesting variation of age-specificgrowth in relation to food availability. This new method to determine age of animals in the field providesopportunities to define age structure and age-specific life history characters that are essential forunderstanding the population dynamics of larvaceans.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Larvaceans are one of the most abundant zooplankton groups andhave several important roles in marine ecosystems. Using theircomplex filtering structure called the ‘house’, these gelatinous pelagictunicates are able to feed on a wide size range of particles rangingfrom DOM to large diatoms (Deibel, 1986; Flood et al., 1992; Urban-Rich et al., 2006). Because larvaceans are prey for many invertebratesand fish (Purcell et al., 2005), they efficiently transfer energy withinfood webs by short-circuiting intermediate trophic links (Gorsky andFenaux, 1998). In favorable conditions larvaceans can grow andmultiply quickly, forming dense blooms that consume up to 50–66% ofthe standing crop of phytoplankton daily (Alldredge, 1981; Deibel,1988; Maar et al., 2004). The sinking of larvacean fecal pellets anddiscarded houses may transport a substantial portion of primaryproduction to the benthos (Alldredge, 2005; Robison et al., 2005; Dagget al., 2008). Although the distribution and abundance of larvaceansare well known, population dynamics have not been studied in depth.To better understand the population dynamics of this organism,demographic parameters need to be determined, in which agedetermination is a crucial factor.

Age determination is essential for the study of populationdynamics, which requires demographic information regarding age-specific fecundity and mortality as well as generation time (Stearns,1992; Vandermeer and Goldberg, 2003). Determination of age is a

ll rights reserved.

major, general problem in the study of soft-bodied, directly develop-ing marine invertebrates (Duchêne and Bhaud, 1988; Lalli and Gilmer,1989; Gordon et al., 2004). For marine invertebrates which lackdiscrete life history stages and have continuous growth, age isconventionally estimated based on body size. However, body sizecan be a poor indicator of age because size-at-agemay vary dependingon the rate of growth, which can be affected by temperature or foodavailability. For example, under severe food limitation, soft-bodiedanimals such as gelatinous zooplankton may experience de-growth asa result of using internal energy reserves and the reabsorption ofsomatic tissue to maintain metabolism (Hamner and Jensen, 1974;Kremer, 1976; Deason and Smayda, 1982). For these reasons, variouschemical and physical characters have been used for age estimation ofaquatic organisms, including lipofuscin concentration in neural tissue,or age increments and dimensions of permanently calcified structuressuch as otoliths and statoliths.

Lipofuscin is autofluorescent material that accumulates over timein lysosomes of postmitotic cells, such as neurons and cardiacmyocytes, in various vertebrates and invertebrates (Porta, 1991; Yin,1996; Terman and Brunk, 1998). Lipofuscin concentration has beenparticularly useful for aging aquatic crustaceans that lack permanentlycalcified structures (Sheehy and Wickins, 1994; Ju et al., 1999; Bluhmand Brey, 2001).

Daily and annual incremental rings that form in the otoliths,statoliths, and shells of fish, cephalopods, gastropods, and bivalves arecommonly used for age and growth rate estimates (Campana andNeilson, 1985; Lipinski, 1993; Sejr et al., 2002; Barroso et al., 2005).Dimensions of otoliths in fish have been found to increase with age.

Page 2: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

90 N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Otolith length, thickness and weight are strongly related to age inseveral fish species (Reznick et al., 1989; Kristoffersen and Klemetsen,1991; Newman, 2002). In some cases, otolith size increases in fish at aconstant rate even when food is limiting and somatic growth ratedecreases (Reznick et al., 1989; Secor et al., 1989; Campana, 1990). Ithas been shown in several fish species that the use of otolith weight-age relationships results in estimated age-frequency distributions thatare not significantly different from those determined from otolithincrement counts (Pilling et al., 2003; McDougall, 2004).

The search for a method of age determination of larvaceans is in anearly stage. So far, one method has been developed using endostylecell number as an index of age of Oikopleura dioica (Troedsson et al.,2007). The endostyle is a pharyngeal organ in larvaceans that secretesthe internal mucous net which collects particles during feeding

Fig.1. (A) Generalmorphology of the trunk ofOikopleura vanhoeffeni. (B) SEM view of the braincalcein under the confocal microscope. (D) SEM view of an isolated statolith. (E) View of a statanus, ‘bg’ buccal gland, ‘br’ brain, ‘en’ endostyle, ‘gd’ gonad, ‘mo’ mouth, ‘nv’ nerve, ‘oe’ oesop

(Olsson,1965; Deibel and Powell,1987). The number of endostyle cellsincreases with age, however, the rate of increase in cell number isdependent on temperature but is independent of food concentration(Troedsson et al., 2007). Thus, age of O. dioica can be predicted fromthe number of endostyle cells and ambient temperature (Troedsson etal., 2007). Lipofuscin concentration in the brain tissue of Oikopleuravanhoeffeni has been explored as an age indicator but no trace ofautofluorescent lipofuscin was detected (Choe, 2008). Daily incre-mental rings on the statolith of O. vanhoeffeni has also been exploredas a determinant of absolute age, but handling difficulties associatedwith the minute size of the statolith precluded proper samplepreparation for SEM and TEM analyses (Choe, 2008). In this study,we demonstrate that statolith diameter can be used as an ageindicator in O. vanhoeffeni.

and sensory vesicle. (C) Dorsal view of the brain, sensory vesicle and statolith stainedwitholith under the light microscope. Scale bar represents (A) 500 μm and (B)–(E) 15 μm. ‘an’hagus, ‘sl’ stomach lobe, ‘sp’ spiracle, ‘st’ statolith, ‘sv’ sensory vesicle.

Page 3: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Fig. 2. Map of Conception Bay, Newfoundland. ‘X’ indicates the sampling site at whichOikopleura vanhoeffeni were collected, and the dotted line indicates the 100 m isobath.The bottom depth at the sampling site was ca. 250 m.

91N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

2. Materials and methods

2.1. Viewing the statolith

The sensory organs of O. vanhoeffeni are located in the anterior–dorsal region of the trunk and consist of the brain and a sensoryvesicle containing the statolith (Fig. 1A). To obtain a closer 3D view ofthe sensory organ using scanning electron microscopy, O. vanhoeffeniwere preserved in 2% buffered glutaraldehyde. The animals weremounted on an SEM stub and dissected to expose the dorsal region ofthe trunk. The dissected trunks were sputter-coated with gold and thesamples were viewed at a magnification of 1500-times with a Hitachi5570 scanning electron microscope (Fig. 1B).

Dorsal sections of the sensory organ were viewed using confocalmicroscopy to confirmthepresence of the statolith in the sensoryvesicle(Fig. 1C). First, animals were incubated for 36 h in seawater (0–1 °C)with 500 mg l−1 calceinwhich binds to calciumwithin live cells and tothenewly formed layerof the statolith. Then the animalswerepreservedin 95%ethanol followedbydehydration in a gradedethanol/water seriesand cleared in xylene. The trunks of the animals were mounted on aglass slide and covered with a non-fluorescent mounting medium (GelMount aqueous mounting medium, Sigma). Specimens were viewedunderanOlympus Fluoview laser scanning confocalmicroscope (FV300laser scanning head) with an argon-ion laser (494 nm emission, 518 nmemission) to capture green fluorescence in the structures labeled withcalcein.

In order to view the entire statolith it was necessary to separate itfrom the sensory vesicle. The animals were preserved in 95% ethanoland were serially hydrated. Manual dissectionwas difficult because ofits minute size (ca. 8 to 18 μm), thus an indirect approachwas taken bydissolving the brain and sensory vesicle in sodium hypochlorite,leaving the statolith intact. Statoliths were rinsed with distilled waterand stained with alizarin red (calcium-specific) to aid visualization.Theywere then collected by filtering onto cellulose acetate membranefilters, which were mounted on a SEM stub and sputter-coated withgold. The samples were viewed at a magnification of 3500-times witha scanning electron microscope (Fig. 1D).

To view the statolith under transmitted light to measure statolithdiameter, animals were preserved in 95% ethanol, cleared in 1% KOHsolution andmounted in glycerol on a glass slide. Statolithswere viewedunder a Zeiss Axiovert 35 microscope at 1000-times magnification andthe statolith diameter was measured to the nearest 0.5 μm (Fig. 1E).

2.2. Growth of the statolith and trunk in the laboratory

Fully mature O. vanhoeffeni were collected from Logy Bay, New-foundland, from April–June 2001. They contained an orange mass ofsperm in the sperm sac and many oocytes packed in the ovary. Eachindividualwaskept suspended ina glass tank containing20 l of seawaterprefiltered through 40 μm mesh. The tanks were maintained at 0–1 °Cduring the experiment in a temperature-controlled room, and eachwasequippedwith a clear, twisted plexiglass paddlemounted vertically. Thepaddles were driven by a 12 V automobile windshield wiper motor(Volkswagen Motors AG, J. L. Acuña, pers. comm.). Rheostats wereconnected to the motors to regulate the speed of rotation. This setupcreated a gentle, rotating current of water which kept the animals insuspension (FenauxandGorsky,1985).Mature individuals spawnedandthe eggs hatched ca. two days after fertilization. The new generation ofanimals (G1) inflated their first houses ca. six days after spawning andbegan feeding (Choe, 2008). G1 animals within houses were moved tonew seawater prefiltered through 40 μmmesh, but supplemented with2×106 cells l−1 of Isochrysis sp. and 2×106 cells l−1 of Thalassiosirapseudonana. G1 animals were moved to new seawater with fresh foodevery two days. G1 animals were removed from the experimentrandomly at intervals from 10–60 days after hatching and werepreserved in 95% ethanol. Trunk lengths, excluding the gonad, were

measured to the nearest 25 μm under a Zeiss stereomicroscope at 40×magnification. Statolith diameter was measured as described above. Inorder to determine the growth pattern of statolith diameter and trunklength, linear and non-linear regressionswerefit to the time-series data.

In this paper we have assumed that statolith diameter is primarily afunction of chronological time and is relatively insensitive to changes intemperature and food concentration. Additional, complex laboratoryexperiments on the effects of variable temperature and food concentra-tion on body size and statolith diameter were beyond the scope of thisstudy, primarily because O. vanhoeffeni is one of the largest oikopleuridspecies in the world and because it is among the most stenothermal,cryophilic and stenohaline larvaceans yet known, preferring tempera-tures from−1.3 to 4.6 °C (Choe and Deibel, 2008). These physiologicalfactors make it extremely difficult to rear in the laboratory.

2.3. Field determination of trunk length-at-age and its relation totemperature and chlorophyll a concentration

Individual O. vanhoeffeni were collected from June 11, 2001 to June25, 2003ata site inConceptionBay,Newfoundland,with abottomdepthof ca. 235 m (47°32.2′N; 53°07.9′W, Fig. 2). Vertical hauls were made

Page 4: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

92 N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

from near the bottom (ca. 225 m) to the surface using a ring net with amesh size of 110 μm. The speed of retrieval of the net was ca. 0.12m s−1.Upon retrieval of the net, the entire sample was immediately preservedin 95% ethanol. The frequency of samplingwas bimonthly except duringwinter, when harsh weather conditions precluded sampling. On eachsampling day, in situ relative fluorescence wasmeasured with a SeabirdSBE 25-01 CTD equipped with a SEATEC fluorometer. Relativefluorescence units (RFU)were converted to chlorophyll a concentration(μg Chl a l−1) using the equation Chl a=0.398×RFU+0.281 (r2=0.72,n=244), which was developed using historical data from ConceptionBay (Cold Ocean Productivity Experiment, unpublished). Temperatureand chlorophyll a data were bin-averaged at 1 m depth intervals beforestatistical analysis and plotting.

Animals from the field samples were rinsedwith clean 95% ethanolbefore measuring trunk length and statolith diameter. Animals werenot rinsed with water before rinsing them with clean ethanol sincedoing so lowered the pH of the sample below 5.0, completelydissolving the statoliths. Trunk length was measured to the nearest

Fig. 3. (A) Statolith diameter of Oikopleura vanhoeffeni vs. age (days post hatch) at 0–1 °C inparents. The solid line shows the least squares linear regression and the dotted lines indicatemean statolith diameter. (C) The coefficient of variation of mean statolith diameter vs. age (dthe least squares linear regression and the dotted lines indicate 95% confidence intervals.coefficient of variation of mean trunk length vs. age.

25 μm under a Zeiss stereomicroscope at 40-times magnification,followed by clearing, mounting and measurement of statolith diameteras described above. Statolith diameters were converted to age using thecalibration curve from the laboratory study, and mean trunk length-at-age was calculated for age classes at 1-μm increments of statolithdiameter from9 to 16 μm(i.e. 8 age classes). Temporal variation inmeanlength-at-age for each of these age classes were described by fitting apolynomial regression (Y=a0+a1x+a2x

2+a3x3…). The order of best-

fit polynomial function to each data set was determined by fitting thesequential orders of polynomials until the sumsof squares for errorwereexplained significantly (Christensen, 1996). Relationships betweenmean trunk length-at-age and temperature and the concentration ofchlorophyll a were explored with a linear regression model. Meantemperature and chlorophyll a in the upper 100 m of the water columnwere used in this model because N70% of the animals were locatedwithin this depth stratum (Choe and Deibel, 2008). Trunk lengths werelog-transformed when residuals were not homogeneous and normallydistributed.

the laboratory. Different data symbols represent offspring from different self-fertilized95% confidence intervals. (B) The coefficient of variation of mean statolith diameter vs.ays post hatch). (D) Trunk length vs. age at 0–1 °C in the laboratory. The solid line shows(E) The coefficient of variation of mean trunk length vs. mean trunk length. (F) The

Page 5: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

93N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

3. Results

3.1. View of the statolith

The ventral view of the brain under SEM revealed that thesensory vesicle is connected to the brain by a pair of nerves(Fig. 1B). The confocal section of the central nervous system showedthat the statolith was located inside this sensory vesicle (Fig. 1C).SEM views of isolated statoliths suggested that the shape ofstatolith is cylindrical (Fig. 1D). The statolith viewed from the topunder the transmitted light was circular with several concentricrings (Fig. 1E).

Fig. 4. Mean trunk length-at-age of Oikopleura vanhoeffeni vs. time from June 2001 to June 20age using laboratory estimates from Fig. 3A. Lines represent polynomial regression of best-fitfrom 57 to 65 days post hatch.

3.2. Relationships between statolith diameter, trunk length and age inlaboratory-reared individuals

Statolith diameter of O. vanhoeffeni increased linearly for over60 days after hatching in the laboratory (Fig. 3A). The relationshipbetween statolith diameter and age does not appear to be dependenton the genetic origin of individuals because each of the data points inFig. 3A comes from the G1 progeny of a single parent. Mostimportantly, the coefficient of variation of mean statolith diameter-at-age did not increase with an increase in mean statolith diameter(Fig. 3B) or with an increase in age (Fig. 3C), meaning that statolithdiameter is a robust age indicator for individuals of all sizes and ages.

03 in Conception Bay for 8 age classes derived from conversion of statolith diameter to-orders. No significant polynomial functionwas found for the mean trunk length-at-age

Page 6: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Fig. 6. Time-depth profiles of (A) temperature and (B) concentration of chlorophyll a inConception Bay from June 2001 to June 2003. The irregular intervals in the chlorophyllprofile are due to a rounding effect. The arrows beneath the X axis indicate when CTDcasts were made.

94 N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Trunk length of the laboratory population increased linearly overtime (Fig. 3D). However, the coefficient of variation of mean trunklength-at-age showed an increasing trend with an increase in meantrunk length (Fig. 3E) and with an increase in age (Fig. 3F), suggestingthat trunk length is a less precise indicator of the age of older,larger individuals than of younger, smaller individuals. In addition,the overall coefficient of variation for mean statolith diameter-at-age(6–11%) was significantly less than that for mean trunk length-at-age(8–27%) (t-test, t (20)=−7.08, pb0.001). Thus, statolith diameterwas a more precise indicator of age for individuals of all sizes and agesthan was body size, even in laboratory conditions where temperatureand food concentration were controlled.

3.3. Trunk length-at-age in field populations

Using the laboratory calibration equation (Fig. 3A), mean trunklength-at-age of O. vanhoeffeni in Conception Bay was estimated for ageclasses from 2 days to 65 days post hatching at an interval of 7–8 days(i.e. 8 age classes corresponding to 1 μm bins of statolith diameter).These age groups were present year round, enabling examination oftheir temporal pattern. Polynomial regression analysis of best-fit orderson the time series of mean trunk length-at-age over the years 2001 to2003 revealed no clear seasonal variation at age classes b23 days old,but a distinct seasonal variation at age classes from 24 to 56 days old(Fig. 4). In general, maximum mean length-at-age of these older ageclasses occurred in spring and minimal values in summer (Fig. 4). Thisseasonal pattern of length-at-age in the older individuals repeated inthe two consecutive years of observation.

Temporal variability in trunk length-at-age, expressed as thecoefficient of variation of the grand mean trunk length for each ageclass over the two year period, fluctuated from 8–10% for the ageclasses b31 days old, but increased linearly up to 31% for the ageclasses from 32 to 65 days old (Fig. 5), suggesting that temporalvariability in length-at-age increased with age.

3.4. Temperature and chlorophyll a in Conception Bay

Temperature fluctuated seasonally in the upper 10 m with anincrease to amaximum of 15.4–16.6 °C in late August and a decrease toa minimum of−1.0 to−0.8 °C in late March to early April (Fig. 6A). Athermocline developed within the upper 60 m from June to Octoberand retreated as winter mixing occurred to a depth of 100 to 150 m.Temperature below 150 m remained b0 °C throughout the study.Seasonal variation in chlorophyll a concentration occurred mostlywithin the upper 100 m (Fig. 6B). The spring bloom began in March

Fig. 5. Temporal variability in trunk length-at-age in Conception Bay, expressed as thecoefficient of variation of the grand mean trunk length-at-age for each of the 8 ageclasses over the entire 2 year study period.

and peaked in May with a maximum chlorophyll a concentration of5.8 μg l−1 in 2002 and 3.5 μg l−1 in 2003. A minor bloom occurred inAugust 2001 (2.4 μg l−1) and a weaker bloom occurred in late July2002 (1.7 μg l−1). Minimum concentrations of chlorophyll a werefound in July 2001 (0.89 μg l−1) and in October 2002 (0.98 μg l−1).

3.5. Relationships between length-at-age and temperature andchlorophyll a in field populations

Mean trunk length-at-age did not vary significantly with tem-perature for the younger age classes but increased significantly astemperature decreased for the three oldest age classes (Fig. 7).Similarly, no significant relationship existed between mean trunklength-at-age and chlorophyll a concentration in the younger ageclasses, but a significant increase was observed with increasingchlorophyll concentration for the five oldest of the eight age classes(Fig. 8).

4. Discussion

The objective of this study was to explore the use of statolithdiameter for age determination of the larvacean O. vanhoeffeni, aspecies commonly found in high concentrations in Arctic and borealwaters (Udvardy, 1954; Shiga, 1993; Ashjian et al., 1997), where it isoften a dominant suspension feeder (Deibel, 1988; Acuña et al., 2002).We demonstrated that statolith diameter is a more precise andaccurate age proxy than is body size. Using statolith diameter as an ageindicator, trunk length-at-age of O. vanhoeffeni in Conception Bay,Newfoundland was examined over a two year period. We discoveredthat seasonal variation in length-at-age was clearly related to foodavailability.

Page 7: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Fig. 7. Mean trunk length-at-age vs. temperature for 8 age classes of Oikopleura vanhoeffeni from June 2001 to June 2003. Temperature was averaged over the upper 100 m of thewater column.

95N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Increase in statolith diameter over time was monotonic withconstant variance under controlled temperature and food concentra-tion in the laboratory. This result was independent of the geneticorigin of the individuals. In contrast, the variance of mean body sizewas not stable over time but increased as the individuals aged, evenunder fixed conditions of temperature and food concentration in thelaboratory, suggesting that increasing variability in body size with ageis an inherent character of O. vanhoeffeni. Furthermore, the overallvariance of statolith diameter-at-age was less than that of body size-at-age, suggesting that statolith diameter is themore precise indicatorof age over all life history stages.

Trunk length-at-age of O. vanhoeffeni in Conception Bay variedannually over the two year study period, particularly in the older agegroups (Fig. 4). Mean length-at-age increased two to four fold fromfall to spring in groups from 41–65 days old (Fig. 4), and temporalvariability in mean length-at-age increasedwith age (Fig. 5). This highand inconsistent variability in length-at-age suggests that body size isnot a reliable age indicator in field populations, perhaps due toseasonally dependent periods of accelerated growth or de-growth.Therefore, using the conventional method of cohort separation basedon modal progression analysis of length frequency data may lead toinaccurate estimation of the population age distribution.

Page 8: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

Fig. 8. Mean trunk length-at-age vs. the concentration of chlorophyll a for 8 age classes of Oikopleura vanhoeffeni from June 2001 to June 2003. Concentration of chlorophyll a wasaveraged over the upper 100 m of the water column.

96 N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Polynomial regression analysis of the time series of trunk length-at-age of O. vanhoeffeni revealed a distinct seasonal cycle in the olderindividuals (24–56 days post hatching) but an absence of seasonalpattern in the younger individuals (b24 days old) (Fig. 4), indicatingage-specific variation in growth. If temperature affected the growth ofO. vanhoeffeni as previous studies of other larvaceans have shown(Lopéz-Urrutia et al., 2003), length-at-age of O. vanhoeffeni wouldhave increased in the summer as the temperature increased. However,a negative relationship between length-at-age and temperature wasfound in O. vanhoeffeni (Fig. 7) suggesting that either the growth was

not affected by the temperature range experienced by the animals orother factors explained the temporal variation in age-specific growth.

The increase in the trunk length-at-age in older age groups duringthe spring bloom (Fig. 4), and a significant positive relationshipbetween the trunk length-at-age and chlorophyll concentration(Fig. 8), suggest that individual growth rates within naturallyoccurring populations of O. vanhoeffeni may be positively related tofood concentration. One possible explanation for the increase in age-specific growth in older individuals in the spring is that older andlarger animals with larger inlet filters on the house may be better able

Page 9: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

97N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

to ingest large diatoms during the spring bloom (Deibel and Turner,1985; Deibel, 1988; Urban et al., 1992). In Conception Bay, nano- andpicoplankton predominate during the fall bloom and larger diatomsduring the spring bloom (Urban et al., 1992; Putland, 2000). Inaddition, studies have shown that inclusion of diatoms that are rich inessential fatty acids in the diet of marine organisms enhances growth(Parrish et al., 1998; Pond et al., 2005; Jones and Flynn, 2005).

Because sample preparation and measurement of statolith dia-meter are relatively simple procedures, use of trunk length-at-age as agrowth index of larvaceans can be practical in field studies and easierand less costly than other growth indices such as RNA content(Sutcliffe, 1970; Båmstedt and Skjoldal, 1980) and RNA/DNA ratio(Saiz et al., 1998; Vrede et al., 2002). The current method ofmeasurement of statolith diameter involves clearing the tissue ofeach animal and mounting several animals on glass slides withglycerol. To define the age structure based on statolith diameterdistribution from many samples in a large scale field study, it ispossible to expedite the procedure by dissolving the trunks of animalsto isolate the statoliths and mounting many statoliths on a slide formeasurements with a microscope. It is possible that this procedurecould then be automated using image analysis.

In conclusion, this study demonstrates for the first time thatstatolith diameter is a precise and accurate indicator of age for thelarvacean O. vanhoeffeni. We found that; (1) length-at-age of fieldpopulations exhibits seasonal variation, indicating temporal variation ingrowth; (2) variation in length-at-age is related to food availability;(3) due to high variability in length-at-age both in controlled laboratoryconditions and in field populations, body size is not a reliable indicatorof age. Thus, age structure in nature can be defined more accuratelyfrom the distribution of statolith diameter than from the distribution ofbody size, and age-specific life history traits can be determined usingstatolith diameter as an age proxy, a finding that has importantimplications for the study of the population dynamics of larvaceans.

Acknowledgements

We thank the OSC divers, captain and crew of RV Karl & Jackie andMVMares for collection of animals and OSC technical staff for buildingthe culture setups. R. Sato and J-L Acuña provided advice on culturingtechniques of appendicularians and M. Sheehy made valuablesuggestions on lipofuscin studies. L. Lee and M. Goldsworthy assistedwith the SEM and confocal microscopical studies. R. Thompson, P.Snelgrove and two anonymous reviewers provided critical commentsthat improved an earlier version of the manuscript. This research wasfunded by an NSERC Discovery Grant to D. Deibel and a GraduateStudent Fellowship from Memorial University. [SS]

References

Acuña, J.L., Deibel, D., Saunders, P.A., Booth, B., Hatfield, E., Klein, B., Mei, Z.-P., Rivkin, R.,2002. Phytoplankton ingestion by appendicularians in the North Water. Deep SeaRes. II 49, 5101–5115.

Alldredge, A.L., 1981. The impact of appendicularian grazing on natural foodconcentrations in situ. Limnol. Oceanogr. 26, 247–257.

Alldredge, A.L., 2005. The contribution of discarded appendicularian houses to the fluxof particulate organic carbon from ocean surface waters. In: Gorsky, G., Youngbluth,M.J., Deibel, D. (Eds.), Response of Marine Ecosystems to Global Change: EcologicalImpact of Appendicularians. Contemporary Publishing International, Paris, France,pp. 309–326.

Ashjian, C., Smith, S., Bignami, F., Hopkins, T., Lane, P., 1997. Distribution of zooplanktonin the Northeast Water Polynya during summer 1992. J. Mar. Syst. 10, 279–298.

Båmstedt, U., Skjoldal, H.R., 1980. RNA concentration of zooplankton: relationship withsize and growth. Limnol. Oceanogr. 25, 304–316.

Barroso, C.M., Nunes, M., Richardson, C.A., Moreira, M.H., 2005. The gastropod statolith:a tool for determining the age of Nassarus reticulatus. Mar. Biol. 146, 1139–1144.

Bluhm, B.A., Brey, T., 2001. Age determination in the Antarctic shrimp Notocrangonantarcticus (Crustacea: Decapoda), using the autofluorescent pigment lipofuscin.Mar. Biol. 138, 247–257.

Campana, S.E., 1990. How reliable are growth back-calculations based on otoliths? Can. J.Fish. Aquat. Sci. 47, 2219–2227.

Campana, S.E., Neilson, J.D., 1985. Microstructure of fish otoliths. Can. J. Fish. Aquat. Sci.42, 1014–1032.

Choe, N., 2008. Population dynamics and life history characters of boreal appendicularianspecies in Conception Bay, Newfoundland. PhD thesis, Memorial University ofNewfoundland, Canada.

Choe, N., Deibel, D., 2008. Temporal and vertical distributions of three appendicularianspecies (Tunicata) in Conception Bay, Newfoundland. J. Plankton Res. 30, 969–979.

Christensen, R., 1996. Analysis of Variance, Design and Regression: Applied StatisticalMethods. Chapman and Hill, London, UK, pp. 204–222.

Dagg, M., Sato, R., Liu, H., Bianchi, T.S., Green, R., Powell, R., 2008. Microbial food webcontributions to bottomwater hypoxia in the northern Gulf of Mexico. Cont. Shelf.Res. 28, 1127–1137.

Deason, E.E., Smayda, T.J., 1982. Experimental evaluation of herbivory in the ctenophoreMnemiopsis leidyi relevant to the ctenophore–zooplankton–phytoplankton inter-actions in Narragansett Bay, Rhode Island. J. Plankton Res. 4, 219–236.

Deibel, D., 1986. Feeding mechanism and house of the appendicularian Oikopleuravanhoeffeni. Mar. Biol. 93, 429–436.

Deibel, D., 1988. Filter feeding by Oikopleura vanhoeffeni: grazing impact on suspendedparticles in cold ocean waters. Mar. Biol. 99, 177–186.

Deibel, D., Powell, C.V.L., 1987. Ultrastructure of the pharyngeal filter of theappendicularian Oikopleura vanhoeffeni: implications for particle size selectionand fluid mechanics. Mar. Ecol. Prog. Ser. 35, 243–250.

Deibel, D., Turner, J.T., 1985. Zooplankton feeding ecology: contents of fecal pellets of theappendicularian Oikopleura vanhoeffeni. Mar. Ecol. Prog. Ser. 27, 67–78.

Duchêne, J.C., Bhaud, M., 1988. Uncinial patterns and age determination in terebellidpolychaetes. Mar. Ecol. Prog. Ser. 49, 267–275.

Fenaux, R., Gorsky, G., 1985. Nouvelle technique d'elevage des Appendiculaires. Rapp.Comm. Int. Mer. Medit. 29, 291–292.

Flood, P.R., Deibel, D., Morris, C.C., 1992. Filtration of colloidal melanin from seawater byplanktonic tunicates. Nature 355, 630–632.

Gordon, M., Hatcher, C., Seymour, J., 2004. Growth and age determination of the tropicalAustralian cubozoan Chiropsalmus sp, 530/531. Hydrobiolgia, pp. 339–345.

Gorsky,G., Fenaux, R.,1998. The roleofAppendicularia inmarine foodwebs. In: Bone,Q. (Ed.),The Biology of Pelagic Tunicates. Oxford University Press, Oxford, UK, pp. 161–170.

Hamner, W.M., Jensen, R.M., 1974. Growth, degrowth and irreversible cell differentia-tion in Aurelia aurita. Am. Zool. 14, 833–849.

Jones, R.H., Flynn, K.J., 2005. Nutritional status and diet composition affect the value ofdiatoms as copepod prey. Science 307, 1457–1459.

Ju, S.J., Secor, D.H., Harvey, R., 1999. Use of extractable lipofuscin for age determinationof blue crab Callinectes sapidus. Mar. Ecol. Prog. Ser. 185, 171–179.

Kremer, P.,1976. Population dynamics and ecological energetics of a pulsed zooplanktonpredator, the ctenophoreMnemiopsis leidyi. In: Wiley, M. (Ed.), Estuarine Processes,vol. 1. Academic Press, New York, pp. 197–215.

Kristoffersen, K., Klemetsen, A., 1991. Age determination of Arctic charr (Salvelinusalpinus) from surface and cross section of otoliths related to otolith growth. Nord. J.Freshw. Res. 66, 98–107.

Lalli, C.M., Gilmer, R.W., 1989. The thecosomes: shelled pteropods. Pelagic snails: TheBiology of Holoplanktonic Gastropod Mollusks. Stanford University Press, Stanford,California, p. 119.

Lipinski,M.R.,1993. Thedepositionof statoliths: aworkinghypothesis. In:Okutani, T., ODor,R.K., Kubodera, T. (Eds.), Recent Advances in Cephalopod Fisheries Biology. TakaiUniversity Press, Tokyo, pp. 241–262.

Lopéz-Urrutia, Á., Acuña, J.L., Irigoien, X., Harris, R., 2003. Food limitation and growth intemperate epipelagic appendicularians (Tunicata). Mar. Ecol. Prog. Ser. 252, 143–157.

Maar, M., Nielsen, T.G., Gooding, S., Tönnesson, K., Tiselius, P., Zervoudaki, S., Christou, E.,Sell, A., Richardson, K., 2004. Trophodynamic function of copepods, appendiculariansand protozooplankton in the late summer zooplankton community in Skagerrak. Mar.Biol. 144, 917–933.

McDougall, A., 2004. Assessing the use of sectioned otoliths and other methods todetermine the age of the centropomid fish, barramundi (Lates calcarifer) (Bloch)using known-age fish. Fish. Bull. 67, 129–141.

Newman, S.J., 2002. Growth rate, age determination, natural mortality and productionpotential of the scarlet seaperch, Lutjanus malabaricus Schneider 1801, off thePilbane coast of north-western Australia. Fish. Res. 58, 215–225.

Olsson, R., 1965. Cytology of endostyle of Oikopleura dioica. Ann. N.Y. Acad. Sci. 118,1038–1051.

Parrish, C.C., Wells, J.S., Yang, Z., Dabinett, P., 1998. Growth and lipid composition ofscallop juveniles, Placopecten magellanicus, fed the flagellate Isochrysis galbanawith varying lipid composition and the diatom Chaetoceros muelleri. Mar. Biol. 133,461–471.

Pilling, G.M., Grancourt, E.M., Kirkwood, G.P., 2003. The utility of otolith weight as apredictor of age in the emperor Lethrinus mahsena and other tropical fish species.Fish. Res. 60, 493–506.

Pond, D.W., Atkinson, A., Shreeve, R.S., Tarling, G., Ward, P., 2005. Diatom fatty acidbiomarkers indicate recent growth rates in Antarctic krill. Limnol. Oceanogr. 50,732–736.

Porta, E.A.,1991. Advances in age pigment research. Arch. Gerontol. Geriatr.12, 303–320.Purcell, J.E., Sturdvant, M.V., Galt, C.P., 2005. A review of appendicularians as prey of

invertebrate and fish predators. In: Gorsky, G., Youngbluth, M.J., Deibel, D. (Eds.),Response of Marine Ecosystems to Global Change: Ecological Impact of Appendicular-ians. Contemporary Publishing International, Paris, France, pp. 359–435.

Putland, J.N., 2000. Microzooplankton herbivory and bacteriovory in Newfoundlandcoastal waters during spring, summer and winter. J. Plankton Res. 22, 253–277.

Reznick, D., Lindbeck, E., Bryga, H., 1989. Slower growth results in larger otoliths: anexperimental test with guppies (Poecilia reticulata). Can. J. Fish. Aquat. Sci. 46,108–112.

Page 10: Statolith diameter as an age indicator in the planktonic tunicate Oikopleura vanhoeffeni: Variability in age-specific growth patterns in Conception Bay, Newfoundland

98 N. Choe, D. Deibel / Journal of Experimental Marine Biology and Ecology 375 (2009) 89–98

Robison, B.H., Reisenbichler, K.R., Sherlock, R.E., 2005. Giant larvacean houses: rapidcarbon transport to the deep sea floor. Science 308, 1609–1611.

Saiz, E., Calbet, A., Fara, A., Berdalet, E., 1998. RNA content of copepods as a tool fordetermining adult growth rates in the field. Limnol. Oceanogr. 43, 465–470.

Secor, D.H., Dean, J.M., Baldevarona, R.B., 1989. Comparison of otolith growth and somaticgrowth in larval and juvenile fishes based on otolith length/fish length relationships.ICES Marine Science Symposia, Copenhagen, Denmark, vol. 191, pp. 431–438.

Sejr, M.K., Jensen, K.T., Rysgaard, S., 2002. Annual growth bands in the bivalve Hiatellaarctica validated by a mark-recapture study in NE Greenland. Polar Biol. 25,794–796.

Sheehy, M.R.J., Wickins, J.F., 1994. Lipofuscin age pigment in the brain of the Europeanlobster Homarus gammarus (L.). Microsc. Anal. 12, 23–25.

Shiga, N., 1993. Regional and vertical distributions of Oikopleura vanhoeffeni on thenorthern Bering Sea shelf in summer. Bull. Plankton Soc. Jpn. 39, 117–126.

Stearns, S.C., 1992. The Evolution of Life Histories. Oxford University Press, Oxford, UK.Sutcliffe, W.H., 1970. Relationship between growth rate and ribonucleic acid

concentration in some invertebrates. J. Fish. Res. Board Can. 27, 606–609.Terman, A., Brunk, U.T., 1998. Ceroid/lipofuscin formation in cultured human

fibroblasts: the role of oxidative stress and lysosomal proteolysis. Mech. AgeingDev. 104, 277–291.

Troedsson, C., Ganot, P., Bouquet, J.-M., Aksnes, D.L., Thompson, E.M., 2007. Endostylecell recruitment as a frame of reference for development and growth in theurochordate Oikopleura dioica. Biol. Bull. 213, 325–334.

Udvardy, M.D.F., 1954. Distribution of appendicularians in relation to the Strait of BelleIsle. J. Fish. Res. Board Can. 11, 431–453.

Urban, J.L., McKenzie, C.H., Deibel, D., 1992. Seasonal differences in the content of Oi-kopleura vanhoeffeni and Calanus finmarchicus fecal pellets: illustrations of planktonfood web shifts in coastal Newfoundland waters. Mar. Ecol. Prog. Ser. 84, 255–264.

Urban-Rich, J., Fernández, D., Acuña, J.L., 2006. Grazing impact on chromorphicdissolved organic matter (CDOM) by the larvacean Oikopleura dioica. Mar. Ecol. Pro.Ser. 317, 101–110.

Vandermeer, J.H., Goldberg, D.E., 2003. Population Ecology: First Principles. PrincetonUniversity Press, Princeton, NJ.

Vrede, T., Persson, J., Aronsen, G., 2002. The influence of food quality (P:C ratio) on RNA:DNA ratio and somatic growth rate of Daphnia. Limnol. Oceanogr. 47, 487–494.

Yin, D., 1996. Biochemical basis of lipofuscin, ceroid, and age pigment-like fluorophores.Free Radical Bio. Med. 21, 871–888.