COMPARISON OF BIOCHEMICAL COMPOSITION AND

DEVELOPMENTAL MODE IN TWO POPULATIONS OF

COSTASIELLA [OPISTHOBRANCHIA: ASCOGLOSSA
( SACOGLOSSA)]
CECELIA M. MILES 1 AND KERRY B. CLARK†
Florida Institute of Technology, Department of Biological Sciences, Melbourne, FL 32901, USA
1
Present address: University of Florida, Department of Zoology, Gainesville, FL 32611, USA
(Received 20 February 2001; accepted 27 August 2001)

ABSTRACT
Egg masses of two populations of the ascoglossan Costasiella with different developmental modes were
examined. Costasiella ocellifera is an encapsulated developer; Costasiella sp. has an obligate plankto-
trophic larval stage. Adults of the two populations were also separable by habitat, but could not be dis-
tinguished by external morphology. Comparison of capsule diameters and calorimetric analyses
showed that C. ocellifera eggs have nearly eight times the capsule volume and 10 times the calories of
eggs of its congener. Histochemical examination of extra-embryonic intra-capsular vesicles of C. ocel-
lifera revealed the contents to be glycoprotein. No inclusions were visible in the capsular fluid of Costa-
siella sp. Egg masses from the two populations differed significantly in the amount of TCA-soluble
carbohydrate, lipid, and NaOH-soluble protein per egg and per milligram dry weight of egg mass.
These components were, however, similar when expressed on an ash-free dry-weight basis. The advant-
age of hatching directly onto a suitable food source that is temporally persistent, but patchily dis-
tributed might have provided the selective pressure to achieve extended intra-capsular development
in C. ocellifera.

INTRODUCTION Types 2 and 3, allowing growth of these smaller eggs into large
larvae or juveniles before hatching.
Pattern of embryonic development is a fundamental character- Use of intra-capsular material as a nutrient source has
istic of marine invertebrate life histories. The life history of a been demonstrated in gastropods, particularly in prosobranchs
species represents the outcome of many interacting selective (e.g. de Mahieu, Penchaszadeh & Casal, 1974; Fioroni, 1977;
pressures. Opisthobranchs (Mollusca: Gastropoda) exhibit Stöckmann-Bosbach & Althoff, 1989; Rivest, 1992; Sautto-
diverse morphological and physiological specializations that Vallejo, 1992; Rivest & Strathmann, 1995; Penchaszadeh &
have enabled adaptation of this group to a wide range of Rincon, 1996; Miloslavich, 1999; Moran, 1999). The existing
environmental conditions. Life-history traits that display high literature on intra-capsular fluids and capsule ultrastructure of
variability in the Opisthobranchia include patterns of develop- nudibranchs was reviewed by Eyster (1986). However, com-
ment (indirect, modified, or direct), egg size, egg number, and pared to other molluscan taxa, little histochemical or bio-
organic composition and caloric content of the egg (Thompson, chemical work on intra-capsular fluids has been done on
1967; Eyster, 1980; DeFreese & Clark, 1983; Hadfield & Switzer- opisthobranchs, and none has been done on the Ascoglossa.
Dunlap, 1984; Hadfield & Miller, 1987). The intent of the present work was to quantify the parental
Thompson (1967) recognized three developmental patterns investment of specific organic resources by comparing the eggs
among the Opisthobranchia. Species with Type-1 development and egg masses of two congeneric ascoglossan molluscs that
have planktotrophic veliger larvae. Species with Type-2 have different reproductive modes. The two populations under
development have lecithotrophic veliger larvae, and those of consideration here are separable by habitat (Figure 1) and
Type 3 have direct encapsulated development. Bonar (1978) reproductive pattern, but are indistinguishable based on ex-
modified Thompson’s scheme by separating Type-3 species ternal morphology. Costasiella ocellifera (Simroth, 1895) is a
into capsular metamorphic (hereafter Type 3a, exhibiting Type-3a developer (DeFreese & Clark, 1983) found at Geiger Key
typical veliger morphology within the capsule) and capsular in the Lower Florida Keys, Florida, USA. Costasiella cf. ocellifera
ametamorphic development (3b, with veliger structures absent, (hereafter Costasiella sp.) is a Type-1 developer (Clark, 1994, as
or only briefly and vestigially present). Costasiella n. sp.) that lives in Lake Surprise, Key Largo, in the
Egg size is a fairly reliable predictor of developmental pat- Upper Florida Keys.
tern for opisthobranchs in general (Clark & Jensen, 1981; Had- Comparison of the composition and location of resources
field & Miller, 1987). Clark & Jensen (1981) found, however, provided for the developing embryos might provide insight
that the limits of egg sizes for Ascoglossa ( Sacoglossa) exhibit- into the differences in life histories of these two congeners.
ing Type-2 and Type-3 development are substantially smaller Ingestion of intra-capsular albumen represents a way to extend
than the limits established by Thompson (1967) for opistho- embryonic development without the complications of increased
branchs in these developmental classes. They observed the amount of yolk. Specific objectives of this work were:
presence of intra-capsular albumen in all three developmental
types in a survey of 32 species of Ascoglossa and hypothesized (1) to measure egg and capsule diameters for each slug taxon
that albumen provides a nutritional supplement for animals of and to calculate their volumes;
†Deceased. (2) to compile a developmental time line for each slug taxon;

J. Moll. Stud. (2002), 68, 101–109 © The Malacological Society of London 2002
 C. M. MILES & K. B. CLARK

(3) to determine both relative (per mg) and absolute (per (24°34.4 N, 81°39.8 W), Florida, USA, on 30 March 1996.
egg) calorimetric values for egg masses of each slug Geiger Key is a euhaline coral-sand habitat, well flushed tidally,
taxon; with oligotrophic to mesotrophic waters. The coarse, well-
(4) to examine sectioned egg masses of each slug taxon to oxidized carbonate sand overlies a limestone base. Animals
identify and locate nutritional components of the capsular reproduced successfully and were maintained in culture in
fluid and ovum; and sub-gravel-filtered marine aquaria, at approximately 25°C, with
(5) to assay for relative (per mg) and absolute (per egg) a photoperiod of 18 h light: 6 h dark, through December 1999.
amounts of carbohydrate, lipid, and soluble protein Living individuals of Costasiella sp. were collected in the same
present in egg masses of each slug taxon. manner every 4–6 weeks from March 1996 to December 1999
at Lake Surprise (25°10.9 N, 80°23.1 W), Key Largo, Florida.
Lake Surprise is a polyhaline mangrove fringe habitat with
MATERIAL AND METHODS
restricted tidal flushing and mesotrophic to eutrophic waters.
Living individuals of Costasiella ocellifera were collected using Sediments are partly organic, partly calcareous, with some shell
a hand-held suction collector (Clark, 1971) at Geiger Key chaff. Specimens of Costasiella sp. were maintained in separate

Figure 1. Collection sites (arrows) for two populations of Costasiella from the Florida Keys, Florida, USA. (A) Lake Surprise, Key Largo. (B) Geiger Key.

102
 COSTASIELLA EGG MASSES

1 aquaria from, but otherwise under the same culture conditions microtome cryostat. Egg masses were fixed for 1 h in 10%
2 as C. ocellifera. Repeated collection of Costasiella sp. was neces- formalin in filtered seawater, mounted in HistoPrep frozen
3 sary because the planktotrophic larval stage made laboratory tissue embedding media (Fisher Scientific Co., Pittsburgh,
4 culture through metamorphosis difficult. Voucher specimens Pennsylvania), and sectioned at 18 m. Sections were stained
5 from both populations are deposited at the American Museum with bromine and Sudan Black B (Bayliss & Adams, 1972).
6 of Natural History (AMNH 232645, Costasiella ocellifera, 10 speci- Again, no reaction was observed on control slides that had
7 mens; AMNH 232646, Costasiella cf. ocellifera, 10 specimens). been delipidized with a mixture of chloroform and methanol at
8 Both populations are trophically specialized, feeding only on 2:1 (v/v) concentration (High, 1984) before staining.
9 the green alga Avrainvillea nigricans Decaisne 1842 (Bryopsi- Protein was detected after egg masses were fixed in 10%
10 dales: Udoteaceae). The Udoteaceae have a haplobiontic life formalin in filtered seawater for 2 h, embedded in LR White,
1 cycle (Silva, 1982; Lobban & Harrison, 1994), and the range of and sectioned at 5 m. Sections were stained with Ninhydrin–
2 this species includes the Gulf of Mexico, Caribbean, East Schiff (Yasuma & Ichikawa, 1953). No reaction for proteins was
3 Africa, Micronesia, Philippines, and western Polynesia (Olsen- observed in control slides that had been exposed to pepsin
4 Stojkovich, 1985). The two populations of Costasiella share the digestion (3 mg pepsin in 100 ml 0.02N hydrochloric acid) for
5 further specialization of kleptoplasty, exhibiting long-term 2 h (Yasuma & Ichikawa, 1953).
6 retention of functional algal chloroplasts (level 6 kleptoplasty; Calorimetric analysis was accomplished using a Phillipson
7 Clark, Jensen & Stirts, 1990) ingested from A. nigricans as a microbomb calorimeter calibrated with a benzoic acid standard
8 supplemental energy source. Avrainvillea nigricans was collected (Paine, 1964; Phillipson, 1964). Mineral oil was used in micro-
9 on the monthly trips to Lake Surprise and the algae collected litre quantities as an organic carrier of high caloric content.
20 from this one site was provided ad libitum to both laboratory Ash content was measured by ashing dried samples in a muffle
1 populations. furnace at 500°C for 4 h (Paine, 1964).
2 A razor blade was used to remove egg masses directly from Carbohydrate content of egg masses was estimated using the
3 the A. nigricans blades on which they were deposited. With the method of Dubois, Gilles, Hamilton, Rebers & Smith (1956) on
4 exception of the sulphophosphovanillin test for lipids, samples individual egg masses from each population. Carbohydrates
5 for biochemical, calorimetric, and ash analyses were dried to a were extracted in 5% (w/v) trichloroacetic acid in a 100°C
6 constant weight at 80°C and stored in a vacuum desiccator over water bath for 1 h. D-glucose was used to generate the standard
7 anhydrous CaSO4 until immediately before the tests were per- curve, and absorbance was read at 490 nm on a spectro-
8 formed. Samples for lipid testing were lyophilized for 24 h to photometer.
9 reduce loss of volatile lipids from heating. The contribution of The sulphophosphovanillin method (Barnes & Blackstock,
30 eggs ( ova, zygotes, embryos), capsular walls, capsular fluid, 1973) was used to estimate total lipids after lyophilizing indi-
1 and extra-capsular jelly were not estimated separately, and were, vidual egg masses from each population for 24 h. Lipids were
2 therefore, combined in results reported as ‘per-egg’ values for extracted in 1:1 (v/v) chloroform-methanol for 30 min at
3 calorimetry, ash, and biochemical constituents. These indi- 40–60°C. Cholesterol was used as the standard, and absorbance
4 vidual components probably vary in contribution with differing was read at 520 nm at 30–60 min after adding phosphovanillin
5 habitats (DeFreese & Clark, 1983) and can be important to reagent.
6 larval fitness (Todd, 1979). Morphometric and histochemical As recommended by Barnes & Blackstock (1973), the accu-
7 analyses were performed on egg masses with uncleaved fertil- racy of the sulphophosphovanillin method was checked by
8 ized eggs. Calorimetric and biochemical analyses were per- using the gravimetric technique of Bligh & Dyer (1959). The
9 formed on egg masses with embryos in the earliest stages of egg masses of Costasiella spp. are too small to use gravimetric
40 development (eight-cell maximum). Calorimetric analysis and techniques. Egg masses of Aplysia californica were obtained
1 determination of ash content could not be performed on indi- from Aplysia Research Facility (University of Miami, Miami,
2 vidual egg masses and pooled samples were needed, but for all Florida), and the colorimetric technique was run simul-
3 other procedures individual egg masses were used. taneously with the gravimetric technique. Aplysia egg masses
4 Measurements of egg and capsule diameters were made on were homogenized in chloroform, methanol, and water
5 uncleaved ova in a depression slide using an ocular micrometer (2.5:5:2 by volume). After repeated extractions followed by
6 on a standard bright field microscope. Volumes were calcu- centrifugation, the liquid phase was dried under nitrogen, and
7 lated from these data using the ellipse of revolution around the the total lipid was weighed to the nearest 10 g.
8 major axis (V  4/3πr 21r2, where r1 is the radius of the minor The content of soluble protein was estimated using the
9 axis and r2 is the radius of the major axis). Three eggs were method of Bradford (1976). Protein was extracted from indi-
50 selected from each of 10 egg masses, one from the centre of the vidual egg masses from each population in 0.2 N NaOH at
1 spiral, one about half way through, and one from the outer room temperature for 5 days. Because the Bradford method is
2 edge. sensitive to pH, an equal volume of 0.2 N HCl was added.
3 All eggs present in individual egg masses of C. ocellifera were Bovine serum albumin was used as the standard, and received
4 counted using a hand-counter. The method described by identical treatment with NaOH and HCl. Absorbance was read
5 Gascoigne & Wallis (1982) was used to count the more numer- at 595 nm at 5–20 min after adding Coomassie Brilliant Blue
6 ous eggs in the egg masses of the planktotrophic Costasiella sp. reagent. The content of insoluble protein was estimated by
7 A Model I linear regression of egg number on dry weight was subtraction (Lawrence & Guille, 1982; Lawrence, McClintock
8 used to convert relative values (per mg dry weight or ash-free & Guille, 1984) of carbohydrate, lipid, and soluble protein
9 dry weight) for organic constituents, ash, and calories (see from the total organic dry weight per egg.
60 below) to absolute values (per egg). Egg masses from both populations of Costasiella were incu-
1 Staining for carbohydrates using the Periodic Acid Schiff bated in the laboratory to hatching. Intact egg masses with
2 (PAS) reaction (Kiernan, 1990) was performed on egg masses uncleaved eggs were placed in enclosed plastic containers
3 fixed for 4 h in cold Rossman’s Fluid (Presnell & Schreibman, (with nylon mesh openings of approximately 30 m) and left
4 1997), embedded in LR White acrylic resin (London Resin Co. in the aquarium, where they were subjected to the same water-
5 Ltd., Woking, UK), and sectioned at 5 m. No reaction was quality and oxygenation regimes as the adults. Development
6 observed on control slides in which the oxidation with periodic was documented daily using high-resolution video equipment
7 acid had been omitted (Kiernan, 1990). and a timetable for development was established for members
8 Frozen sections for lipid staining were prepared using a of each population.

103
 C. M. MILES & K. B. CLARK

RESULTS Mean egg diameter of C. ocellifera was 38% greater than that
of Costasiella sp. (106 and 76.6 m, respectively; Student’s t-test,
The two populations of Costasiella were easily distinguished by t  20.218, P < 0.0001) and egg volume was 158% greater (0.62
their egg masses. The spawn of C. ocellifera contained fewer, and 0.24 nl, respectively; Student’s t-test, t  16.118, P < 0.0001;
larger eggs in a less rigidly structured spiral than those of Table 1). The mean capsule volume for the Type 3a C. ocellifera
Costasiella sp. (Figure 2). The range of number of eggs per (15 nl) was nearly eight times that of the Type 1 Costasiella sp.
mass used in constructing the Model I regressions was 26–313 (1.9 nl) (Student’s t-test, t  16.618, P < 0.0001).
(n  10) for C. ocellifera and 562-2616 (n  10) for Costasiella Egg masses of C. ocellifera contained 58.3% ash on a dry-
sp. and these are significantly different (data normal log trans- weight basis, and the egg masses of C. sp. contained 49.5% ash
formed, Student’s t-test, t  –8.7918, P < 0.0001). Model I linear (Table 2). Total organic content was 3.84 g/egg for C. ocellif-
regression of egg number on dry weight yielded two equations: era and 0.407 g/egg for C. sp. Calorimetric analyses gave rela-
C. ocellifera, Y  18.832  106.698*X (r2  0.948); Costasiella sp., tive values for egg masses of C. ocellifera of 2.09 cal/mg dry
Y  115.948  1229.914*X (r2  0.925). Slopes of both lines weight (DW) and 5.01 cal/mg ash-free dry weight (AFDW).
were significantly different from zero (ANOVA, F  145.4; 98.5, These values were not significantly different from those of
respectively, P < 0.0001) and from each other (F-test for homo- C. sp. [2.24 and 4.14, respectively; Student’s t-test, t  –0.61611,
geneity of slope, F  108.05[1,16]). P  0.550 (DW); t  1.5611, P  0.148 (AFDW)]. The caloric
content of eggs of C. ocellifera (0.019 cal/egg) was, however,
significantly greater (about 10-fold larger) than that of C. sp.
(0.0018 cal/egg; data normal log transformed, Student’s t-test,
t  –21.711, P < 0.0001).
Capsular fluids of egg masses from members of both popula-
tions were PAS-positive, as were the intra-capsular vesicles of C.
ocellifera. No PAS-positive inclusions were visible in the capsules
of C. sp. The eggs of both populations were PAS-negative.
Staining with Bromine Sudan Black B for lipids gave no strik-
ingly positive results for intra-capsular fluid, and no vesicular
inclusions were observed for either population; the egg stained
positively in both populations. The Ninhydrin–Schiff reaction
for proteins was strongly positive for intra-capsular vesicles, and
moderately positive for the egg and the capsular fluid in C. ocel-
lifera. The egg and intra-capsular fluid of Costasiella sp. reacted
moderately, but no intra-capsular vesicles were observed.
Estimates for biochemical assays followed a different trend
than the calorimetric data. There was a significant difference
for both relative (g/mg DW) and absolute (g/egg) values
between members of the two populations for all three assays

–
Table 1. Egg and capsule dimensions (X  SE) for two populations of
Costasiella.

Costasiella ocellifera Costasiella sp.

(n  10) (n  10)

Egg diameter (m) 106  1.4 76.6  0.41

Egg volume (nl) 0.62  0.024 0.24  0.003
Capsule diameter (m)
Major axis 346  5.3 180  2.7
Minor axis 289  6.7 142  1.8
Capsule volume (nl) 15  0.8 1.9  0.07

–
Table 2. Ash, organic, and caloric composition (X  SE) for egg masses
from two populations of Costasiella.

Costasiella ocellifera Costasiella sp.

(n  5) (n  5)
Ash level (% DW) 58.3  2.80 49.5  1.10
Ash content (g/egg) 5.38  0.254 0.398  0.0090
Organic level (% DW) 41.7  2.80 50.5  1.10
Organic content (g/egg) 3.84  0.257 0.407  0.0087
(n  7) (n  6)
Caloric level (cal/mg DW) 2.09  0.190 2.24  0.151
Caloric level (cal/mg AFDW) 5.01  0.453 4.14  0.293
Figure 2. Egg masses from two populations of Costasiella with different Caloric content (cal/egg) 0.019  0.0017 0.0018  0.00012
developmental modes, deposited on Avrainvillea nigricans. (A) Costasiella
ocellifera. (B) Costasiella sp. Abbreviations: AFDW, ash-free dry weight; DW, dry weight.

104
 COSTASIELLA EGG MASSES

1 Table 3. Organic components of egg masses from two populations of DISCUSSION

–
2 Costasiella (X  SE). Means that differ significantly (P < 0.05) between
3 congeners are indicated by asterisks (*). Insoluble protein determined by Values for egg diameter, capsule diameter, and capsule volume
4 subtraction. for Costasiella ocellifera were similar to published values for this
5 species (Clark & Goetzfried, 1978; DeFreese & Clark, 1983).
6 Component Costasiella ocellifera Costasiella sp. Calculations of capsule volume for C. ocellifera were nearly seven
7 times greater than for Costasiella sp. Capsule volume has been
TCA-soluble carbohydrates (n  14) (n  14) found to be more accurate than egg size as an indicator of
8 g/mg DW 67.7  2.10 76.0  1.58 *
9 developmental type in ascoglossans (Clark & Jensen, 1981).
g/mg AFDW 162  5.0 150  3.1 The relative (per mg) caloric content was similar for egg masses
10
g/egg 0.52  0.018 0.058  0.0011 * from each population. However, because there are more and
1
2 Lipids (n  14) (n  15) smaller eggs in Costasiella sp. masses, the absolute (per egg)
3 g/mg DW 32.2  0.59 27.1  0.82 * caloric content was 10-fold greater for C. ocellifera. The absolute
4 g/mg AFDW 77.2  1.41 53.6  1.62 * caloric content for C. ocellifera (0.019 cal/egg) was slightly less
5 g/egg 0.25  0.004 0.020  0.0007 *
than the range of values reported for the same species by
6 DeFreese & Clark (1983; 0.051–0.455 cal/egg).
NaOH-soluble protein (n  21) (n  17)
7 DeFreese & Clark (1983) found that the extremes in egg
g/mg DW 55.9  2.57 75.8  3.16 *
8 caloric content of Type-1 and Type-3 developers overlapped
g/mg AFDW 134  6.2 150  6.3 substantially, indicating that absolute caloric content alone
9 g/egg 0.41  0.014 0.054  0.0019 *
20 is not the major determinant of developmental pattern in
1
Insoluble protein opisthobranchs. The resources necessary to double the develop-
g/egg 2.66 0.259 mental time from 6 days to hatching (as a veliger larva in C. sp.)
2
3 to 12 days to hatching (as a juvenile in C. ocellifera) might be
4 in accessory nutritional components provided by the adult C.
5 ocellifera. The presence of intra-capsular vesicles in C. ocellifera
6 was reported by Clark & Goetzfried (1978) and their use as a
7 (carbohydrate, lipid, and NaOH-soluble protein). Values for nutritional supplement during development was hypothesized.
8 TCA-soluble carbohydrates were 67.7 and 76.0 g/mg DW, and The vesicles were clearly visible under the light microscope in
9 0.52 and 0.058 g/egg for C. ocellifera and C. sp., respectively this study, but were not seen in the capsular fluid of C. sp.
30 [Table 3; Student’s t-test, t  3.1426, P  0.0042 (DW); data Histochemical tests revealed that the capsular fluid in both
1 normal log transformed, t  55.026, P < 0.0001 (g/egg)]. The populations was similar (PAS- and Ninhydrin–Schiff-positive,
2 mean estimates of lipid were 32.2 and 27.1 g/mg DW, and Bromine Sudan Black B-negative). Ghiselin (1965) found that
3 0.25 and 0.020 g/egg for C. ocellifera and C. sp., respectively the intra-capsular fluid of the dorid nudibranch Dendrodoris
4 [Student’s t-test, t  –4.9827, P < 0.0001 (DW); data normal log albopunctata (Cooper, 1863) was PAS-positive and concluded
5 transformed, t  67.127, P < 0.0001 (g/egg)]. No significant that it was a neutral carbohydrate. Kuzirian (1973) found the
6 difference was found between the gravimetric (n  11) and intra-capsular fluid in three aeolid nudibranchs from the genus
7 colorimetric (n  11) methods for estimating total lipid con- Coryphella to be composed of a weakly acidic, sulphated muco-
8 tent of Aplysia californica spawn (Mann–Whitney U-test, U  40, polysaccharide. Eyster (1986) found no periodate reactive
9 P  0.178). Values obtained colorimetrically for Costasiella carbohydrates in the intra-capsular fluids of Aeolidia papillosa
40 were, therefore, not converted to gravimetric values. The mean (Linnaeus).
1 estimates of NaOH-soluble protein were 55.9 and 75.8 g/mg Eggs from members of the two populations of Costasiella were
2 DW, and 0.41 and 0.054 g/egg for C. ocellifera and for C. sp., also similar, exhibiting a high lipid content, a positive reaction
3 respectively [Student’s t-test, t  –4.9336, P < 0.0001 (DW); data for protein, but no reaction for carbohydrate. Provisioning the
4 normal log transformed, t  39.236, P < 0.0001 (g/egg)]. eggs of C. ocellifera with intra-capsular vesicles (PAS- and
5 When data based on dry weight were adjusted for ash content, Ninhydrin–Schiff-positive glycoprotein) was one of the defin-
6 no significant difference between C. ocellifera and C. sp. was ing differences between the two populations. Bayne (1968),
7 found for relative amounts of carbohydrates (162 and 150 working with egg masses of the anaspidean opisthobranch
8 g/mg AFDW, respectively; Student’s t-test, t  2.0126, P  Aplysia punctata (Cuvier, 1803), concluded that all nutritive
9 0.0549) or soluble proteins (134 and 150 g/mg AFDW; reserves were contained within the egg cells, and these tested
50 Student’s t-test, t  –1.8136, P  0.0793). However, lipid values strongly PAS-positive and positive for protein.
1 differed significantly between the two groups (77.2 and 53.6 Statistically significant differences in total lipids, carbo-
2 g/mg AFDW for C. ocellifera and C. sp., respectively; Student’s hydrates, and NaOH-soluble proteins were found in both relative
3 t-test, t  10.927, P < 0.0001). Conversion of ash-weight data into and absolute values for egg masses of the two populations. As
4 total organic weight per egg gave values of 3.84 g/egg for with the caloric contents, all absolute (per egg) biochemical
5 C. ocellifera and 0.407 g/egg for C. sp. (Table 2). Insoluble assay results were significantly higher in C. ocellifera egg masses.
6 protein—calculated by subtraction of carbohydrate, lipid, and The mean relative value of lipid (i.e. per milligram) was also
7 soluble protein from total organic matter—comprised a large significantly higher for egg masses of C. ocellifera, whether calcu-
8 proportion of the eggs of both species (Figure 3). Because lated on a dry-weight or ash-free dry-weight basis. Because lipid
9 insoluble protein was estimated by subtracting mean values for was not detected in the capsular fluid, the higher lipid level
60 other constituents from the total mean organics, no test for probably accounts for the greater egg volume of C. ocellifera
1 significance was performed. than of C. sp.
2 Embryogenesis in the two populations progressed at similar Both TCA-soluble carbohydrate and NaOH-soluble protein
3 rates through the fifth day. Notations on specific larval struc- data revealed significantly higher values per mg in the egg
4 tures or stages were not made, but the progression through masses of C. sp. Members of both populations reach the veliger
5 embryogenesis was similar in appearance. On the sixth day, the stage within the capsule and it can be assumed that the
6 planktotrophic larvae of C. sp. (n  5) hatched, but embryo- resources necessary to reach this stage are provided to both.
7 genesis of C. ocellifera (n  5) continued until embryos hatched The egg masses of C. sp. were consistently found to contain at
8 as crawl-away juveniles on the twelfth day. least 10 times the number of eggs per mass (though no associa-

105
 C. M. MILES & K. B. CLARK

Figure 3. Relative biochemical composition of egg masses from two populations of Costasiella. A. Costasiella ocellifera. B. Costasiella sp. Units are g/egg. Error
bars indicate standard error. Insoluble protein calculated by subtraction.

tion of individual adults, or their weights, and the specific egg planktotrophic larvae of C. sp. could also increase fitness by
masses they deposited was made for either group). This means facilitating growth, shortening the planktonic phase preceding
that, within the same time frame, 10 times the number of metamorphosis, or increasing tolerance to starvation periods
veligers must be assembled with 10 times the number of larval encountered during the planktonic phase.
shells, and each must be provided with energy for spinning Costasiella ocellifera, the encapsulated developer, must be pro-
while still within the capsule. Jaeckle (1995) found a larger pro- vided with additional resources in order to complete develop-
portion of protein in planktotrophic echinoderm eggs than in ment beyond the veliger, a process that includes formation of a
lecithotrophic eggs, and he felt that this indicated a relatively functional radula, resorption of the velum, and exit from the
greater need for presynthesized structural materials during capsule. Biochemical assays estimate the quantity of organic
early development. Additional resources provided to the constituents provided to offspring, but do not address the

106
 COSTASIELLA EGG MASSES

1 qualitative aspect. Provisioning of developing C. ocellifera with mode. By making modifications in the provisioning of off-
2 accessory glycoprotein vesicles, separate and distinct from spring through accessory structures, perhaps C. ocellifera is able
3 the capsular fluid, seems to affirm that some of the resources to achieve the advantages of encapsulated development (e.g.
4 needed to extend development are provided extra-embryonic- assure hatching on an appropriate food source, avoid exposure
5 ally. Penchaszadeh & Rincon (1996) analysed the capsular to planktonic predation) without the need to produce a large
6 fluid at different developmental stages in Prunum prunum yolky egg. The amount of provisioning for C. sp. suggests that
7 (Gmelin, 1791), a marginellid gastropod, and concluded some of the advantages of encapsulated development can be
8 that the embryo appears to require extra-embryonic food to incorporated without loss of dispersal. Evidence is mounting
9 complete development despite a generous supply of yolk. that life history variability, and response flexibility and plasticity
10 Stöckmann-Bosbach & Althoff (1989) examined the capsular characterize the reproduction and development of many
1 fluid of the muricid gastropod Nucella lapillus L. and found that benthic marine invertebrates (Hadfield & Strathmann, 1996).
2 albumen can be important even in cases in which nurse eggs Classification of developmental type is necessarily a subjective
3 are the major food supply for the developing embryo. During determination because it seeks to describe a continuum of
4 development it is not only relevant what specific resources are variable characters as if they were discrete. Calorimetric and
5 provided to offspring and in what amounts, but also how they biochemical results support the idea that some ascoglossans
6 are provided. Differences in location and availability of bio- have the potential to achieve more than one developmental
7 chemical constituents can be essential for controlling the pattern, especially through modification in quantity and
8 timing of events or for maintaining reserves of a particular quality of accessory provisioning, but selective forces constrain
9 resource until specific developmental stages are achieved. the expression to a single pattern (DeFreese & Clark, 1983).
20 Extra-zygotic nutrition might be ingested directly, as suggest- Two explanations can be offered for the presence of two
1 ed for C. ocellifera by Clark & Goetzfried (1978). Moran (1999) reproductive patterns in these populations: poecilogony and
2 showed that uptake of capsular proteins occurred in ciliated sibling species. Poecilogony is the presence of two develop-
3 cells of the velum and foot in species of Littorina that have mental modes within a species. Poecilogony in gastropods was
4 encapsulated development. Other larval structures, such as reviewed by Bouchet (1989) and in marine invertebrates in
5 ‘larval kidneys’ (Rivest, 1992), are involved in endocytotic general by Chia, Gibson & Qian (1996). Poecilogonous species
6 uptake of capsular proteins. However, Rivest (1992) found no are rare and the literature reveals many reported cases that
7 absorptive cells in the two opisthobranchs included in his were later defined as sibling species. Sibling or cryptic species
8 survey mostly of prosobranch gastropods. The mechanism of are difficult or impossible to distinguish by morphological
9 uptake of intra-capsular nutrients by either of these congeners characters (Mayr & Ashlock, 1991). The abundance of marine
30 could include both ingestion (after formation of feeding and sibling species was reviewed by Knowlton (1993), Knowlton
1 digestive organs) and endocytosis. and Jackson (1994), and Palumbi (1994); they are ubiquitous.
2 Members of both populations of Costasiella, like many While poecilogonous species are rare, the phenomenon has
3 ascoglossans, exhibit specialized traits that act to narrow their been documented in opisthobranchs [e.g. Elysia chlorotica (West,
4 ecological niches [e.g. dietary specificity (Jensen, 1980) and Harrigan & Pierce, 1984), Haminoea callidegenita (Gibson &
5 kleptoplasty (Clark, Jensen, Stirts & Fermin, 1981)]. These two Chia, 1991), Alderia modesta (Krug, 1998)], and polychaetes
6 congeners are indistinguishable based on external morph- [e.g. Capitella sp. (Qian & Chia, 1991, 1992a,b, 1994), Streblospio
7 ology, but they can be distinguished by developmental mode benedicti (Levin, 1984b; Levin & Bridges, 1994)]. Support for a
8 and by habitat. Clark (1994) referred to Costasiella sp. as a hypothesis of poecilogony must include at least one of three
9 reproductively isolated undescribed species. Costasiella sp. was types of data (Hoagland & Robertson, 1988): genetic data to
40 found in the Upper Florida Keys, occupying a mangrove fringe evaluate if inter-breeding occurs between sympatric individ-
1 habitat where Avrainvillea nigricans is tall and abundant. uals, successful cross-breeding of normally allopatric individ-
2 Costasiella ocellifera was found in the Lower Florida Keys, occupy- uals, or observation of a single individual producing two types
3 ing an exposed coral sand habitat where A. nigricans is shorter of larva in the laboratory, e.g. under varying environmental or
4 and less densely spaced. No quantitative data on A. nigricans nutritional conditions. In this study the environmental and
5 densities are reported here, but both of these habitats were nutritional conditions of both populations were intentionally
6 described fully by Clark & DeFreese (1987). Degree of patchi- kept identical in the laboratory, and no indication of variability
7 ness of the substratum can influence the success of a particular in pattern of development was observed.
8 developmental mode (Levin & Bridges, 1995). Even small In order to perform cross-breeding experiments, it would
9 changes in habitat resources can produce shifts in distribution be necessary to raise virgin individuals from each of the two
50 and abundance patterns of a species (Eggleston, Elis, Ether- populations in the laboratory from hatching to reproductive
1 ington, Dahlgren & Posey, 1999). Levin (1984a) observed that maturity to insure parentage. These experiments were not per-
2 mudflats in Californian lagoons that were separated by lengths formed because of difficulties in culturing C. sp. through its
3 of high-energy coastline exhibited higher incidence of species planktotrophic stage, but are planned for the near future.
4 with reduced potential for larval dispersal. Island ecologists have Allozyme or gene-sequence analyses of these two congeners
5 theorized that the dispersal inherent in many forms with plank- might also help determine if this is a case of poecilogony or of
6 tonic larvae might not be beneficial to organisms in patchy sibling species.
7 and unpredictable habitats (Carlquist, 1974). The advantage of
8 hatching directly onto a suitable food source that is temporally ACKNOWLEDGEMENTS
9 persistent, but patchily distributed, might have provided the
60 selective pressure to evolve extended development in C. ocellif- R. L. Turner, P. M. Mikkelsen, Aplysia Resource Facility, two
1 era. This hypothesis was also offered for the evolution of anonymous referees and The Astronaut Trail Shell Club of
2 lecithotrophic development in the nudibranch Doto amyra Melbourne Beach, FL, USA, for partially funding this research.
3 Marcus, 1961 by Goddard (1996).
4 Opisthobranch larval development is regulated by strong REFERENCES
5 phylogenetic constraints (Hadfield & Miller, 1987). By pro- BARNES, H. & BLACKSTOCK, J. 1973. Estimation of lipids in marine
6 visioning the embryo with the appropriate quantity and quality animals and tissues: detailed investigation of the sulphophospho-
7 of nutrients, largely through modifications of accessory struc- vanillin method for ‘total’ lipids. Journal of Experimental Marine
8 tures, C. ocellifera might retain flexibility in its developmental Biology and Ecology, 12: 103–118.

107
 C. M. MILES & K. B. CLARK

BAYLISS, O.B. & ADAMS, C.W.M. 1972. Bromine Sudan Black (BSB): a FIORONI, V.P. 1977. On the peripheric resorption of albumen in
general stain for tissue lipids including free cholesterol. Histo- gastropods and its significance for general embryology. Zoologische
chemical Journal, 4: 505–515. Jahrbücher Abteilung für Anatomie und Ontogenie der Tiere, 98: 181–202.
BAYNE, C.J. 1968. Histochemical studies on the egg capsules of eight GASCOIGNE, T. & WALLIS, D.E. 1982. On counting the eggs of
gastropod molluscs. Proceedings of the Malacological Society of London, ascoglossans. Journal of Molluscan Studies, 48: 219–224.
38: 199–212. GHISELIN, M.T. 1965. Reproductive function and the phylogeny of
BLIGH, E.G. & DYER, W.J. 1959. A rapid method of total lipid extrac- opisthobranch gastropods. Malacologia, 3: 327–378.
tion and purification. Canadian Journal of Biochemistry and Physiology, GIBSON, G.D. & CHIA, F-S. 1991. Contrasting reproductive modes in
37: 911–917. two sympatric species of Haminaea (Opisthobranchia: Cepha-
BONAR, D.B. 1978. Morphogenesis at metamorphosis in opistho- laspidea). Journal of Molluscan Studies, 57: 49–60.
branch molluscs. In: Settlement and metamorphosis of marine invertebrate GODDARD, J.H.R. 1996. Lecithotrophic development in Doto amyra
larvae (F. Chia & M. Rice, eds), 177–196. Elsevier, New York. (Nudibranchia: Dendronotacea), with a review of developmental
BOUCHET, P. 1989. A review of poecilogony in gastropods. Journal of mode in the genus. Veliger, 39: 43–54.
Molluscan Studies, 55: 67–78. HADFIELD, M.G. & MILLER, S.E. 1987. On developmental patterns of
BRADFORD, M.M. 1976. A rapid and sensitive method for the quanti- opisthobranchs. American Malacological Bulletin, 5: 197–214.
tation of microgram quantities of protein utilizing the principle of HADFIELD, M.G. & STRATHMANN, M.F. 1996. Variability, flexibility,
protein-dye binding. Analytical Biochemistry, 72: 248–254. and plasticity in life histories of marine invertebrates. Oceanologica
CARLQUIST, S. 1974. Island biology. Columbia University Press, New Acta, 19: 323–334.
York. HADFIELD, M.G. & SWITZER-DUNLAP, M. 1984. Opisthobranchs.
CHIA, F-S., GIBSON, G. & QIAN, P-Y. 1996. Poecilogony as a repro- In: The Mollusca (A.S. Tompa, N.H. Verdonk & J.A.M. van den
ductive strategy of marine invertebrates. Oceanologica Acta, 19: Biggelaar, eds), Vol. 7, 209–350. Academic Press Inc., Orlando,
203–208. Florida.
CLARK, K.B. 1971. The construction of a collecting device for small HIGH, O.B. 1984. Lipid histochemistry. Oxford University Press, London.
aquatic organisms and a method for rapid weighing of small inverte- HOAGLAND, K.E. & ROBERTSON, R. 1988. An assessment of poecilo-
brates. Veliger, 13: 364–367. gony in marine invertebrates: phenomenon or fantasy? Biological
Bulletin (Woods Hole), 174: 109–125.
CLARK, K.B. 1994. Ascoglossan (Sacoglossa) molluscs in the Florida
Keys: rare marine invertebrates at special risk. Bulletin of Marine JAECKLE, W.B. 1995. Variation in the size, energy content, and bio-
Science, 54: 900–916. chemical composition of invertebrate eggs: correlates to the mode
of larval development. In: Ecology of marine invertebrate larvae (L.
CLARK, K.B. & DEFREESE, D. 1987. Population ecology of Caribbean
McEdward, ed.), 49–77. CRC Press, Boca Raton.
Ascoglossa (Mollusca: Opisthobranchia): a study of specialized algal
herbivores. American Malacological Bulletin, 5: 259–280. JENSEN, K. 1980. A review of sacoglossan diets, with comparative notes
on radular and buccal anatomy. Malacological Review, 13: 55–77.
CLARK, K.B. & GOETZFRIED, A. 1978. Zoogeographic influences on
developmental patterns of North Atlantic Ascoglossa and Nudi- KIERNAN, J.A. 1990. Histological and histochemical methods: theory and
branchia, with a discussion of factors affecting egg size and number. practice. Pergamon Press, Oxford.
Journal of Molluscan Studies, 44: 283–294. KNOWLTON, N. 1993. Sibling species in the sea. Annual Review of
CLARK, K.B. & JENSEN, K.R. 1981. A comparison of egg size, capsule Ecology and Systematics, 24: 189–216.
size, and development patterns in the order Ascoglossa (Saco- KNOWLTON, N. & JACKSON, J.B.C. 1994. New taxonomy and niche
glossa) (Mollusca: Opisthobranchia). International Journal of Inverte- partitioning on coral reefs: jack of all trades or master of some?
brate Reproduction, 3: 57–64. Trends in Ecology and Evolution, 9: 7–9.
CLARK, K.B., JENSEN, K.R. & STIRTS, H.M. 1990. Survey for function- KRUG, P.J. 1998. Poecilogony in an estuarine opisthobranch: plank-
al kleptoplasty among West Atlantic Ascoglossa ( Sacoglossa). totrophy, lecithotrophy, and mixed clutches in a population of the
Veliger, 33: 339–345. ascoglossan Alderia modesta. Marine Biology (Berlin), 132: 483–494.
CLARK, K.B., JENSEN, K.R., STIRTS, H.M. & FERMIN, C. 1981. KUZIRIAN, A.M. 1973. The anatomy and functional morphology of
Chloroplast symbiosis in a non-elysiid mollusc, Costasiella lilianae the reproductive systems of three coryphellid nudibranchs (Gastro-
(Marcus) (Hermaeidae: Ascoglossa  Sacoglossa): effects of tem- poda: Opisthobranchia). MS thesis, Northeastern University,
perature, light intensity, and starvation on carbon fixation. Biological Boston.
Bulletin (Woods Hole), 160: 43–54. LAWRENCE, J.M. & GUILLE, A. 1982. Organic composition of tropical,
DEFREESE, D.E. & CLARK, K.B. 1983. Analysis of reproductive polar and temperate-water echinoderms. Comparative Biochemistry
energetics of Florida Opisthobranchia (Mollusca: Gastropoda). and Physiology, 72B: 283–287.
International Journal of Invertebrate Reproduction, 6: 1–10. LAWRENCE, J.M., MCCLINTOCK, J.B. & GUILLE, A. 1984. Organic
DE MAHIEU, G.C., PENCHASZADEH, P.E. & CASAL, A.B. 1974. level and caloric content of eggs of brooding asteroids and an
Algunos aspectos de las variaciones de proteínas y aminoácidos echinoid (Echinodermata) from Kerguelen (South Indian Ocean).
libres totales del líquido intra-capsular en relación al desarrollo International Journal of Invertebrate Reproduction and Development, 7:
embrionario en Adelomelon brasiliana (Lamarck, 1811) (Gastropoda, 249–257.
Prosobranchia, Volutidae). Cahiers de Biologie Marine, 15: 215–227. LEVIN, L. 1984a. Life history and dispersal patterns in a dense infaunal
DUBOIS, M., GILLES, K.A., HAMILTON, J.K., REBERS, P.A. & polychaete assemblage: community structure and response to
SMITH, F. 1956. Colorimetric method for determination of sugars disturbance. Ecology, 65: 1185–1200.
and related substances. Analytical Chemistry, 28: 350–356. LEVIN, L. 1984b. Multiple patterns of development in Streblospio bene-
EGGLESTON, D.B., ELIS, W.E., ETHERINGTON, L.L., DAHLGREN, dicti Webster (Spionidae) from three coasts of North America.
C.P. & POSEY, M.H. 1999. Organism responses to habitat frag- Biological Bulletin (Woods Hole), 166: 494–508.
mentation and diversity: habitat colonization by estuarine macro- LEVIN, L. & BRIDGES, T.S. 1994. Control and consequences of altern-
fauna. Journal of Experimental Marine Biology and Ecology, 236: ative developmental modes in a poecilogonous polychaete.
107–132. American Zoologist, 34: 323–332.
EYSTER, L.S. 1980. Distribution and reproduction of shell-less opistho- LEVIN, L. & BRIDGES, T.S. 1995. Pattern and diversity in reproduc-
branchs from South Carolina. Bulletin of Marine Science, 30: 580–599. tion and development. In: Ecology of marine invertebrate larvae (L.
EYSTER, L.S. 1986. The embryonic capsules of nudibranch molluscs: McEdward, ed.), 1–48. CRC Press, Boca Raton.
literature review and new studies on albumen and capsule wall ultra- LOBBAN, C.S. & HARRISON, P.J. 1994. Seaweed ecology and physiology.
structure. American Malalcological Bulletin, 4: 205–216. Cambridge University Press, Cambridge.

108
 COSTASIELLA EGG MASSES

1 MAYR, E. & ASHLOCK, P.D. 1991. Principles of systematic zoology. RIVEST, B.R. 1992. Studies on the structure and function of the larval
2 McGraw-Hill Inc., New York. kidney complex of prosobranch gastropods. Biological Bulletin
3 MILOSLAVICH, P. 1999. Nutritional value of the intracapsular liquid (Woods Hole), 182: 305–323.
4 of Engoniophos unicinctus Say, 1825 (Caenogastropoda: Buccinidae). RIVEST, B.R. & STRATHMANN, R.R. 1995. Uptake of protein by an
5 Journal of Molluscan Studies, 65: 502–503. independently evolved transitory cell complex in encapsulated
6 MORAN, A.L. 1999. Intracapsular feeding by embryos of the gastropod embryos of Neritoidean gastropods. In: Reproduction and Development
7 genus Littorina. Biological Bulletin (Woods Hole), 196: 229–244. of Marine Invertebrates (W.H. Wilson; S.A. Strickler & G.L. Shinn,
eds), 166–176. Johns Hopkins University Press, Baltimore.
8 OLSEN-STOJKOVICH, J. 1985. A systematic study of the genus
9 Avrainvillea Decaisne (Chlorphyta, Udoteaceae). Nova Hedwigia, 41: SAUTTO-VALLEJO, L.G. 1992. Nutrient reserves and their mobiliza-
1–68. tion during the development of embryos of Littorina saxatilis
10
(Gastropoda, Prosobranchia, Littorinidae). In: Proceedings of the
1 PAINE, R.T. 1964. Ash and calorie determinations of sponge and Ninth International Malacological Congress (E. Gittenberger & J. Goud,
2 opisthobranch tissues. Ecology, 45: 384–387. eds), 335–340. Unitas Malacologia, Leiden.
3 PALUMBI, S.R. 1994. Genetic divergence, reproductive isolation, and SILVA, P.C. 1982. Chlorophyceae. In: Synopsis and classification of living
4 marine speciation. Annual Review of Ecology and Systematics, 25: organisms (S.P. Parker, ed.), Vol. 1, 133–161. McGraw-Hill, New
5 547–572. York.
6 PENCHASZADEH, P.E. & RINCON, A. 1996. Egg capsules and devel- STÖCKMANN-BOSBACH, R. & ALTHOFF, J. 1989. A correlated
7 opment of Prunum prunum (Gmelin, 1791) (Prosobranchia: morphological and biochemical study of capsular fluid of Nucella
8 Marginellidae) from the Venezuelan Caribbean. Veliger, 39: 83–86. lapillus (Gastropoda: Prosobranchia: Muricidae). Marine Biology
9 PHILLIPSON, J. 1964. A miniature bomb calorimeter for small biologi- (Berlin), 102: 283–289.
20 cal samples. Oikos, 15: 130–139. THOMPSON, T.E. 1967. Direct development in a nudibranch Cadlina
1 PRESNELL, J.K. & SCHREIBMAN, M.P. 1997. Humason’s animal tissue laevis, with a discussion of developmental processes in Opistho-
2 techniques, 5th edn. John’s Hopkins University Press, Baltimore. branchia. Journal of the Marine Biological Association of the United
3 QIAN, P-Y. & CHIA, F-S. 1991. Fecundity and egg size are mediated by Kingdom, 47: 1–22.
4 food quality in the polychaete worm Capitella sp. Journal of Experi- TODD, C.D. 1979. Reproductive energetics of two species of dorid
5 mental Marine Biology and Ecology, 148: 11–25. nudibranchs with planktotrophic and lecithotrophic larval stages.
6 QIAN, P-Y. & CHIA, F-S. 1992a. Effect of aging on reproduction in a Marine Biology (Berlin), 53: 57–68.
7 marine polychaete Capitella sp. Journal of Experimental Marine Biology WEST, H.H., HARRIGAN, J.F. & PIERCE, S.K. 1984. Hybridization of
8 and Ecology, 156: 23–38. two populations of a marine opisthobranch with different develop-
9 QIAN, P-Y. & CHIA, F-S. 1992b. Effects of diet type on the demograph- mental patterns. Veliger, 26: 199–206.
30 ics of Capitella sp. (Annelida: Polychaeta): lecithotrophic develop- YASUMA, A. & ICHIKAWA, T. 1953. Ninhydrin–Schiff and Alloxan–
1 ment vs planktotrophic development. Journal of Experimental Marine Schiff staining: a new histochemical staining method for protein.
2 Biology and Ecology, 157: 159–179. Journal of Laboratory and Clinical Medicine, 41: 296–299.
3 QIAN, P-Y. & CHIA, F-S. 1994. In situ measurement of recruitment,
4 mortality, growth, and fecundity of Capitella sp. (Annelida: Poly-
5 chaeta). Marine Ecology Progress Series, 111: 53–62.
6
7
8
9
40
1
2
3
4
5
6
7
8
9
50
1
2
3
4
5
6
7
8
9
60
1
2
3
4
5
6
7
8

109