Leonardo’s Mountain of Clams and the Diet of Worms

by Stephen Jay Gould

  19

  TRIUMPH OF THE ROOT-HEADS

  I AM NOT MUCH OF A BETTING MAN. FOR ME, A MAN O’ WAR IS AN OLD British fighting ship, and a Native Dancer inhabits Tahiti, wears grass skirts, and gyrates on the beach for Fletcher Christian and Captain Bligh in various Hollywood versions of Mutiny on the Bounty. Nonetheless, if compelled to put up or shut up, I would make an unconventional wager on the controversial subject of progress in evolution.

  In our culture’s focal misunderstanding of evolution, most people assume that trends to increasing complexity through time must impart a primary and predictable direction to the history of life. But Darwinian natural selection only yields adaptation to changing local environments, and better function in an immediate habitat might just as well be achieved by greater simplicity in form and behavior as by ever-increasing complexity. Thus, one might predict that cases of evolutionary simplification will be just about as common as increases in complexity.

  But I would be tempted to bet on culture’s underdog, and to suspect that examples of simplification might actually hold a small overall edge. I hazard this unconventional proposition because a common lifestyle assumed by tens to hundreds of thousands of animal species—namely, parasitism—usually involves evolutionary simplification of adult form in comparison with free-living ancestors. Since I know of no comparable phenomenon that could supply a countervailing bias for complexity, a compendium of all cases might produce a majority for simplification—as natural selection in free-living forms imparts no bias in either direction, while parasitism gives a clear edge to simplification.

  I regard this argument as impeccable—in its own restricted way. But nature scorns such crimping limits imposed by frailties of human cognition upon her wonderful and multifarious variety. This argument about parasites only works under the aegis of another bias almost as serious as our equation of evolution with progress: our prejudice for regarding adult anatomy as the organism, and our failure to consider entire life cycles and complexities of physiological function.

  Consider one of the standard “laments” or “stories of wonder” in conventional tales of natural history: the mayfly that lives but a single day (a sadness even recorded in the technical name for this biological group—Ephemoptera). Yes, the adult fly may enjoy only one moment in the sun, but we should honor the entire life cycle and recognize that the larvae, or juvenile stages, live and develop for months. Larvae are not mere preparations for a brief adulthood. We might better read the entire life cycle as a division of labor, with larvae as feeding and growing stages, and the adult as a short-lived reproductive machine. In this sense, we could well view the adult fly’s day as the larva’s clever and transient device for making a new generation of truly fundamental feeders—the insect equivalent of Butler’s famous quip that a chicken is merely the egg’s way of making another egg.

  This essay treats the most celebrated story of extreme simplification in an adult parasite—in the interests of illuminating, reconciling, and, perhaps, even resolving two major biases that have so hindered our understanding of natural history: the misequation of evolution with progress, and the undervaluing of an organism by considering only its adult form and not the entire life cycle.

  The adult of Sacculina, the standard representative of a larger group with some two hundred species, the Rhizocephala, could hardly be more different from its barnacle ancestors—or more simplified in anatomy and appearance. The two names accurately record this dramatic evolutionary change—for Sacculina is a Latin “little sac,” while Rhizocephala is a Greek “root-head.” As we shall see, the rhizocephalans are clearly barnacles by ancestry, but the adult preserves not a hint of this crustacean past. Rhizocephalans are parasites upon other crustaceans, and nearly all infest decapods (crabs and their relatives). The adult consists of two parts with names that (in a refreshing change from usual practice) almost count as vernacular expressions, rather than jargon. From the outside, a human observer sees only a formless sac (called the externa) attached to the underside of the crab’s abdomen. The sac is little more than a reproductive device, containing the ovary and a passageway for introduction of males and their sperm. The externa contains no other differentiated parts—no appendages, no sense organs, no digestive tract, and no sign of segmentation at all. The fertilized eggs develop within the externa (which then operates as a brood pouch).

  But how can the externa function without any evident source of nutrition? Closer examination reveals a stalk that pierces the crab’s abdomen and connects the externa to an elaborate network of roots (called the interna). These roots may pervade the entire body of the crab. They penetrate through the hemocoelic spaces (the analogs of blood vessels) and invest many of the crab’s internal organs. They provide nutrition to the parasite by absorption from the crab’s vital fluids. In some species, roots are restricted to the abdomen, but in Sacculina they may run through the entire body, right to the ends of the appendages. (This system is not so grisly—in inappropriate human terms—as first glance might suggest. The parasite does not devour the host, but rather maintains the crab as a “life support” system.) The name Sacculina (for the most common genus) honors the externa, while the designation of the entire group—Rhizocephala, or root-head—recognizes the interna.

  These barnacle parasites have been known to zoologists since the 1780s (though the first recorder, correctly observing the release of crustacean larvae from the externa, misinterpreted the sac as an organ of the crab, induced by the parasite much as some insect larvae can commandeer a plant to grow a protective gall). Ever since this early discovery, rhizocephalans have played a classic role in conventional natural history as the standard example of maximal degeneration in parasites. Many of the foremost zoologists of Darwin’s generation highlighted Sacculina as one of evolution’s primary marvels.

  The German biologist Fritz Müller wrote a famous book in 1863 that provided Darwin with crucial early support. Müller’s book deals almost entirely with the anatomy of crustaceans, but bears the general title Für Darwin (For Darwin). Müller cited Sacculina, and its undoubted relationship with free-living barnacles, as a primary example of “retrogressive metamorphosis” in evolution. He referred to this genus as “these ne plus ultras in the series of retrogressively metamorphosed Crustacea,” and he wrote of their limited activity:

  The only manifestations of life which persist . . . are powerful contractions of the roots and an alternate expansion and contraction of the body, in consequence of which water flows into the brood-cavity, and is again expelled through a wide orifice.

  E. Ray Lankester (1847–1929), later director of the Natural History Division of the British Museum, published a famous essay in 1880 titled Degeneration: A Chapter in Darwinism. He defined degeneration as “a loss of organization making the descendant far simpler or lower in structure than its ancestor,” using Sacculina as a primary example. Lankester described the barnacle parasite as “a mere sac, absorbing nourishment and laying eggs.”

  Yves Delage (1854–1920), one of France’s finest natural historians and a patriotic Lamarckian, published a major empirical study on Sacculina in 1884. He referred to the genus as “this singular parasite, reduced to a sac containing the genital organs.” “Sacculina,” he added, “seems to be one of those beings made to chill adventurous imaginations” (faits pour refroidir les imaginations aventureuses).

  Thus, all major authors and experts used the Rhizocephala as primary illustrations of degeneration in the evolution of parasites (or, at least, of simplification if we wish to avoid the taint of moral opprobrium). I will not challenge this assertion for a restricted view of the adult as an external sac attached to internal roots. But I do wish to oppose the myopia of such a restriction. From a properly expanded viewpoint—and for three major reasons that I shall discuss in sequence—rhizocephalans are remarkably intricate animals, as bizarre in their elaborate uniquenesses as any creature on earth. In this expanded perspective, however, they remain as wonderfully provocat
ive as ever—as superbly illustrative of the meaning of evolution as when Europe’s greatest zoologists falsely appointed them as chief exemplars of Darwinian degeneration.

  1. THE FULL LIFE CYCLE OF THE RHIZOCEPHALA. How did we ever discern the barnacle ancestry of Sacculina? We could now gain this information by sequencing DNA, but early-nineteenth-century zoologists correctly identified the affinity of rhizocephalans. How did they know, especially when studies of the adult externa and interna could not provide the slightest clue?

  Observations of the complex life cycle in female rhizocephalans solved this zoological puzzle. (I shall discuss the growth of males later, as my third argument.) The first two phases of growth differ very little from the development of ordinary barnacles, and therefore seal the identification. The larvae exit from the externa’s brood pouch as a conventional dispersal stage, common in many crustaceans, called the nauplius. The rhizocephalan nauplius passes through as many as four instars (molting stages) and, except for the absence of all feeding structures, looks like an ordinary crustacean nauplius, right down to the most distinctive feature of a single median eye.

  I am trying to suppress my usual lateral excursions in this essay—if only because I find the main line of the story so exciting—but I cannot resist one digression for its striking illustration of science’s human face. Yves Delage’s 1884 monograph on Sacculina, undoubtedly the most important early study of rhizocephalans, runs to more than three hundred pages of dry anatomical description, devoted mainly to these early stages of the life cycle. But at several points he vents his anger at a German colleague, R. Kossmann. Delage took particular delight in exposing Kossmann’s error in identifying two larval eyes. Early in his monograph, this French patriot admits the source of his venom and consequent pleasure in Kossmann’s mistakes. Kossmann had previously skewered a Frenchman, a certain Monsieur Hesse, for errors in interpreting the life cycle of Sacculina. Delage took offense for two reasons. First of all, poor Hesse was a dedicated amateur who only took up the study of marine zoology in retirement, “at an age when so many others, in Germany as elsewhere, are only seeking to enjoy the inactivity of repose merited by their long service.” Kossmann should have been more generous. But second, and impossible to forgive, Kossmann had explicitly attacked Hesse as a Frenchman in clear violation of the norms of science as a cooperative and international enterprise. Delage then speculated about Kossmann’s motives and recalled his own bitter feelings at the defeat of his country in the Franco-Prussian War of 1872:

  What I cannot excuse is that this gentleman [Mr. Kossmann] expressed pleasure in seeing a scientist fall into error because that scientist is a Frenchman. This illustrates the workings of a narrow mind, and such thinking will quickly destroy the characteristic nobility of scientific discussion. But Mr. Kossmann has an excuse. Note that he wrote in 1872, at a moment when Germany was still tipsy from its recent military successes, and he just didn’t have enough fortitude to resist the temptation to give the proverbial kick in the behind to the defeated.

  The one-eyed nauplius only identifies rhizocephalans as crustaceans, but the next phase, the cyprid larva, occurs only in barnacles and thus specifies the ancestry of the root-heads. If the nauplius acts as a waterborne dispersal phase, the subsequent cyprid explores the substrate by crawling about on a pair of frontal appendages called antennules, securing a good spot for attachment, and then secreting cement for permanent fastening. This cement fixes most barnacles to rocks, but some species attach to whales or turtles, and one species sinks deep into whale skin to live as a near parasite. Thus, we can easily envisage the evolutionary transition from fastening to rock, to external attachment upon another animal, to internal burrowing for protection, and finally to true internal parasitism. In any case, the rhizocephalan cyprid functions like its barnacle counterpart and searches for an appropriate site of attachment upon a crustacean host. (Favored sites vary from species to species; some settle on the gills, others on the limbs.)

  We now reach the crux of the argument in considering the curious uniqueness of rhizocephalans as defined by newly evolved stages in an intricate life cycle. How does the cyprid, now attached to an external part of the host, manage to get inside the host’s body to become an adult root-head? The rhizocephalan life cycle proceeds from taxonomic generality to uniqueness. The initial nauplius identified the creature as a crustacean; the subsequent cyprid proves barnacle affinities within the Crustacea. But the next phase belongs to root-heads alone.

  The female cyprid, now attached to the host by its antennules, metamorphoses to a phase unique to the rhizocephalan life cycle, as discovered by Delage in 1884 and named the kentrogon (meaning “dart larva”). The kentrogon, smaller and simpler than the cyprid, develops a crucial and special organ—Delage’s “dart” (now generally called an “injection stylet”). The kentrogon’s dart functions as a hypodermic needle to inject the precursors of the adult stage into the body of the host!

  This delivery system for the adult’s primordium shows great diversity across the two hundred or so species of rhizocephalans. In one group, the kentrogon cements its entire ventral surface to the host. The dart then pierces the host through this ventral surface, requiring a passage through three layers—the kentrogon’s cuticle, the attaching cement, and the host’s cuticle. In another group, the kentrogon’s ventral surface does not cement, and antennules continue to function as attachments to the host. In these forms, including the genus Sacculina itself, the injecting dart goes right through one of the antennules, and thence into the body of the host! A third group skips the kentrogon stage entirely; the cyprid’s antennule penetrates the host and transfers the primordial cells of the adult parasite.

  Yves Delage, who discovered the kentrogon and its injecting device in 1884, could not hide his amazement. He wrote:

  All these facts are so remarkable, so unexpected, so strange compared with anything known either in barnacles or anywhere in the entire animal kingdom, that readers will excuse me for providing such a thorough factual documentation.

  But the next observation strikes me as even more amazing—the high-point of rhizocephalan oddity, and a near invitation to disbelief (if the data were not so firm). What constitutes the primordium of the adult parasite? What can be injected through the narrow opening of the dart’s hypodermic device?

  Delage, who discovered the mechanism, concluded that several cells, maintaining some organization as precursors to different tissues of the adult, entered the host. He could hardly come to grips with the concept of this much reduction separating larval and adult life. Imagine going through such complexity as nauplius, cyprid, and kentrogon—and then paring yourself down to just a few cells for a quick and hazardous transition to the adult stage. What a minimal bridge at such a crucial transition! “The Sacculina,” Delage wrote, “has been led to make something of a tabula rasa [blank slate] of its immediate past.” Delage then groped for analogies, and could only come up with a balloonist jettisoning all conceivable excess weight upon springing a leak.

  All can be explained by the necessity for the parasite to make itself very small in order to pass more easily through the narrow canal, whose dimensions are set by the orifice of the dart. [The transferred cells] are in the same condition as an aeronaut whose balloon has lost part of its gas, and who, needing to rise again at all cost, lightens his load by throwing out everything not absolutely indispensable to the integrity of his machine.

  Well, Monsieur Delage, the actual situation far exceeds your own source of amazement. You were quite right; many species do transfer several cells through the dart. But other species have achieved the ultimate reduction to a single cell! The dart injects just one cell into the host’s interior, and the two parts of the life cycle maintain their indispensable continuity by an absolutely minimal connection—as though, within the rhizocephalan life cycle, nature has inserted a stage analogous to the fertilized egg that establishes minimal connection between generations in ordinary sexual organisms.

&
nbsp; The evidence for transfer of a single cell has been provided in recent articles by our leading contemporary student of rhizocephalans, Jens T. Høeg of the zoological institute of the University of Copenhagen. (I read about a dozen of Høeg’s fascinating papers in preparing this essay, and thank him for so much information and stimulation.) In a 1985 article on the species Lernaeodiscus porcellanae, published in Acta Zoologica, Høeg documented the settlement of cyprids, formation of kentrogons, and injection of only a single cell, recognizable within the kentrogon, into the host. Høeg writes of the tenuous bridge within the kentrogon: “Because of its size and apparent lack of specialization the invasion cell stands out conspicuously against the surrounding epithelial, nerve and gland cells.”

  The September 14, 1995, issue of Nature, my original inspiration for this essay, reports an even more remarkable discovery: “A new motile, multicellular stage involved in host invasion by parasitic barnacles (Rhizocephala),” by Henrik Glenner and Jens T. Høeg. The authors found that the kentrogon of Loxothylacus panopaei injects a previously unknown structure into the host: a wormlike body containing several cells enclosed in an acellular sheath. This “worm” breaks up within the host’s body, and the individual cells, about twenty-five in number, then disperse separately “by alternating flexure and rotating movements.” Apparently, each cell maintains the potential to develop into an entire adult parasite, though only one usually succeeds (a few crabs develop multiple externa with independent root systems inside the body).

  This minimal transition helps to explain why the adult root-head shows no sign of barnacle affinities. If the adult parasite develops anew from a single transferred cell, then all architectural constraints of building an adult from parts of a taxonomically recognizable larva have been shed. In any case, the primordial cell or cells then migrate from the site of injection, through the circulatory spaces of the host, find a site for settlement, build an internal root system, and finally emerge through the host’s abdomen as a new structure bearing the charming name of “virgin externa.”