The vestigiality of the human vermiform appendix

"The vermiform appendage—in which some recent medical writers have vainly endeavoured to find a utility—is the shrunken remainder of a large and normal intestine of a remote ancestor. This interpretation of it would stand even if it were found to have a certain use in the human body. Vestigial organs are sometimes pressed into a secondary use when their original function has been lost." Joseph McCabe The Story of Evolution (1912), p. 264

"Its major importance would appear to be financial support of the surgical profession." Alfred Sherwood Romer and Thomas S. Parsons The Vertebrate Body (1986), p. 389.

Outline

• Introduction
• The vermiform appendix: background information
• The caecum
• The caecal apex and the appendix are homologous
• Intermediates between the caecum and the appendix
• Possible functions of the appendix
• The appendix is a suboptimal design
• Evolutionary misconceptions in the medical literature
• How to disprove that the appendix is a vestige
• Conclusion
• Acknowledgements
• References

Introduction

Many biological structures can be considered vestiges given our current evolutionary knowledge of comparative anatomy and phylogenetics. In evolutionary discussions the human vermiform appendix is one of the most commonly cited vestigial structures, and one of the most disputed. Evolutionary vestiges are, technically, any diminished structure that previously had a greater physiological significance in an ancestor than at present. Independently of evolutionary theory, a vestige can also be defined typologically as a reduced and rudimentary structure compared to the same homologous structure in other organisms, as one that lacks the complex functions usually found for that structure in other organisms (see, e.g. Geoffroy 1798).

Classic examples of vestiges are the wings of the ostrich and the eyes of blind cavefish. These vestigial structures may have functions of some sort. Nevertheless, what matters is that rudimentary ostrich wings are useless as normal flying wings, and that rudimentary cavefish eyes are useless as normal sighted eyes. Vestiges can be functional, and speculative arguments against vestiges based upon their possible functions completely miss the point.

For more discussion of the vestigial concept, extensive modern and historical references concerning its definition (especially the allowance for functionality), see the Citing Scadding (1981) and Misunderstanding Vestigiality and 29+ Evidences for Macroevolution: Anatomical vestiges FAQs.

The following discussion makes four main points:

1. The human appendix may have bona fide functions, but this is currently controversial, undemonstrated in humans, and irrelevant as to whether the appendix is a true vestige or not.
2. The appendix is a prime example of dysteleology (i.e. suboptimal structural design), a prediction of genetically gradual evolution.
3. The appendix is a rudimentary tip of the caecum and is useless as a normal, cellulose-digesting caecum.
4. Thus, the appendix is vestigial by both the evolutionary and non-evolutionary, typological definitions of vestigiality.

The vermiform appendix: background info

Figure 1: The human vermiform appendix (image reproduced with modifications from Gray 1918)

In humans, the vermiform appendix is a small, finger-sized structure, found at the end of our small caecum and located near the beginning of the large intestine (Fawcett and Raviola 1994, p. 636; Oxford Companion to the Body 2001, pp. 42-43; Williams and Myers 1994). The adjective "vermiform" literally means "worm-like" and reflects the narrow, elongated shape of this intestinal appendage. The appendix is typically between two and eight inches long, but its length can vary from less than an inch (when present) to over a foot. The appendix is longest in childhood and gradually shrinks throughout adult life. The wall of the appendix is composed of all layers typical of the intestine, but it is thickened and contains a concentration of lymphoid tissue. Similar to the tonsils, the lymphatic tissue in the appendix is typically in a constant state of chronic inflammation, and it is generally difficult to tell the difference between pathological disease and the "normal" condition (Fawcett and Raviola 1994, p. 636). The internal diameter of the appendix, when open, has been compared to the size of a matchstick. The small opening to the appendix eventually closes in most people by middle age. A vermiform appendix is not unique to humans. It is found in all the hominoid apes, including humans, chimpanzees, gorillas, orangutans, and gibbons, and it exists to varying degrees in several species of New World and Old World monkeys (Fisher 2000; Hill 1974; Scott 1980).

The caecum: a specialized herbivorous organ

Our appendix is a developmental derivative and evolutionary vestige of the end of the much larger herbivorous caecum found in our primate ancestors (Condon and Telford 1991; Williams and Myers 1994, p. 9). The word "caecum" actually means "blind" in Latin, reflecting the fact that the bottom of the caecum is a blind pouch (a dead-end or cul-de-sac).

In most vertebrates, the caecum is a large, complex gastrointestinal organ, enriched in mucosal lymphatic tissue (Berry 1900), and specialized for digestion of plants (see Figure 2; Kardong 2002, pp. 510-515). The caecum varies in size among species, but in general the size of the caecum is proportional to the amount of plant matter in a given organism's diet. It is largest in obligate herbivores, animals whose diets consist entirely of plant matter. In many herbivorous mammals the caecum is as large as the rest of the intestines, and it may even be coiled and longer than the length of the entire organism (as in the koala). In herbivorous mammals, the caecum is essential for digestion of cellulose, a common plant molecule. The caecum houses specialized, symbiotic bacteria that secrete cellulase, an enzyme that digests cellulose. Otherwise cellulose is impossible for mammals to digest.

The structure of the caecum is specialized to increase the efficiency of cellulose fermentation. As a "side branch" from the gut it is able to house a large, dense, and permanent colony of specialized bacteria. Being a dead-end sac at the beginning of the large intestine, it allows more time for digesting food to reside in the gut and ferment more completely, before passing through the large intestine where the resulting nutrients are absorbed. However, even though humans are herbivorous, the small human caecum does not house significant quantities of cellulase-excreting bacteria, and we cannot digest more than but a few grams of cellulose per day (Slavin, Brower, and Marlett 1980).

Figure 2: Gastrointestinal tracts of various mammals. For each species, the stomach is shown at top, the small intestine at left, the caecum and associated appendix (if present) in magenta, and the large intestine at bottom right. Scale differs between species. Reproduced with modifications from Kardong 2002, p. 511. Copyright © 2002 McGraw-Hill.

The human appendix is homologous to the end of the mammalian caecum

In vertebrate comparative anatomy, it has long been known that the human appendix and the end of the mammalian caecum are structurally homologous (Berry 1900; Fisher 2000; Hill 1974; Hyman 1979, p. 412; Kardong 2002, pp. 513-515; Kluge 1977, p. 1977; Neal and Rand 1936, p. 315; Romer and Parsons 1986, p. 389; Royster 1927, p. 27; Smith 1960, p. 305; Weichert 1967, p. 189; Wiedersheim 1886, p. 236; Wolff 1991, p. 384). Of course, the end of the caecum and the appendix can be homologous and have different functions. Being the termination of the caecum, the human vermiform appendix is also a "blind pouch," and another name for the appendix is in fact the "true caecal apex" (Berry 1900). Within the gastrointestinal tract of many vertebrates, mammals, and primates in particular, the termination of the caecum and the vermiform appendix share the same relative position (Figure 2), both have a similar structure and form, both are blind sacs enriched with lymphatic tissue (Berry 1900), both have a common developmental origin (Condon and Telford 1991; Williams and Myers 1994, p. 9), and, as discussed below, in the primates both are connected by an extensive series of intermediates. These observations firmly establish these structures as homologous by standard systematic criteria (Kitching et al. 1998, pp. 26-27; Remane 1952; Schuh 2000, pp. 63-64; Rieppel 1988, p. 202), a conclusion confirmed by cladistic systematic analysis (Goodman et al. 1998; Shoshani 1996).

A few other mammals appear to have a structure similar to the hominoid vermiform appendix, including the wombat, South American opossum (both marsupials), some rodents, and the rabbit. However, extensive comparative analysis has shown that the caecal appendixes of humans and these other mammals were derived from the caecum independently; these structures are not homologous as appendixes (Shoshani and McKenna 1998). The relationship between these other caecal structures and the hominoid vermiform appendix is similar to the homology of bat and bird wings. The wings of bats and birds are homologous as modified forelimbs, yet they are not homologous as wings. Likewise, the appendixes of rabbits and humans are homologous as modified caeca, yet they are not homologous as appendixes. If the rabbit and human appendix have similar functions (which has never been experimentally demonstrated), they are the result of independent convergence of function and form (Shoshani and McKenna 1998). The many significant morphological, histological, and cellular differences between the rabbit and human appendixes (discussed below) all are consistent with a superficial convergent relationship.

Structural intermediates of the appendix and caecum in primate phylogeny

The primate family tree provides a rather complete set of intermediates between the states "large caecum/appendix absent" to "small caecum/appendix present". Many primates have both a caecum and an appendix, or a structure intermediate between the two. The anatomical definition of a vermiform appendix is a narrowed, thickened, lymphoid-rich caecal apex (Fisher 2000 and references therein). As already mentioned, the hominoid apes all bear a vermiform appendix, but many non-anthropoid primates also have structures that fit the above definition to varying degrees. In fact, recent reevaluation of the anatomy of the primate caecum and appendix has highlighted the difficulties in determining exactly where the caecum ends and the appendix begins in different species (Fisher 2000). This complication arises from the continuous, variable, and overlapping nature of caecal and appendicular tissues, both histologically and anatomically. For example, in most primates the caecal apex is rich in lymphoid tissue and is thickened, but whether it is narrowed into a conical or "worm-like" structure is variable (Fisher 2000).

From systematic analysis of comparative anatomy, it is known that in primates a large caecum with a small or absent appendix is the ancestral, primitive state (Goodman et al. 1998; Shoshani 1996). In general, the length of the caecum, relative to that of the colon, decreases as one traverses the primate phylogenetic tree from monkeys to humans. Concurrently, the size of the appendix increases. The appendix is mostly absent in prosimians and New World monkeys, yet they have a large caecum. In Old World monkeys the appendix is more recognizable, and it is well-developed in the anthropoid apes, which lack the large cellulose-fermenting caecum found in their ancestors and other primates (Fisher 2000; Goodman et al. 1998; Hill 1974; Shoshani 1996; Scott 1980).

Possible function of the appendix

"The appendix n, of the colon n m, is a part of the caecum and is capable of contracting and dilating so that excessive wind does not rupture the caecum." Leonardo da Vinci FB 14v (1504-1506). from O'Mally and Saunders 1952, Leonardo da Vinci on the human body, p. 185.

Earliest known drawing of the appendix, by Leonardo da Vinci.

Throughout medical history many possible functions for the appendix have been offered, examined, and refuted, including exocrine, endocrine, and neuromuscular functions (Williams and Myers 1994, pp. 28-29). Today, a growing consensus of medical specialists holds that the most likely candidate for the function of the human appendix is as a part of the gastrointestinal immune system. Several reasonable arguments exist for suspecting that the appendix may have a function in immunity. Like the rest of the caecum in humans and other primates, the appendix is highly vascular, is lymphoid-rich, and produces immune system cells normally involved with the gut-associated lymphoid tissue (GALT) (Fisher 2000; Nagler-Anderson 2001; Neiburger et al. 1976; Somekh et al. 2000; Spencer et al. 1985). Animal models, such as the rabbit and mouse, indicate that the appendix is involved in mammalian mucosal immune function, particularly the B and T lymphocyte immune response (Craig and Cebra 1975). Animal studies provide limited evidence that the appendix may function in proper development of the immune system in young juveniles (Dasso and Howell 1997; Dasso et al. 2000; Pospisil and Mage 1998).

However, contrary to what one is apt to read in anti-evolutionary literature, there is currently no evidence demonstrating that the appendix, as a separate organ, has a specific immune function in humans (Judge and Lichtenstein 2001; Dasso et al. 2000; Williams and Myers 1994, pp. 5, 26-29). To date, all experimental studies of the function of an appendix (other than routine human appendectomies) have been exclusively in rabbits and, to a lesser extent, rodents. Currently it is unclear whether the lymphoid tissue in the human appendix performs any specialized function apart from the much larger amount of lymphatic tissue already distributed throughout the gut. Most importantly with regard to vestigiality, there is no evidence from any mammal suggesting that the hominoid vermiform appendix performs functions above and beyond those of the lymphoid-rich caeca of other primates and mammals that lack distinct appendixes.

As mentioned above, important differences exist in nearly all respects between the human and rabbit appendixes (Dasso et al. 2000; Williams and Myers 1994, p. 57). The rabbit appendix, for instance, is very difficult to identify as separate from the rest of its voluminous caecum (see Figure 2). Unlike the human appendix, the rabbit's appendix is extremely large, relative to the colon, and is the seat of extensive cellulose degradation due to a specialized microflora. The large rabbit appendix houses half of its GALT lymphoid tissue, whereas the contribution of the human appendix to GALT is significantly less (Dasso et al. 2000). In humans the vast majority of GALT tissue is found in hundreds of Peyer's patches coating the small intestine and in nearly 10,000 similar patches found in the large intestine. Additionally, there are important differences in lymphoid follicular structure, in T-cell distribution, and in immunoglobulin density (Dasso et al. 2000). Furthermore, from systematic analysis we know that the rabbit, rodent, and human appendixes are convergent as outgrowths and constrictions of the caecum (Shoshani and McKenna 1998). It is thus very questionable to conclude from these animal studies that the human appendix has the same function as the other non-primate appendixes.

Of course, over a century of medical evidence has firmly shown that the removal of the human appendix after infancy has no obvious ill effects (apart from surgical complications, Williams and Myers 1994). Earlier reports of an association between appendectomy and certain types of cancer were artifactual (Andersen and Isager 1978; Gledovic and Radovanovic 1991; Mellemkjaer et al. 1998). In fact, congenital absence of the appendix also appears to have no discernable effect. From investigative laparoscopies for suspected appendicitis, many people have been found who completely lack an appendix from birth, apparently without any physiological detriment (Anyanwu 1994; Chevre et al. 2000; Collins 1955; Hei 2003; Host et al. 1972; Iuchtman 1993; Kalyshev et al. 1995; Manoil 1957; Pester 1965; Piquet et al. 1986; Ponomarenko and Novikova 1978; Rolff et al. 1992; Saave 1955; Shperber 1983; Tilson and Touloukian 1972; Williams and Myers 1994, p. 22).

In sum, an enormous amount of medical research has centered on the human appendix, but to date the specific function of the appendix, if any, is still unclear and controversial in human physiology (Williams and Myers 1994, pp. 5, 26-29).

The appendix is suboptimally designed

The human appendix is notorious for the life-threatening complications it can cause. Deadly infection of the appendix at a young age is common, and the lifetime risk of acute appendicitis is 7% (Addiss et al. 1990; Hardin 1999; Korner et al. 1997; Pieper and Kager 1982). The most common age for acute appendicitis is in prepubescent children, between 8 and 13 years of age. Before modern 20th-century surgical techniques were available, a case of acute appendicitis was usually fatal. Even today, appendicitis fatalities are significant (Blomqvist et al. 2001; Luckmann 1989).

The small entrance to this dead-end pocket makes the appendix difficult to clean out and prone to physical blockage, which ultimately is the cause of appendicitis (Liu and McFadden 1997). This peculiar structural layout is quite beneficial for a larger cellulose-fermenting caecum, but it is unclear why gut lymphoid tissue would need to be housed in a remote, dead-end tube with negligible surface area. In fact, 60% of appendicitis cases are due to lymphoid hyperplasia leading to occlusion of the interior of the appendix, indicating that the appendix is unusually prone to abnormal proliferation of its lymphoid tissue (Liu and McFadden 1997). Such an occurrence would be much less problematic if the interior of the appendix were not so small, confined, and inaccessible from the rest of the gut. In many other primates and mammals, the GALT lymphoid tissue appears to function without difficulty in a much more open, bulbous caecum with ample surface area.

Furthermore, there is mounting evidence that removing the appendix helps prevent ulcerative colitis, a nasty inflammatory disease of the colon (Andersson et al. 2001; Buergel et al. 2002; Judge and Lichtenstein 2001; Koutroubakis and Vlachonikolis 2000; Koutroubakis et al. 2002; Naganuma 2001; Rutgeerts 1994). This evidence suggests that the appendix is actually maladaptive, and that the lymphoid tissue contained in the appendix is prone to chronic pathological inflammatory states. If the appendix does have an important function that we have yet to find, it is a leading candidate for the worst designed organ in the human body. How nice if the appendix would just degenerate away after it is no longer needed, so it could never get infected and kill us needlessly. Any biological structure that supposedly ensures our livelihood by its functions, yet paradoxically and unnecessarily kills a large fraction of its bearers prematurely, is poorly designed indeed.

Why do some medical sources question the vestigiality of the appendix?

The reasons for this are multiple, but they largely stem from the simple fact that most physicians are not trained in evolutionary biology. The erroneous "completely nonfunctional" definition of a vestige is primarily found in medical papers, textbooks, and dictionaries (e.g. Williams and Myers 1994, p. ix). Using this incorrect and nonevolutionary definition, it is logical to conclude that a structure is not vestigial if its function is discovered. For instance, based upon this incorrect definition, Williams and Myers 1994 incorrectly argue that an evolutionary vestige cannot be both a complex and a "regressive" structure (p. 27). Similarly, a modern version of Gray's Anatomy confusingly implies that the appendix cannot be both vestigial and specialized (Williams and Warwick 1980). However, vestiges are very often complex or specialized structures, in fact overly complex for their functions, and prime examples are the wing of the ostrich and the eyes of blind cavefish. A vestige can be a complex structure, in an absolute sense, while simultaneously being rudimentary or degenerate relative to the same homologous structure in other organisms.

Perhaps most important is the fact that a vestige can be identified only via comparative analysis. Physicians are experts on human anatomy and physiology, but rarely do discussions in medical publications consider phylogenetic and comparative issues. Medical articles that attempt to consider phylogenetics often provide a gross misconception of evolutionary fundamentals. For instance, the most thorough and in depth source on the physiology of the human appendix, Williams and Myers 1994, refers to how the appendix changes as "the primate scale is ascended" and to the "evolutionary scale" with humans at its end (pp. 26-27). These are long-refuted orthogenetic concepts which have been contradicted by basic evolutionary theory since Darwin. Scott 1980 similarly argues against vestigiality based upon orthogenetic concepts and a belief in evolutionary "progress." Fisher 2000 (p. 229) and Scott 1980 both incorrectly imply that a vestige cannot be a derived character, a curious assessment since a vestige must be a phylogenetically derived character by definition.

How would we know if the appendix were not vestigial?

Whether the appendix has a function of some sort or not has no direct bearing on whether it is a bona fide vestige. However, at least three possible observations would help negate the conclusion that the human appendix is vestigial, using either the evolutionary or the typological definitions of vestigiality:

• (1) if the human appendix were actually as large and developed as, say, the caecum of a prosimian or New World monkey;
• (2) if the human appendix contributed significantly to cellulose fermentation and contained a large amount of cellulose-digesting bacteria;
• (3) if we could demonstrate via phylogenetic or systematic methods that the apex of the cellulose-fermenting caecum in other primates and the vermiform appendix were not structurally homologous as side branches from the intestine.

An additional possible observation would contradict the conclusion of vestigiality by the evolutionary definition:

• (4) if phylogenetic methods indicated that no predicted ancestors of humans ever had a large, cellulose-fermenting caecum (i.e., that a large, cellulose-digesting caecum is actually a derived primate character, not a primitive one).

All four of these potential observations are demonstrably false. Additionally, each is based upon positive scientific evidence: (1) we can measure and quantitate the size of the appendix; (2) we can measure and quantitate the amount of cellulose digestion occurring in the appendix; (3) we can observe and compare the relative positions, underlying structures, forms, and development of the organs in the gastrointestinal tracts of various organisms; and (4) we can determine primitive and derived characters by independent phylogenetic analysis. Therefore, the conclusion of vestigiality is susceptible and open to scientific testing against empirical observation. As such the concept of vestigiality is not an "argument from ignorance." It is clearly scientific in nature, based completely upon positive evidence.

Conclusion: The vermiform appendix is vestigial

Currently, arguments against the vestigiality of the human vermiform appendix have been based upon misunderstandings of what constitutes a vestige and of how vestiges are identified.

From an evolutionary perspective, the human appendix is a derivative of the end of the phylogenetically primitive herbivorous caecum found in our primate ancestors (Goodman et al. 1998; Shoshani 1996). The human appendix has lost a major and previously essential function, namely cellulose digestion. Though during primate evolution it has decreased in size to a mere rudiment, the appendix retains a structure that was originally specifically adapted for housing bacteria and extending the time course of digestion. For these reasons the human vermiform appendix is vestigial, regardless of whether or not the human appendix functions in the development of the immune system.

From a nonevolutionary, typological perspective, the human appendix is homologous to the end of the physiologically important, large, cellulose-fermenting caeca of other mammals. Even though humans eat cellulose, the contribution to cellulose digestion by both the human caecum and its associated appendix is negligible. Regardless of whether one accepts evolutionary theory or not, the human appendix is a rudiment of the caecum that is useless as a normal mammalian, cellulose-digesting caecum. Thus, by all accounts the vermiform appendix remains a valid and classic example of a human vestige.

Acknowledgements

Thanks to Colin Groves for helpful discussion on the comparative anatomy and homology of the primate appendix and caecal apex, and to John Harshman for insightful comments and clarification of levels of homology.

References