Of all gastropods, whelks in the Family Muricidae have the most extravagant shell ornamentation, and the function of these spires, flanges, and spikes has been a subject of much speculation.  In most cases, too little is known of the biology of a species to make more than an educated guess.  Suggestions for functions fall in the following 4 broad categories: 1) affecting falling and subsequent landing orientations, 2) protecting through camouflaging or shell-strengthening, 3) aiding in feeding or sensory input, and 4) stabilising the shell in currents, waves, or shifting substrata.  Of these, only the first 2 have received specific research attention.  For example, although there are at least 26 west-coast species of shell-crushing fishes, their effect, if any, on local whelk populations is basically unknown.  Moreover, defensive shell sculpturing in gastropods as, for example, stout spines around the periphery of the shell, while common in tropical species, is virtually absent in gastropods north of 38^o N latitude. Decrease in presence of such spines appears to be directly correlated with fewer shell-crushing fishes towards the north.  For review, see Palmer 1979 Evolution 33: 697.

Murex sp. shells are found in the Indo-Pacific
region. If the function of its spines is for
defense, the predators are unknown 1X

Among northern west-coast shallow-water whelks, the leafy hornmouth shell Ceratostoma foliatum, a predator on barnacles and bivalves, has probably the most elaborate shell ornamentation.  There are 3 varices, sometimes termed flanges, left, middle, and right.  The left and right varices form a broad platform surrounding the shell aperture.  Growth involves adding a new right varix, which causes the former right varix to shift to the middle, and the former middle varix to shift to the left.  This happens about once per year in older animals held under laboratory conditions and takes 1-2mo for completion.  Field animals may do it quicker and/or in secrecy, as animals with partially formed varices are not commonly seen in the field.  Spight & Lyons 1974 Mar Biol 24: 77.

One proposed function for the varices in Certatostoma is to destabilise its falling orientation such that if dislodged from attachment on a vertical surface, and provided there is an unimpeded fall path of at least 20 body lengths, it will land upright 35-70% of the time.  Experiments at the Bamfield Marine Sciences Centre, British Columbia show that upright landings occur only slightly more than predicted by chance (56 versus 50%) but the results are statistically significant.  Tests on shells with different varices removed show that the destabilisation owes to the “rudder”-like effect of the middle varix that tends to rotate the shell during the fall to an aperture-down orientation.  Without the middle varix the animal’s centre of gravity would cause it to fall in an aperture-up orientation. The protective advantage to the snail in landing upright is obvious.  Palmer 1977 Science 197: 1293.

NOTE  a stable falling orientation would be one with centre of gravity downwards, and the animal would tend to land on the left or middle varices with the aperture facing upwards

Later studies at the Bamfield Marine Sciences Centre, British Columbia, however, suggest that most Ceratostoma do not occupy habitats that on dislodgement would allow unimpeded falls of at least 20 body lengths.  Rather, they live in rock- and seaweed-strewn habitats in which dislodgement leads to no fall at all, or to only short bump-and-tumble falls.  For example, of 100 individual Ceratostoma tested in the field by SCUBA-divers, only 14 when dislodged fell more than 5 body lengths, and none fell further than 9 body lengths.

If a Ceratostoma is dislodged by a wave or predator and doesn’t safely end up aperture-down, then it must right itself as quickly as possible to minimise risk of predation on its exposed soft parts.  Of the 2 possible non-aperture-up landing orientations, one is with the shell resting on the right and middle varices, and righting is through an arc of 130^o (see left-hand drawing and photo in figure on Right).  This is termed the "hard" landing orientation. The second landing possibility is with the shell resting on the left and middle varices . The offset nature of the middle varix means that the aperture is close to vertical, and righting is through an arc of 110^o (see right-hand drawing and photo in figure on Right). This is termed the "easy" landing orientation. Note that larger snails will have relatively greater difficulty in righting than smaller snails, as expected.

Not only does it take almost twice as long to right from the “hard” landing orientation (see graph upper Left), but oxygen uptake is significantly greater and, therefore, overall energy expenditure is significantly greater by 4-fold.  Righting costs for a 10g individual are 261 mJoules from the “hard” orientation, and 66 mJ from the “easy” orientation. Clearly, if not landing aperture down after dislodgement, it is advantageous for Ceratostoma to end up in the “easy” landing orientation. 

If unimpeded falls of sufficient height to lead to destabilisation are rare, then what are the chances of a dislodged snail landing aperture-down or in the “easy” orientation from more realistic short-distance heights?  To determine this snails are dropped from a height of 5 body-lengths through seawater from random starting orientations and landing frequencies are recorded.  Results show that 48% land in the aperture-down orientation, 15% in the “hard” orientation, and 37% in the “easy” orientation.  

How do these results compare with expected landing frequencies? To calculate these, imagine the shell as a 3-sided die, with expected landing frequencies being equivalent to the extent of circumference of the shell occupied by each varix pair (see drawing lower Left).  These are 50% for the right/left varix pair or aperture-down position), 31% for the “hard” orientation resting on right and middle varices, and 19% for the “easy” orientation resting on left and middle varices.  Note that while the actual landing frequency for aperture-down is similar to expected landing frequency based on extent of circumference occupied, the other 2 sets of frequencies differ significantly, with a preference for landing in the "easy" orientation. In summary, after short falls Ceratostoma lands significantly more on the varix pair from which righting is easier, and significantly less on the varix pair from which righting is harder. Carefoot & Donovan  1995 Biol Bull 189: 59.

NOTE  137 live but withdrawn adult Ceratostoma foliatum each dropped 10 times in water from random starting orientations, with landing frequencies being averaged for each individual

NOTE expected landing probabilities for a 3-sided die would, of course, be 33 1/3% for each of the 3 sides

The broad implication from the above studies on Ceratostoma is that a quick return to the stable, aperture-down, foot-attached position after dislodgement is favoured, presumably as protection against predators. Because the right varix ultimately becomes the middle varix at each growth spurt, then its presence and size and, thus, its later effect, is important to a snail’s survival.  In fact, measurements show that the area of the right varix does not scale isometrically with body length; rather, it scales allometrically with a slope of 2.15.  What this means is that as the snail grows, its right varix area and, by association, later the middle varix area, increases out of proportion, thus providing relatively greater effect with increasing size and age of the snail.  Palmer 1977 Science 197: 1293; Carefoot & Donovan  1995 Biol Bull 189: 59.

NOTE if the relationship of varix area to length were isometric, then the slope would be 2.  Anything that differs significantly from this value,in this case, 2.15, is allometric

What animals prey on Ceratostoma foliatum and do the shell varices provide protection?  This is tested in cafeteria-style experiments at the Bamfield Marine Science Centre, British Columbia using sunflower stars Pycnopodia helianthoides and red-rock crabs Cancer productus. Each predator is presented with 5 treatment-types of Ceratostoma: 1) intact (CONTROL), 2) right varix removed (R), 3) middle varix removed (M), 4) left varix removed (L, see photograph on Left), and 5) ALL varices removed.  Several replicates are run for each predator.  Eaten snails are replaced, so the number of prey available remains the same over time. 

After 10wk the researchers find that sea stars Pycnopodia preferentially eat snails with ALL varices removed and there is no significant difference in numbers eaten of the other treatment types (see graph lower Right).  The prey is swallowed whole, and it seems that Pycnopodia prefers a smoother-bodied prey, for ease of intake and/or less pokey-ness around the soft mouth tissues. 

Red-rock crabs Cancer productus also prefer snails with all varices removed, but with less well-defined preferences than shown by Pycnopodia.  Generally, snails lacking varices (ALL) or with the right varix removed (R) are preferred over the other treatment types.  Cancer’s mode of eating is to crack and peel starting at the right side of the aperture, so absence of the right varix makes the job easier (see graph upper Right).  If both types of predator prefer smooth-shelled prey, then how will they respond to a choice of intact Ceratostoma, varices-removed Ceratostoma, and smooth-shelled morphs of Nucella lamellosa of identical body lengths? 

The answer for Pycnopodia after an 8-wk test is a strong preference for smooth-shelled Nucella, again possibly owing to degree of smoothness of shell (see histogram upper Left).  When Ceratostoma’s varices are removed by grinding, they are not removed completely.  Rather, a small ridge of thickened shell representing the varix base is intentionally left to avoid risk of serious weakening the shell.  This extra roughness may be enough to turn off the predator. 

The comparable data for predation on these different shell types by Cancer productus is a moderate, but significant, preference for Ceratostoma with ALL varices removed, followed by intact (CONTROL) Ceratostoma, and then by smooth-shelled Nucella (see histogram lower Left).  This makes sense because Nucella’s shell is much stronger than Ceratostoma’s, and the crab’s usual behaviour of chipping open a snail from the aperture, then cracking it in half, would make an intact Ceratostoma more difficult to open than one with its varices ALL removed. 

Overall, the experiments show that the varices on C. foliatum provide protection from both crab and sea-star predators.  A nice follow-up experiment to test this might be to offer to Pycnopodia a choice of smooth-morph versus sculpted-morph of Nucella lamellosa.  Donovan et al. 1999 J Exp Mar Biol Ecol 236: 235.

What other reasons might there be for Pycnopodia’s preference for Nucella?  Think about the suggestions below, then CLICK HERE for explanations.

Nucella has more nutrients and/or energy. 

Nucella may be more tasty to the predator than Ceratostoma. 

Nucella might be easier for Pycnopodia to digest. 

Nucella is easier for the sea star to capture.