Shaving a shell: Effect of manipulated sculpture and feeding on shell growth and sculpture development in Nucella lamellosa (Muricidae: Ocenebrinae).
Gastropod shell sculpture offers a potentially intriguing tool to study morphological patterning. Shell sculpture is highly variable, yet is made up of relatively discrete elements. As a snail grows, it must control when, where, and to what extent sculpture is produced. Although shell shape and, to a much lesser degree, shell sculpture have been modeled computationally and studied geometrically (Ackerly, 1989; Hammer and Bucher, 2005; Meinhardt, 2009; Chirat et al., 2013), the factors affecting sculpture growth have not been studied experimentally.
For spiral sculpture, which is oriented parallel to the direction of shell growth, a sculpture pattern could be set at the shell lip initially and remain fixed for the rest of a snail's life; mantle shape does not need to vary over time to maintain sculpture patterning (Meinhardt, 2009). The potential mechanism is more complicated for axial sculpture, where ribs, varices, or lamellae are oriented perpendicularly to the direction of shell growth (parallel to the apertural margin). The mantle's form or positioning must vary over time to produce axial sculpture periodically. The degree of regularity in axial sculpture varies greatly. Varices, or periodic thickenings of the aperture that uplift from the shell surface, are usually regularly spaced (Webster et al., unpubl. data; Vermeij, 1995; Savazzi and Sasaki, 2004). This spacing yields a synchronized pattern, where each newly produced varix lines up with a varix on the previous whorl. Previous varices on the shell are thought to provide a cue for the location of these new varices (Savazzi and Sasaki, 2004). Varices are also thought to be associated with episodic shell growth, in which shell length does not increase for some time after the growth of a varix. In episodic shell growth, reinforcement of existing shell can still occur, and the body may continue to expand to fill the new living space. This hiatus between periods of shell growth often produces a thickened apertural lip, and can last for several months (MacKenzie, 1961; Inaba, 1967; MacGinitie and MacGinitie, 1968; Spight et al., 1974; Spight and Lyons, 1974; Illert, 1981).
No hypothesis exists to explain how snails control less regular patterns of axial sculpture. Nucella lamellosa (Gmelin, 1791) is an intertidal muricid gastropod from the Pacific coast of North America, and is best known for its highly plastic shell morphology and its ability to change shell form adaptively in response to predators (Kincaid, 1957; Spight, 1973; Palmer, 1985; Appleton and Palmer, 1988). In the presence of crab predators, N. lamellosa grow a thicker shell with minimal shell sculpture and better developed apertural teeth (Appleton and Palmer, 1988). Where crab predation is less of a threat, N. lamellosa generally exhibit the frilled sculpture form (Fig. 1). This thin-shelled, frilled form bears semi-regular axial sculpture, variously called "foliate ribs" or "lamellose ridges" (Webster et al., unpubl. data; Kincaid, 1957; Abbott, 1974), but which we will refer to as "lamellae." These axial lamellae resemble the varices of other muricids. Although they are thinner and more variable, they may have a similar function in shell defense (Palmer, 1979; Miller and LaBarbera, 1995; Donovan et al., 1999). The axial lamellae of N. lamellosa are thought to reflect a trade-off between the slow-growing, but well defended thick form, and a thin-shelled, faster-growing alternative, in which lamellae may provide some reinforcement of the lip (Spight and Lyons, 1974; Palmer, 1981, 1985). Similar lamellae occur in many other muricids (Radwin and D'Attilio, 1976).
Elaborate sculpture is possible only because most gastropods can dissolve shell material that might otherwise impede subsequent shell growth (Vermeij, 1977). As a snail grows, it must first remove sculpture from the previous whorl that may obstruct the aperture. The cost of such resorption in energy or time is unknown. Resorption of sculpture could impose additional energetic costs, or constrain the maximal rate of shell growth by requiring more time to remove preexisting sculpture.
Here we describe in detail--for the first time--the growth of axial lamellae in N. lamellosa. We explore the factors that affect lamellar growth and how lamellae differ in form from varices, and attempt to answer these questions: Are lamellae associated with a growth hiatus? Is the spacing between axial lamellae regular? How quickly are lamellae produced? Does the feeding rate affect lamellar spacing or shell growth rate? Do existing lamellae affect the body growth rate or the addition of new lamellae?
Materials and Methods
Field collections and husbandry
Small to medium-sized, "frilled" Nucella lamellosa (15-36 mm shell length; Fig. 1) were collected from the Ross Islets, Barkley Sound, British Columbia (BC), Canada (48.872200-125.161783), for all experiments. Snails reared in the laboratory were maintained in flow-through seawater tables (9-12 [degrees]C) at the Bamfield Marine Sciences Centre, Bamfield, BC, Canada. Six to 10 snails were held together in individual, perforated Ziploc containers. For food, snails were given small stones covered in barnacles (Balanus glandula), which were changed weekly.
Measurements and analyses
The starting location of shell growth was marked by painting the margin of the aperture with nail polish (Gosselin, 1993). Shell weight was calculated from immersed weight (MetderP153 Balance; Mettler-Toledo Inti., Inc., Columbus, OH) following Palmer (1982). Shell length (apex to tip of siphonal canal) was measured with digital calipers (MaiCal 16 EWR; Mahr GmbH, Gottingen, Germany). Shell diameter was measured, using digital calipers, on the body whorl, starting at the aperture and excluding lamellae (Fig. 1A). Angular growth of the aperture and angular lamellar spacing were measured from apical shell images using Imaged (Rasband, 1997). Using the apex as the center point, the angles were measured at the suture between the body whorl and the penultimate whorl. Angular lamellar spacing was measured from where the edge of the lamellae contacted the body whorl. This point was determined by drawing lines along the edge of each lamella from the shoulder (P1) to the suture (Fig. 1A), to account for shell curvature. Spiral growth of the aperture and lamellar spacing (distance between two adjacent lamellae) were calculated as linear distances along the arc of the shell in the direction of growth. They were calculated from the shell diameter (d; mm) and angular growth ([theta]; degrees) or the angular spacing between adjacent lamellae ([theta]; degrees), such that arc [pi] ml X [theta]/360[degrees] for both measurements. Forty N. lamellosa were measured in order to calculate the relationship between shell diameter and shell length (diameter (mm) = 0.380 X shell length (mm) + 3.323; [r.sup.2] = 0.81).
Measurement error was estimated by taking blind repeated .measurements on different days: shell length: [+ or -] 0.8% (n = 30); shell diameter: [+ or -] 2.5% (n = 40); lamellar angles, different photos: [+ or -] 16.6% (n = 28); and lamellar angles, same photos: [+ or -] 6.9% (n = 28).
The state of the apertural sculpture was scored in two stages. First, apertures were scored according to the state of the axial lamella: "no lamella" (no outward bend of the aperture edge), "incipient lamella" (evidence of some outward bending of the aperture), or "full lamella" (a complete lamella was clearly present; see inset drawing, Fig. 2). Lamellae were considered distinct when their tips at the primary shoulder cords (P1; Fig. 1) were not touching. Second, the aperture was scored for the presence or absence of a frill (multiple lamellae stacked closely together; Fig. 1C). A frill was defined as axial sculpture at least three times as thick as other lamellae on the shell, and apparently made up of multiple lamellae stacked together.
When quantifying the timing of lamellar growth, frills were considered to be single lamellae, and the time to grow a lamella was counted from the start of one "no lamella" phase to the beginning of the next "no lamella" phase.
Statistics were run in RStudio/R (RStudio Team, 2012); the "Least-Squares Means" package (Lenth, 2014) was used to compute the mean response for each treatment at a standardized body size from the ANCOVA. Least-squares means were calculated only when the slopes did not differ significantly among treatments (all cases). Images were edited with Adobe Photoshop, version CS6 (Adobe Systems, Inc., Waltham, MA), to adjust brightness and contrast, and to clean up the background.
In June 2012, 36 Nucella lamellosa (16-27 mm in length) were collected and separated into 3 feeding treatments, with a comparable size distribution of snails in each treatment. Snails were acclimatized to laboratory conditions for two months. Those receiving the "High" feeding treatment received ad libitum barnacles; those given the "Medium" feeding treatment received barnacles for four consecutive days each week; and snails receiving the "Low" feeding treatment were given barnacles for two consecutive days each week. A bare rock was placed in the cage when snails were not being fed. Snails were grown for 112 days. Immersed weight, shell length, number of axial lamellae, and spiral growth of the aperture were measured approximately every three weeks.
Sculpture growth rate
To quantify the rates of addition of new lamellae and spiral shell growth, 33 N. lamellosa (16-36 mm shell length) were collected in April 2014, and grown for 50 days in individual containers. The number of barnacles eaten was scored weekly by counting empty tests (stones were cleared of all empty tests before being placed in the cage). The state of apertural sculpture was scored every 1-2 days for 2 periods of 22 days each (i.e., the first and second time periods). Periods were separated by 12 days, during which time there was no scoring.
Field sculpture growth
In April 2014, 148 N. lamellosa were scored in the field for shell length and state of apertural sculpture, in order to document the proportion of each stage found in the field.
Sculpture removal experiment
In April 2013, 90 N. lamellosa (15-33 mm shell length) were collected and separated into 3 treatments, each with an even size distribution. Snails were acclimatized to laboratory conditions for one month. "Lamellae removed" snails had all shell sculpture except the apertural lamella removed with a Dremel grinding tool (Dremel 300 series; Robert Bosch Tool Corp., Anaheim, CA). Grinding did not damage the body whorl or otherwise harm the snail (Fig. 1E). "Aperture removed" snails had the apertural lamella removed, including the portion of the body whorl extending to the penultimate lamella (Fig. IF). "Control" snails experienced no shell manipulation. Snails were grown for 141 days, during which time immersed weight, shell length, number of axial lamellae, and angular growth were measured at 3 times.
Are lamellae associated with a growth hiatus?
Nucella lamellosa of all sizes showed evidence of recent spiral growth in the field: 50%-70% of snails of all size classes were observed without a complete apertural lamella. The proportion of "growing" snails did not differ significantly among size classes ([chi square] = 5.9, df = 6, P = 0.44; Fig. 2). Frills--areas where many lamellae were grown all together--were found only in snails larger than 20 mm in shell length. The proportion of snails with frills increased with shell size (Fig. 2).
Is the spacing between axial lamellae regular?
The spacing between adjacent lamellae varied from 0.4-9.0 mm (mean = 4.0 [+ or -] 0.05 mm; n = 888; Fig. 3A). The level of variation in lamellar spacing differed among snails, with the standard deviation (SD) of lamellar spacing in an individual ranging from 0.55 to 2.1 mm (mean = 1.3 [+ or -] 0.035 mm, n = 87; Fig. 3B). Qualitatively, some individuals had fairly regularly spaced lamellae, where each lamella approximately lined up to one on the previous whorl (Fig. 3C); others clearly had an irregular arrangement of lamellae (Fig. 3D).
How quickly are lamellae produced?
All 33 snails in the sculpture growth rate experiment grew at least one lamella. One snail took longer than the duration of the experiment to grow a single lamella (> 51 days), and was excluded from the analysis. Two snails took longer than 22 days to grow a lamella (30 and 31 days), and were considered outliers (i.e., more than 3 SD away from the mean). Overall, the average time to grow one complete lamella, including the intervening flat portion of shell produced since the last lamella, was 9.46 [+ or -] 0.6 days (SE) (median = 8 days, mode = 8 days), or 8.3 [+ or -] 0.4 days (SE), with data from outliers removed (Fig. 4A). During the growth of a lamella, the "no lamella" phase took an average of 1.9 [+ or -] 0.2 days (SE) (excluding outliers). The "incipient lamella" phase was much more subjective to score, but took a similar length of time to complete. It took an average of 5.5 [+ or -] 0.4 days (SE) (excluding outliers) from the end of the "incipient lamella" phase to complete the axial sculpture and begin the next "no lamella" phase. This time period would include any pause in growth between lamellae.
The spacing between adjacent lamellae generally declined .with increasing time to grow a lamella, but this relationship was not statistically significant unless outliers were included (Fig. 4B). When outliers were included, lamellae that took less time to grow were spaced farther apart. No small snails (smallest one third of the group had < 25 mm shell length) required more than 10 days to grow a lamella.
How does feeding rate affect shell sculpture and growth rate?
For shell length, shell weight, and spiral growth, snails in the "High" feeding treatment grew significantly and, not surprisingly, more than snails in the "Low" feeding treatment (P < 0.007; Appendix, Fig. A1A, B; data for growth in shell length and weight not shown).
When compared to initial shell length, the spacing between lamellae decreased in larger snails (Appendix, Fig. A 1C, D). "High" feeding rate snails had significantly greater spacing between lamellae than "Low" feeding rate snails, when adjusted for size (P = 0.03). In contrast, relative to daily spiral shell growth, lamellar spacing increased in snails that grew more (Fig. 5A), and did not differ between feeding treatments after standardizing for the amount grown (Fig. 5B).
Time to grow a lamella
The average time to grow a lamella increased in larger snails (Appendix, Fig. A1E, F), and decreased in snails that grew more. "High" feeding rate snails took 16.3 [+ or -] 3.8 days, "Medium" feeding rate snails took 19.8 [+ or -] 2.8 days, and "Low" feeding rate snails required 23.0 [+ or -] 3.9 days, on average, to grow a lamella. When adjusted for body size, "High" treatment rate snails grew lamellae significantly faster than "Low" feeding rate snails (P = 0.02). The time needed to grow a lamella was approximated by dividing the time by the number of lamellae grown.
Snails ate more barnacles later in the sculpture growth rate experiment (second time period vs. the first: Wilcoxon signed-rank test: W = 65, P = 0.0001), and larger snails ate more barnacles than did smaller snails (second time period; Appendix, Fig. A2A). The number of barnacles eaten in 1 week correlated significantly with the number of lamellae grown in the same week (Spearman's rank correlation coefficient: [r.sub.s] = 0.21, P = 0.003; Appendix, Fig. A2B). Of the 33 snails, 20 completed more lamellae when more barnacles were eaten, whereas only 7 snails completed more lamellae when fewer barnacles were eaten. When the total number of lamellae grown was compared to the average feeding rate, the relationship was not significant (P = 0.28). Although more lamellae were grown, snails that ate more barnacles did not grow lamellae significantly faster (Spearman's correlation: [R.sub.s] = -0.02, P = 0.84; Appendix, Fig. A2C).
Do existing lamellae affect shell growth rate or growth of new lamellae?
Neither removal of the apertural lamella nor removal of all body whorl lamellae had a significant effect on the subsequent shell growth rate of Nucella lamellosa, or on the spacing and number of lamellae produced. This finding was true with log-log regressions against shell length, as well as when least-squares means were calculated to account for body size (see next section). All analyses of this experiment were therefore pooled to look at growth rate trends.
Shell growth rate
Larger snails grew more slowly than smaller snails, both in shell length change (df = 10, [r.sup.2] = 0.46, P < 0.0001; data not shown) and spiral growth (Fig. 6A). Snails whose shell length was greater than 29 mm appeared to grow more slowly, although the change in slope was not significant for length (P < 0.077) or spiral growth (P < 0.17); few snails in the experiment were that large. Neither removal of lamellae nor removal of the aperture had a significant effect on the change in shell length (P > 0.98; data not shown) or on spiral growth (Fig. 6B).
Increases in shell weight did not vary significantly with initial shell length when all snails were included (P = 0.77; data not shown). However, when only the snails whose initial weight was less than 1.8 g were included, larger snails increased in mass slightly faster than the smaller snails ([r.sup.2] = 0.05, P = 0.023; data not shown). The sculpture removal treatment had no effect on the rate of shell weight gain (data not shown).
Time to grow a lamella
The time to grow a lamella was approximated by dividing the total time in the treatment by the number of lamellae grown. The average time taken to grow a lamella was 14.3 [+ or- ] 0.5 days (SE) (13.7 days [+ or- ] 0.4 (SE) with snails smaller than 29 mm). It increased with initial shell length (Fig. 6C), and decreased with the amount of spiral growth (df =85, [r.sup.2] = 0.62, P < 0.0001; data not shown).
Lamellar spacing decreased in larger snails (Fig. 6D), and increased with the rate of spiral growth (df = 85, [r.sup.2] = 0.87, P < 0.0001; data not shown). Lamellar spacing did not differ significantly between treatments in the sculpture removal experiment (data not shown).
Growth rate (increase in shell length) was the primary determinant of lamellar growth dynamics in this study. Across the various experiments, neither feeding rate nor presence of past varices had any direct effect on the rate of lamellar production or on lamellar spacing. As expected, increased size and decreased feeding rate were generally associated with lower rates of shell length increase (Spight, 1981). However, in our experiments, there was no evidence of an interactive effect of these factors on the rate of lamellar growth. Shell manipulation (removal of shell sculpture or the aperture) had no effect on overall shell growth rate, and did not change the spacing or timing of lamellae.
Are lamellae associated with a growth hiatus?
Although lamellae in Nucella lamellosa resemble axial varices in other muricid shells, and a growth hiatus is typically observed after completion of a varix (MacKenzie, 1961; Inaba, 1967; MacGinitie and MacGinitie, 1968; Spight et ai, 1974; Spight and Lyons, 1974; Illert, 1981), we observed no evidence of a growth hiatus after completion of a lamella. If spiral growth paused significantly between lamellae in N. lamellosa, we would predict that in a static sample of snails from the field, most would bear a complete apertural lamella at the apertural lip, and few snails would be observed between lamellae. Larger snails would also be expected to spend more time paused due to their slower overall growth rate. Most N. lamellosa of all size classes that were collected from the field were in the process of growing a lamella (Fig. 2), a finding that suggests that significant pauses are not associated with lamellar growth, unlike varices.
Lamellar spacing and lamellar growth rate
In most varix-bearing muricids, varices are regularly spaced and aligned with previous varices on the preceding shell whorl (Webster et ai, unpubl. data; Vermeij, 1995; Savazzi and Sasaki, 2004); however, regular spacing and alignment between whorls were not apparent in the lamellae of N. lamellosa. The spacing between lamellae was highly variable within individuals, and variability also differed among individuals (Fig. 3B). Although some snails showed fairly regular lamellar spacing (Fig. 3C), no snail showed a complete whorl, in which lamellae were aligned with those on the previous whorl. Therefore, the developmental control of lamellar placement appears to differ quite significantly from the presumed mechanism for varices (Vermeij, 1995; Savazzi and Sasaki, 2004; Seilacher and Gishlick, 2014). Overall shell growth rate clearly affected lamellar spacing. Larger snails, as well as snails that were fed less, had lower rates of shell length increase, resulting in lamellae grown at a slower rate and lamellae that were closer together, on average. Lamellar spacing was directly related to spiral growth (growth rate; Fig. 5A) in the feeding experiment, but there was no additional effect of feeding treatment (Fig. 5). Snails that were fed more did grow lamellae farther apart when compared to initial shell length, a proxy for body size (Appendix, Fig. AID); the feeding rate indirectly changed lamellar spacing by affecting the spiral growth rate.
Although shell growth has been modeled geometrically in many types of shells (Ackerly, 1989; Hammer and Bucher, 2005; Meinhardt, 2009; Chirat et ai, 2013), these models have not been tested for biological relevance. Surprisingly, little is known about how the rate of lamellar accretion compares to normal spiral accretionary growth. When fed ad libitum, N. lamellosa grew a new lamella in approximately 1-2 weeks. Over this cycle, from one lamella to the next, we observed that about one third of the time was spent growing intervening shell ("no lamella" and "incipient lamella;" Fig. 2, inset), and the other two thirds was devoted to growing the lamella itself, including any possible pause in growth. This finding corresponded roughly to an accretion rate of 1.6 [+ or- ] 0.1 mm/day (SE) in between lamellae, and 0.8 [+ or- ] 0.08 mm/day (SE) when growing a lamella (n = 27, measured at the P1 shoulder cord).
The lamellar accretion rate may be significantly lower than the spiral accretion rate (Paired r-test; t = 3.89, df = 26, P = 0.0006) for many possible reasons. 1) A short growth hiatus may be associated with completion of a lamella, although not nearly as long what was seen for varices. 2) The relative thickness of the lamella compared to the body whorl may differ. 3) The growing margin of a lamella is greatly expanded, with added corrugation not seen between lamellae (Fig. 1). These undulations increase the total length of the secretory edge, as well as the length and marginal area of the mantle during lamellar growth; time may be required to physically expand the mantle, then shorten it again for the next phase. This last point is consistent with the models of Chirat et al. (2013), who showed that wavy or spiny edges could arise from excessive marginal growth (expansion of the mantle). It would be interesting to document how the mantle expands, then retracts, during successive iterations of lamellar growth. Mantle expansion is not simply a muscular stretching. Individuals relaxed in Mg[Cl.sub.2] had much shorter (perpendicular to the growth edge) mantles when they were between lamellae than when they were in mid-lamella (N. B. Webster, pers. obs.). Instead, the mantle must be physically expanding, but exactly how it expands remains unclear.
Even though individuals were held under the same conditions, the time taken to grow a lamella varied widely (Fig. 4A). A few snails took much longer (e.g., one snail required up to 51 days), but these were likely outliers due to extreme natural variation, a slowing of growth with increasing size, or some other factor. Such wide, among-individual variation is not uncommon, even under controlled conditions (Spight, 1981; Koehn and Shumway, 1982; Burrows and Hughes, 1990).
The time taken to grow a lamella did appear to depend, at least weakly, on its distance from the previous one. The more rapidly grown lamellae were generally spaced more widely apart (Fig. 4B), although this relationship was not statistically significant if outliers were excluded. The weakness of this association, at least in part, was that lamellae were scored only once per day. So, in an extreme case, if a lamella were completed after 4.1 days, it would be scored as 5 days, which is a difference of 18%. This overestimate of time taken would be more pronounced among faster-growing snails, and would artificially decrease the slope in Fig. 4B.
Finally, the lamellar growth rate and spacing likely depend on how close snails are to maturity. The size at which snails reach maturity varies greatly, even within species (Spight, 1973), and some of the variation that we observed may have arisen from some snails reaching maturity at smaller sizes. Nucella lamellosa, like many other snails (Spight, 1981), drastically decrease growth rate with maturity, even though growth is not determinate (Spight, 1973). Maturity also appeared to have an effect on lamellar spacing in N. lamellosa. Many larger snails stopped producing separate lamellae, and instead grew a terminal frill (Fig. 1C). This frill appears to be the result of lamellae being grown closer together when the spiral growth rate is reduced at larger sizes. However, not all large snails grew a frill, nor were all frills terminal, and no snails were dissected to confirm sexual maturity.
Effect of existing lamellae on shell growth rate and growth of new lamellae
Previously produced axial sculpture is a potential impediment to subsequent shell growth, because it must be reabsorbed before additional spiral growth may proceed (Carriker, 1972; Vermeij, 1977). We therefore expected that removal of N. lamellosa lamellae would affect either the spiral growth rate or the spacing of new lamellae. However, despite thoroughly shaving the shells, neither removing the aperture (damaging the shell) nor removing all remaining lamellae from the body whorl had any effect on any metric of shell or sculpture growth. We therefore reject the hypothesis that the rate of spiral growth or the placement of new lamellae is affected by previous lamellae.
This result was surprising. Prominent shell sculpture necessarily impedes further growth if it is not removed from the front of the aperture, although nothing is known about the costs of this process (Vermeij, 1977). This suggests a relatively low cost of sculpture removal, and does not contradict the findings of Palmer (1981), who reported that the deposition rate of the shell is the rate-limiting step, not energy, somatic growth, or sculpture resorption. Muricids use a system of shell removal to drill their prey (Carriker, 1981), and a similar mechanism may aid in the removal of sculpture in front of the expanding aperture. The ability to dissolve shell sculpture appears to be derived in gastropods, as internal shell remodeling is absent in most basal groups except the Neritidae (Vermeij, 1973). More basal lineages of gastropod taxa generally do not have elaborate shell sculpture, probably due to an inability to resorb it efficiently (Vermeij, 1977). We therefore predict that more basal snails bearing sculpture grow faster when shell sculpture is removed experimentally.
Comparing lamellae and varices
Muricid varices, whether blade-like or an array of spines, are typically regularly spaced and exhibit a fixed angle within species (Webster et al., unpubl. data; Vermeij, 1995; Savazzi and Sasaki, 2004). The lamellae of N. lamellosa do not follow this stereotypical growth pattern, and have relatively plastic positioning. Famellae and blade-like varices appear superficially similar, and are thought to be primarily defensive (Vermeij et al., 1981; Carefoot and Donovan, 1995; Miller and FaBarbera, 1995; Donovan et al, 1999). But the varices of other muricids are associated with periodic bursts of shell growth. In larger specimens with the mature arrangement of varices, long pauses of up to several months occur between bursts of fast growth from one varix to the next (Inaba, 1967; MacGinitie and MacGinitie, 1968; Spight and Fyons, 1974; Illert, 1981). Famellar growth in N. lamellosa, on the other hand, was completed in 1-2 weeks and was not associated with any such hiatus. In addition, varices are also associated with a period of shell thickening (Inaba, 1967; Carriker, 1972) and a stereotypical lip, but neither was present in N. lamellosa (Faxton, 1970). Thus, we predict that in muricids with varices, factors influencing shell growth rate affect the length of the pause between varices, but not the time taken to grow the varix itself, nor the spacing of varices; some muricids do not eat during shell growth (Carriker, 1972). We feel that these differences in growth between the axial sculpture of N. lamellosa differ sufficiently from the stereotypical varix pattern to warrant the separate term "lamellae." An in-depth discussion of the origin and evolution of varices, including clarification of the different forms of axial sculpture, is forthcoming (Webster et al., unpubl. data).
Growth and development of lamellae
Most shell sculpture is thought to have a defensive function (Vermeij, 1987, 1995), and the alignment of varices appears to add to their functionality (Spight and Lyons, 1974; Carefoot and Donovan, 1995; Donovan et al., 1999). But what about lamellae? Our results suggest that the rate at which lamellae are produced depends mostly on factors that affect shell growth rate. When spiral growth rate is relatively constant, the spacing and timing of lamellae should also be relatively constant. The variation in sculpture patterning suggests that the spacing of lamellae is neither a priority nor regulated by some higher-level control mechanism. If a particular lamellar spacing were adaptive, then it should be constant regardless of shell growth rate. However, lamellae appear to be added in a pattern in which the likelihood of having an apertural lamella at a given point in time remains relatively constant. When conditions are good and growth is fast, the snail generally adds lamellae faster, yet farther apart. When growth is slow, N. lamellosa place lamellae closer together. The likelihood of producing an apertural lamella, therefore, appears to remain relatively constant regardless of environmental conditions, and may ensure that apertural reinforcement does not depend on external factors, such as variation in food availability or water temperature.
The Nucella lamellosa shell form is known to be quite plastic, changing readily between thin-shelled, with frilly lamellae, and thick-shelled, with few or no lamellae (Appleton and Palmer, 1988). This plasticity would not be possible if new lamellae depended on physical cues from previous lamellae for positioning. Furthermore, a mechanism to control lamellar spacing might be difficult to evolve under these circumstances. If snails regularly transition to a form without lamellae, the opportunity for selection of coordinated lamellar placement would be lessened. It is unclear how actively the plastic shell response to predator cues and the apparently passive plasticity of axial lamellar growth interact with one another. Even though shell plasticity is not uncommon (Trussed, 2000; Hoverman and Relyea, 2009; Pascoal et al., 2012), our understanding of the role of plasticity in evolution is just beginning (Laland et al., 2014; Forsman, 2015).
In snails, larger juveniles generally grow more quickly than smaller snails, but growth rate then drops off again with maturity (Spight, 1981). If the rate of production of lamellae is correlated with the growth rate, then similar trends should be seen for lamellar spacing. Diplommatinid land snails bear sharp axial sculpture, termed "ribs," which are quite similar to axial lamellae. One species (Plectostoma retrovens) reportedly grows ribs nightly, and will grow faster if it is kept in the dark (Berry, 1962). Another species, Plectostoma concinnum, grows ribs more slowly, and the spacing of ribs correlates with shell growth rate (Liew et al., 2014). Rib spacing increases with increased growth rate initially, then begins to decrease again with maturity, as diplommatinids have determinate growth. Similarly, Taylor et al. (2004) examined the spacing of lamellae in the bivalve Lucina pensylvanica. Within individuals, lamellae grew further apart as the bivalve grew larger, up to a length corresponding to sexual maturity; then lamellar spacing decreased with an increase in variability. In Nucella lamellosa, larger snails had lamellae closer together, but this was an intraspecific pattern, and did not include very small snails. When the distance between all measurable lamellae (including those grown prior to collection) of an individual snail was measured, the general trend in Nucella lamellosa followed that of Plectostoma concinnum and Lucina pensylvanica: increased lamellar spacing with increasing juvenile size, followed by an irregular decrease as maturity approached (Appendix, Fig. A3). The variation in spacing is much higher in N. lamellosa than in these other two mollusc species, and the size at which lamellar spacing begins to decrease also varies. The most distinct period at which lamellar spacing increased occurred prior to collection, which explains why we did not detect it during our experiments. Interestingly, laboratory rearing had no obvious effect on lamellar spacing. That such disparate species, with different shell sculpture, exhibit such a similar growth pattern strongly suggests a conserved process, which may be widespread among mollusc taxa bearing axial sculpture. Without a specific and separate mechanism to regulate the spacing of such sculpture, though, the underlying pattern seems to be to produce lamellae at more or less regular time intervals, which results in a spacing that depends on the rate of spiral shell growth.
Parallel color patterns on shells have been modeled based on reaction-diffusion, lateral inhibition, and neurosecretory models (Ermentrout et al., 1986; Boettiger et al., 2009; Meinhardt, 2009). Similar mechanisms may also apply to the control of axial sculpture. Although several theoretical mechanisms have been proposed (Hammer, 2000; Moulton et al., 2012; Chirat et al., 2013), the actual biological processes responsible for the growth of axial sculpture remain poorly studied from either an experimental or a developmental point of view. Both morphological and physiological changes to the mantle are theoretically required each time a lamella is produced, leaving room for developmental plasticity to affect each lamella separately, and adding to the overall variability in shell sculpture growth. Clearly, much remains to be learned about the signaling pathways and mantle feedback mechanisms underlying the control of shell sculpture (Urdy, 2015). If we can understand how the mantle grows a shell, we can begin to understand how the striking diversity of gastropod shells evolved.
We thank Eric Clelland; our assistants, Jared Sykes, Carissa Keates, and Anna Smith; and the Bamfield Marine Sciences Centre for enabling us to do this work. We also thank Regis Chirat, Severine Urdy, and Lindsey Leighton for thoughtful comments on the manuscript. This research was funded by NSERC Canada (PGSD graduate scholarship to NBW, and Discovery Grant A7245 to ARP).
Abbott, R. T. 1974. American Seashells; the Marine Molluska of the Atlantic and Pacific Coasts of North America, 2nd ed. Van Nostrand Reinhold, New York.
Ackerly, S. C. 1989. Kinematics of accretionary shell growth, with examples from brachiopods and molluscs. Paleobiology 15: 147-164.
Appleton, R. D., and A. R. Palmer. 1988. Water-borne stimuli released by predatory crabs and damaged prey induce more predator-resistant shells in a marine gastropod. Proc. Natl. Acad. Sci. USA 85: 4387-4391.
Berry, A. J. 1962. The growth of Opisthostoma (Plectostoma) retrovertens Tomlin, a minute cyclophorid from a Malayan limestone hill. J. Molluscan Stud. 35: 46-49.
Boettiger, A., B. Ermentrout, and G. Oster. 2009. The neural origins of shell structure and pattern in aquatic mollusks. Proc. Natl. Acad. Sci. USA 106: 6837-6842.
Burrows, M. T., and R. N. Hughes. 1990. Variation in growth and consumption among individuals and populations of dogwhelks, Nucella lapillus: a link between foraging behaviour and fitness. J. Anim. Ecol. 59: 723-742.
Carefoot, T. H., and D. A. Donovan. 1995. Functional significance of varices in the muricid gastropod Ceratostoma foliatum. Biol. Bull. 189: 59-68.
Carriker, M. R. 1972. Observations on removal of spines by muricid gastropods during shell growth. Veliger 15: 69-74.
Carriker, M. R. 1981. Shell penetration and feeding by naticacean and muricacean predatory gastropods: a synthesis. Malacologia 20: 403-422.
Chirat. R.. D. E. Moulton, and A. Goriely. 2013. Mechanical basis of morphogenesis and convergent evolution of spiny seashells. Proc. Natl. Acad. Sci. USA 110: 6015-6020.
Donovan, D. A., J. P. Danko, and T. H. Carefoot. 1999. Functional significance of shell sculpture in gastropod molluscs: test of a predator-deterrent hypothesis in Ceratostoma foliatum (Gmelin). J. Exp. Mar. Biol. Ecol. 236: 235-251.
Ermentrout, B., J. Campbell, and G. Oster. 1986. A model for shell patterns based on neural activity. Veliger 28: 369-388.
Forsman, A. 2015. Rethinking phenotypic plasticity and its consequences for individuals, populations and species. Heredity 115: 276-284.
Gosselin, L. A. 1993. A method for marking small juvenile gastropods. J. Mar. Biol. Assoc. UK 73: 963-966.
Hammer, 0. 2000. A theory for the formation of commarginal ribs in mollusc shells by regulative oscillation. J. Molluscan Stud. 66: 383-392.
Hammer, 0.. and H. Bucher. 2005. Models for the morphogenesis of the molluscan shell. Lethaia 38: 111-122.
Hoverman, J. T., and R. A. Relyea. 2009. Survival trade-offs associated with inducible defences in snails: the roles of multiple predators and developmental plasticity. Fund. Ecol. 23: 1179-1188.
Illert, C. 1981. The growth and feeding habits of a South Australian murex. Sea Shore 12: 8-10.
Inaba, A. 1967. The growth of Chicoreus asianus. Venus 26: 5-7.
Kincaid, T. 1957. Local Races and Clines in the Marine Gastropod Thais lamellosa Gmelin: a Population Study. The Calliostoma Company, Seattle, WA.
Koehn, R. K., and S. E. Shumway. 1982. A genetic/physiological explanation for differential growth rate among individuals of the American oyster, Crassostrea virginica (Gmelin). Mar. Biol. Lett. 3: 35-42.
Laland, K., T. Uller, M. Feldman. K. Sterelny, G. B. Muller, A. Moczek, E. Jablonka, J. Odling-Smee, G. A. Wray, H. E. Hoekstra et al. 2014. Does evolutionary theory need a rethink? Nature 514: 161-164.
Laxton, J. H. 1970. Shell growth in some New Zealand Cymatiidae (Gastropoda: Prosobranchia). J. Exp. Mar. Biol. Ecol. 4: 250-260.
Lenth, R. V. 2014. R Package Lsmeans: Least-Squares Means [Online]. Available: http://CRAN.R-project.org/package=lsmeans. [2015, August 29],
Liew, T.-S., A. C. M. Kok, M. Schilthuizen, and S. Urdy. 2014. On growth and form of irregular coiled-shell of a terrestrial snail: Plectostoma concinnum (Fulton, 1901) (Mollusca: Caenogastropoda: Diplommatinidae). PeerJ 2: e383.
MacGinitie, G. E., and N. MacGinitie. 1968. Natural History of Marine Animals, 2nd ed. McGraw-Hill, New York.
MaeKenzie, C. L., Jr. 1961. Growth and reproduction of the oyster drill Eupleura caudata in the York River, Virginia. Ecology 42: 317-338.
Meinhardt, H. 2009. The Algorithmic Beauty of Sea Shells. Springer, Berlin.
Miller, D. J., and M. LaBarbera. 1995. Effects of foliaceous varices on the mechanical properties of Chicoreus dilectus (Gastropoda: Muricidae). J. Zool. 236: 151-160.
Moulton, D. E., A. Goriely, and R. Chirat. 2012. Mechanical growth and morphogenesis of seashells. J. Theor. Biol. 311: 69-79.
Palmer, A. R. 1979. Fish predation and the evolution of gastropod shell sculpture: experimental and geographic evidence. Evolution 33: 697-713.
Palmer, A. R. 1981. Do carbonate skeletons limit the rate of body growth? Nature 292: 150-152.
Palmer, A. R. 1982. Growth in marine gastropods: a non-destructive technique for independently measuring shell and body weight. Malacologia 23: 63-73.
Palmer, A. R. 1985. Adaptive value of shell variation in Thais lamellosa: effect of thick shells on vulnerability to and preference by crabs. Veliger 27: 349-356.
Pascoal, S., G. Carvalho, S. Creer, J. Rock, K. Kawaii, S. Mendo, and R. Hughes. 2012. Plastic and heritable components of phenotypic variation in Nucella lapillus: an assessment using reciprocal transplant and common garden experiments. PLoS One 7: e30289.
Radwin, G. E., and A. D'Attilio. 1976. Murex Shells of the World: an Illustrated Guide to the Muricidae. Stanford University Press, Stanford, CA.
Rasband, W. S. 1997. Image], U. S. National Institutes of Health, Bethesda, MD. [Online], Available: http://imagej.nih.gov/ij/. [2015, August 28].
Savazzi, E., and T. Sasaki. 2004. Synchronized sculpture in gastropod shells. Am. Malacol. Bull. 18: 87-114.
Seilacher, A., and A. D. Gishlick. 2014. Gastropod heteromorphy: or how to get out of the spiral syndrome. Pp. 256-273 in Morphodynamics. CRC Press, Boca Raton. FL.
Spight, T. M. 1973. Ontogeny, environment, and shape of a marine snail Thais lamellosa Gmelin. J. Exp. Mar. Biol. Ecol. 13: 215-228.
Spight, T. M. 1981. The ecology of body growth: environmental influences on the growth of marine snails. Ecosynthesis 1: 257-344.
Spight, T. M., and A. Lyons. 1974. Development and functions of the shell sculpture of the marine snail Ceratostoma foliatum. Mar. Biol. 24: 77-83.
Spight, T. M., C. Birkeland, and A. Lyons. 1974. Life histories of large and small murexes (Prosobranchia: Muricidae). Mar. Biol. 24: 229-242.
Taylor, J. D., E. A. Glover, M. Peharda, G. Bigatti, and A. Ball. 2004. Extraordinary flexible shell sculpture: the structure and formation of calcified periostracal lamellae in Lucina pensylvanica (Bivalvia: Lucinidae). Malacologia 46: 277-294.
Trussed, G. C. 2000. Phenotypic dines, plasticity, and morphological trade-offs in an intertidal snail. Evolution 54: 151-166.
Urdy, S. 2015. Theoretical modelling of the molluscan shell: what has been learned from the comparison among molluscan taxa? Pp. 207-251 in Ammonoid Paleobiology: from Anatomy to Ecology, Topics in Geobiology 43, C. Klug, D. Korn, K. De Baets, I. Kruta, and R. H. Mapes. eds. Springer. New York.
Vermeij, G. J. 1973. Adaptation, versatility, and evolution. Syst. Biol. 22: 466-477.
Vermeij, G. J. 1977. The mesozoic marine revolution: evidence from snails, predators and grazers. Paleobiology 3: 245-258.
Vermeij, G. J. 1981. Apertural form in gastropods. Lethaia 14: 104.
Vermeij, G. J. 1987. Evolution and Escalation: an Ecological History of Life. Princeton University Press, Princeton.
Vermeij, G. J. 1995. A Natural History of Shells. Princeton University Press. Princeton, NJ.
Shell growth in feeding experiment
Number of barnacles eaten
Cumulative lamellar spacing
Cumulative lamellar spacing
NICOLE B. WEBSTER (1,2,*) AND A. RICHARD PALMER (1,2)
(1) Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9; and (2) Bamfield Marine Sciences Centre, Bamfield, British Columbia, Canada V0R 1B0
Received 28 April 2015; accepted 13 November 2015.
(*) To whom correspondence should be addressed. E-mail: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Webster, Nicole B.; Palmer, A. Richard|
|Publication:||The Biological Bulletin|
|Date:||Feb 1, 2016|
|Previous Article:||Role of the substrate in feeding and growth of the marine suspension-feeding gastropods Crepidula fornicata and Crepipatella peruviana.|
|Next Article:||Entrainment of the circadian rhythm in egg hatching of the crab Dyspanopeus sayi by chemical cues from ovigerous females.|