Deceit is not a trait that is normally considered desirable in human social behavior. Deceit, however, when applied to different aspects of biology can yield fascinating results. Over millions of years, nature has evolved several species with the remarkable ability to effectively mimic other things in their environment. Examples of such mimicry in nature can range from displays of simple camouflage, to animals such as a stick insect that can effectively mimic a stick "down to the last fine details of fake buds and leaf-scars" (Dawkins Watchmaker 82). The clever development of mimicry is not unique to animals or even genetic biology in general. Several species of plants have evolved with intricate methods of deception to propagate their genes. Many examples of mimicry can also be seen in artifacts; these artifacts may mimic other artifacts, or organisms in genetic biology. In the world of biology we can find some especially interesting comparisons between different types of mimics. Often the survival of replicators for mimicry exist for different reasons in different areas of biology, but throughout biology there are many niches that mimics may find to occupy. Mimicry can be found throughout the biological world; we can see mimicry in genetic biology in organisms, and we can see mimicry in memetic biology in artifacts.

There are two different categories of genetic mimicry that can be looked at in depth, but first the general principles for mimicry in genetic biology must be established. The precise accuracy of mimics in nature is so great that we may have a hard time imagining how genetic evolution could produce such vehicles. As often is the case, Professor Richard Dawkins provides a complete, yet easy to understand, explanation of how such close mimics may evolve in his book, The Blind Watchmaker. Critics of gradualism, or Darwinian theory in general, assert that in order for an organism such as some sort of “stick insect” to have gotten its high level of mimicry to a stick, the insect must have gone through a long series of gradual changes from, presumably, some sort of other insect that did not resemble a stick. This is the case according to gradualism, but, as the critics often persist, what are the benefits in looking only slightly, say one percent, like a stick? Would that not only fool a predator with very poor vision? If critics of gradualism are right, this would mean that all forms of mimicry would have to coevolve in some way with the eye sight of predators. Mimics and predators coevolving in all cases of evolution would be highly unlikely.

The claim that looking only slightly like a stick would only fool a predator with very poor vision, however, is not accurate. An example of why the claim is not accurate can be found when we consider distance into the equation. If we imagine some sort of insect resting on a branch where the insect resembles part of the branch to a predator by only one percent, that insect will probably not last long to a predator with keen eyesight from a moderate distance away. If, however, that predator happens to be farther away, the predator may not be able to tell the difference at glance between the insect and a branch. Whatever this critical distance is that the predator would overlook the insect is irrelevant; no matter “how poor is the resemblance of an insect to a stick, there must be . . . some degree of distance away from the eye, or some degree of distraction of the predators attention, such that even a very good eye will be fooled by the remote resemblance” (Dawkins Watchmaker 83). The insects that look more like a branch would naturally be able to deceive predators at shorter distances, so we can find reason to see a gradual process leading up to near perfect mimicry. Thus in the natural world we can see a gradual evolution of animals that are selected based on their closer resemblance to whatever they are mimicking.

In dealing with the concept of mimicry in genetic biology, we may separate mimicry into the two categories of Batesian mimicry and Mullerian mimicry. Batesian mimicry gets its name from a man named Henry Bates who discovered that "many butterflies which had similar color and pattern of the wings were in fact not closely related but were members of different families" (Mimicry online). Realizing the mistakes he had made in classification, Bates quickly recognized how such mimicry could be shaped by natural selection. Darwin accepted Bate’s results, as the results were “what [Darwin] needed to confirm his own theory” (Ferrari 124). The concept of "Batesian mimicry involves a palatable, unprotected species (the mimic) that closely resembles an unpalatable or protected species (the model)" (Salvato online). The "signal receiver" is the organism that is to be deceived by the mimic species (Miami online). In Batesian mimicry, the mimic species will take on characteristics of the model species to obtain some sort of advantage in fitness by deceiving the signal receiver.

Perhaps one of the best examples of Batesian mimicry found in nature would be that of Orchids such as Orphyrus sphegodes. Orchids that belong to the "genus Ophrys" deceive their signal receivers by way of sexual mimicry by offering "male wasps, bees, and flies-depending on the species of Ophrys-the illusion of sexual contact with a female of their own species" (Trivers 401). Male insects are "strongly attracted to these flowers," and attempt to copulate with the flowers. As the male enters this sort of "pseudo-copulation" as Trivers describes it, the male picks up pollen from the flower (403). This pollen is then transferred to other flowers as the confused insect searches for a mate. 

Although the flower may be remarkably similar in appearance to a female of the species, the most important mimicry is likely the strong scent that the flower will produce. In a study of bees, the powerful perfume of the orchid strongly resembled that of the female. In fact, of fifteen substances that are sexually stimulating to male bees, the study found that "the orchid flower turned out to produce 14 of them" (Milius 11). These scents that are put out by the flower may even be stronger than those made by the female. Because the orchid is so much larger than the insect that it mimics, "the cost of producing sexual odors should, other things being equal, be lower to the plant" (Trivers 405). In other words, since the orchid has more resources overall, more total energy may be invested in attractive scents for male insects. After the scent of the orchid gets the signal receiver’s attention, the insect will make its way to the flower that resembles the female of the insect’s species. The orchid has also evolved  “a series of rigid hairs that mimic the hairy abdomen of the female insect” that ensures the male will be well aligned on the flower to receive pollen from the orchid (Trivers 405). The male receives the pollen, though is unable to successfully mate with the flower. He flies off in search of another female of his species.

Although the evolutionary development of such an accurate mimic may be difficult to imagine, scientists have uncovered several clues to the origins of the orchid. The powerful aphrodisiac appears to have originated from “the waxy coating on part of the bloom;” this suggests that the powerful scent “may have had its evolutionary roots in waterproofing” (Milius 11). All that would be required for an evolutionary advantage would be a scent similar enough to the female insect to confuse a few males of the same species. Those orchids that developed “molecules most similar to those of the female” of the signal receiver would have the best chance of pollination, and thus reproduction (Luntz 9). Other similarities between the female companion of the male signal receiver and the orchids would be profitable for the orchids to adopt. Finally the orchid would be shaped into the plant the orchid is today, bearing a striking resemblance to the female in a species of insect.

Mimicry does not have to be so clear cut, where one species mimics another species. Sometimes, such as in the case of "the ten-spined stickleback fish," the male of the species may actually mimic the female (Trivers 406). This is an especially interesting example of deception, and I will not go into detail with the example; instead I refer the reader to Robert Triver's book Social Evolution to show that Batesian mimicry is not confined to one species deceiving another species.

The second form of mimicry in genetic biology to discuss is that of Mullerian mimicry. Mullerian mimicry was suggested by a man named Fritz Muller in the late nineteenth century. Muller studied species of butterfly and determined that "two brighly [sic] coloured distasteful butterfly species (co-models) that share a single warning-colour pattern would benefit" in sharing the burden of teaching predators to avoid the species (Kapan 338). By both species sharing similar traits, fewer butterflies of each species will pay the price of being eaten because a predator need only try one species before knowing to avoid the bright colors that the two species have in common.

An example of Mullerian mimicry occurs between the monarch and the viceroy butterflies. This is an especially interesting example because the viceroy and the monarch were originally thought to exhibit Batesian mimicry, but now several “relationships involving what were once thought to be Batesian mimicry are being reevaluated” (Salvato online). The viceroy was once “thought to mimic the Monarch,” but has been discovered in recent studies “to be as distasteful to birds as the Monarch” (Salvato online). This discovery was made by David B. Ritland and has been published in Evolution. In Ritland’s experiment, he used “red-winged blackbird predators” to confirm that the viceroy displayed Mullerian mimicry (918). Ritland’s results gave quite a shake to the biological community. The results overturned “a theory that has been part of standard textbooks for decades” (Maugh 27). Until Ritland, no one had bothered to study the relations between the viceroy and the monarch. Ritland certainly deserves due credit for his just skepticism of a long held assumption.

Monarchs rely heavily on the milkweed plant for their survival. The milkweed "serves both as a food source and breeding ground for the butterfly throughout its life cycle" (Krasean j1). The milkweed also provides the Monarch with the poison the butterfly needs to evade predators. This poison wasn’t discovered “until the 1960's,” when researchers found that “cardenolides were the chemicals in milkweed that made monarchs toxic and bitter tasting” (Interactions online). This toxic tastes horrible to predators, and similar toxins in the viceroy butterfly also make the viceroy foul tasting. As both the viceroy and the monarch developed these nasty chemicals, predators learned to avoid the colors of each species. After attacking one butterfly of the species, a predator would be unlikely to make the same mistake in attacking another butterfly of the species. The predator might, however, attack a butterfly of another species with similar toxic chemicals, for that species would exhibit a different set of traits to learn. In this situation, any butterfly that might look similar to another species of toxic butterfly would benefit; or, more properly, any gene that resulted in a phenotypic effect of a butterfly such as the viceroy that is similar to the phenotype of a monarch would spread well because a predator would be more likely to know to avoid such traits. This principle, naturally, works in reverse as well, so butterflies evolve to share similar traits because predators only have to learn the warning signs of one species of butterfly before knowing not to attack. This benefits both species, though some scientists now question exactly how much. Researchers have pointed out that both species exerting Mullerian mimicry “are unlikely to be equally unpalatable,” and have argued that “the more unpalatable species will suffer the cost of increased predation because the presence of the more palatable mimic will increase its perceived palatability”  (Dawkins & McDougall 8). Whether this is the case remains unclear, and more studies of mimicry are necessary to make confirmations. Other scientists have gone so far to say that Batesian “mimicry is unlikely in the vast majority of cases (Maugh 27). Whatever the case may be, we can still see the principle behind Mullerian mimicry and the fundamentals of mimicry in genetic biology in general. I don’t believe these concepts of mimicry are restricted to genetic biology, however, and I would like to provide some examples of how mimicry might work in the world of memetics and their vehicles.

In the world of memetics, artifacts often mimic similar traits of other objects. Though we obviously can see similarities in artifacts, the principles of mimicry in artifacts are different than in organisms. An important aspect of memetic replicators to take into account is the high amount of horizontal transmission that goes on. In genetic biology, horizontal transmission is rare; genes are passed via vertical transmission from parent to offspring. Memetic transmission is different. If a single meme spreads well in one artifact, that meme may quickly spread to other artifacts as well. If the  meme for an artifact to be constructed of plastic spreads exceptionally well in one type of artifact, that meme may be extracted and applied to another artifact by means of the human brain. For this reason there is a difficulty in separating a true artifactual mimic from two artifacts with common memes. The reader might even wonder if, in fact, any type of mimicry can be expressed in artifacts. I will return to this question later.

Another important principle of mimicry in artifacts to point out before providing examples is the reason artifacts may be mimicked, or, in other words, the reason replicators that produce an effect mimicking the effects of other replicators survive. In organisms, the reason organisms evolved to mimic other species is often to avoid predators or to gain some sort of reproductive advantage. Genes that proved likely to express certain phenotypes confusing the signal receiver would survive because confusing the signal receiver heightened the gene’s chance of propagation. A similar phenomenon occurs with memes, but the reason for confusing the signal receiver is often different. Instead of avoiding predators, artifacts that mimic other artifacts may survive because they require fewer resources than their model. We can examine a type of Batesian mimicry in artifacts as a cheaper artifact may parasitize a more expensive artifact by confusing the signal receiver; the signal receiver is almost always a human being. In the case of this mimicry, the signal receiver may be fooled into acquiring the artifact; the signal receiver is likely to have memes that move the signal receiver to invest as little of the receivers own resources into acquiring the artifact as possible, and benefiting by receiving the most resources as possible. When I use the term resources, I am referring to a value, not a pure amount. Two tons of stone probably will not be considered as valuable to the signal receiver as a pound of gold. To simplify the concept, the signal receiver, most likely a human being, will invest as little as possible while trying to benefit by receiving as much as possible from the exchange. This opens up a niche for artifacts to try and deceive the signal receiver by making their value appear higher than their value truly is, and making their cost seem lower than the artifact’s cost should be. A type of mimicry occurs when one artifact of lesser value attempts to be passed off as another artifact of greater value for more than the mimic is worth.

To illustrate this, perhaps confusing, concept, I will use an example from the clothing industry. Sometimes in the clothing industry, designer clothes are found to be counterfeit. The clothes that are produced are crafted to look like a more expensive brand of apparel with ultimately less cost of production. The popular name brand Tommy Hilfiger has often been the model for mimicry in clothing artifacts. In the year 2000, Hilfiger announced that the company was suing Goody’s Family Clothing for “selling counterfeits of its products” (Tommy C.4). The company of Tommy Hilfiger announced that Goody’s Family Clothing was profiting off of “unauthorized imitations of the company’s flag trademark” (Tommy C.4). Pairs of jeans were marketed by Goody’s stores that very closely resembled those jeans of Tommy Hilfiger. Naturally the mimic jeans were cheaper to produce, so the jeans could be sold at a cheaper price with some consumers unable to see the difference between the mimic and the model jeans. The mimic expresses a type of Batesian mimicry on the model because the imitation jeans reduce the amount of model jeans being sold. Goody’s jeans weren’t the only mimics of Hilfiger. A year earlier the company of Tommy Hilfiger “charged [Wal-Mart Stores Inc.] with selling counterfeit clothes” (Moore 2). Wal-Mart “agreed to pay $6.4 million” to settle the dispute (Moore 2). At another store, Hilfiger jeans were found to be sold “for a third of the normal price” (Bryan B.1). The jeans were, as all the other examples, mimics of the actual brand, but the memes for the mimics survived as long as they were being sold and bought.

Clothing is not the only example of this type of memetic mimicry. Beautiful and expensive jewelry that can be imitated when less expensive but similar material opens up a suitable niche for mimics. A diamond, being expensive and rare, is often substituted by the less valuable cubic zirconium. Because a cubic zirconium bears such a strong resemblance to a diamond, the stone may confuse the signal receiver into investing more resources in the cubic zirconium than the cubic zirconium is actually worth. One woman took her grandmother’s ring to a jeweler hoping to update the ring’s style. When the ring was returned to the woman, one diamond that was returned “was actually [a] cubic zirconium” (Budish 7L). The less valuable stone was returned with the effect of mimicking a more expensive diamond, so memes for the effective mimic survive. In this woman’s case, another jeweler “verified the substitutions and damages by comparing the returned ring with the prior description” (Budish 7L). The woman was fortunate, but her case is actually not uncommon, and effective mimics of more expensive jewelry are able to survive 

But all of this talk about people swindling their way to greater gains may confuse the basic principle of the nature of replicators. When counterfeiters craft products that make the counterfeiters more money, the forgers are the ones who benefit from the artifacts mimicry. Shouldn’t the survival of artifacts, like the survival of organisms, depend on the replicator as the fundamental unit? The key concept to remember here is that when memes are replicated their success will depend on how well the memes can survive in their environment. The environment that memes replicate in is, almost entirely, the human brain. When a counterfeiter enjoys reaping the benefits of making a mimic of an artifact, the memes for making artifacts will have a better chance of surviving. As a result, memes that appear to benefit the counterfeiter will also benefit memes for counterfeiting. This is not always the case, however, and sometimes memes may indeed spread that do not benefit, but actually harm, the person that houses a particular meme or set of memes. A simple example of such a case would be a meme for committing suicide.

One replicator is often complex enough, but throwing two different replicators together is enough to boggle the mind. Figuring out the dynamics between genetics and memetics may be very difficult, and I will not attempt to calculate how one replicator influences the other at this time. What I do intend to show, however, is that the two will occasionally overlap in mimicry. There is no reason for us not to expect memetic life not to mimic genetic life, and there is no reason, in due time, that genetic life should not mimic memetic life. The latter may be more difficult to show because of the much higher rate that memes are exchanged. Organisms that mimic artifacts are abound. The most recently hyped example of such an artifact came when “Sony came out with its first four-legged Aibo robot in 1999" (Guernsey 3). The Aibo robot will mimic the phenotype of a dog by being able to “respond to 75 voice commands and move around” similar to a normal canine produced by strings of DNA (Guernsey 3). The reason memes for such an artifact survives are not hard to figure out. The idea of a robotic mimic to a dog that doesn’t make a mess of the backyard or tear up the morning paper may be quite appealing to some owners. In a study “at Purdue University,” the robotic pets were shown to exhibit “enormous psychological value to the elderly” (Goddard 13). Another artifactual mimic of biological life arises for a completely different reason. Decoy hunting ducks are used by duck hunters to mimic real ducks into the area. The recent invention of “motorized, revolving- wing duck decoys, or moto-ducks” have been so effective, that the use of these “moto-ducks” has caused a “30-percent decrease in resident mallards” in the state of California (Doty 25). In the case of artifactual mimicry, both the model organisms and the mimics will be effected, so interactions between memes and genes are to be expected. An example of an organism mimicking an artifact will be a lot more difficult to find because of the fact that the rate of evolution for artifacts is so much faster than the rate of evolution for organisms. The only candidate that came to my attention was that “of the peppered moth, Biston betularia” (Miller & Levine). The peppered moth is an example of camouflage in that the species of moth changed to a darker color during the industrial revolution. The darker moths hid from predators better against the dark soot that was being produced on trees as a result of industrial activity. While this more effective camouflage is not as impressive as an orchid mimicking a female wasp, the concept still is much the same.

From orchids to jewelry, we can see a examples of mimicry throughout the biological world. Organisms can mimic other organisms; artifacts can mimic other artifacts, and both organisms and artifacts may mimic each other. There is no reason to think that the two replicators will remain separate and genes and memes should and do interact in ways that effect each others survival. Mimicry when dealing with organisms can be separated fairly cleanly into the categories of Batesian mimicry and Mullerian mimicry, but artifactual mimics often evolve to imitate other artifacts of greater value. My reason for pointing out these examples is simply to open the reader up to similarities between artifacts and organisms that may or may not be caused by common properties of replicators that produce these survival machines. Although the idea of mimicry being produced by an effect because of the properties of replicators is appealing, and I hope my own argument stands ground, we must still be careful not to get out of hand with the parallels between organisms and artifacts. Organisms and artifacts are both phenotypes of different replicators, so they both operate on the principles of any replicator, but that is where the similarities end.

Mimicry as a whole is a fascinating aspect of biology that lately has not deserved enough attention. More studies in the field of genetics in particular might open up new insights to the dynamics of mimicry in organisms. There is still much confusion over how the mimicry of works, and whether or not many mimics function through Batesian mimicry or Mullerian mimicry. In contrast, mimicry in artifacts is my own new idea, and thus is still very new. My hope is that this idea takes off naturally along with the concept of artifacts as vehicles for memes. If nothing else, the idea of mimicry in both artifacts and organisms is an interesting thought experiment, and might stir some more interest into studying the lives of mimics.

Works Cited

"Appraisals Save Day When Jeweler Swaps Stones." The Plain Dealer 20 Feb. 2000: : L7. Bryan, Bill. "Raid on Discount Shop Nets Counterfeit Clothing, Police Say Truckloads of 'Name-Brand' Products, Labels are Confiscated." St. Louis Post - Dispatch 20 Jan. 2000: : B1.

Dawkins, Marian S., and Angus McDougall. "Predator Discrimination Error and the Benefits of Mullerian Mimicry." Animal Behaviour May 1998: 1281-1288.

Dawkins, Richard. The Blind Watchmaker. 2nd ed. New York, NY: W.W. Norton & Company Inc., 1996.

Doty, Jonah. "Moto-Duck Mayhem?" Sports Afield Dec. 2001: 25.

Ferrari, Marco. Colors for Survival. 2nd ed. New York: Barnes & Noble Books, 1997.

Goddard, Jacqui. "Just Because Dog's a Robot Doesn't Mean You Can't Love it." South China Morning Post 8 Jan. 2003: : 13.

Guernsey, Lisa. "A Programmable Dog That Will Bite Back." The New York Times 8 Nov. 2001: : 3.

Interactions With Milkweed. University of Minnesota. 4 Feb. 2003  <http://www.monarchlab.umn.edu/research/Mwd/mwd.html>.

Kapan, Durrell D. "Three-Butterfly System Provides a Field Test of Mullerian Mimicry." Nature 18 Jan. 2001: 338-340.

Krasean, Bill. "Monarchs Take Entomologist on Adventure of a Lifetime." The Grand Rapids Press 16 July 1995: : J1.

Levine, and Ken Miller. The Peppered Moth - An Update. Aug. 1999. Brown University. 17 Jan. 2003  <http://biocrs.biomed.brown.edu/Ele...ers/Ch%2014/Moths/Moth-Update.html>.

Luntz, Stephen. "The Way to a Bee's Heart is Through its Stomach." Australasian Science Nov. 2000: 9.

Maugh II, Thomas H. "Research Exposes Myth on Butterfly Coloring Evolution: Century-Old Belief is Discredited. It Held that Viceroy Butterfly Avoids Being Eaten by Mimicking Coloration of Two Kinds that Birds Dislike." The Los Angeles Times 11 Apr. 1991: : 27.

Milius, Susan. "Recipe Found for Orchid Aphrodisiac." Science News 3 July 1999: 11.

Mimicry. University of Miami. 15 Jan. 2003 <http://fig.cox.miami.edu/Faculty/Tom/bil101sp99/23_mimicry.html>.

Moore, Booth. "Page 2 / News, Trends, Gossip and Stuff to do; Notes and News; Wal-Mart Settles Counterfeiting Suit with Hilfiger for $6.4 Million." The Los Angeles Times 16 July 1999: : 2.

Ritland, David B. "Revising a Classic Butterfly Mimicry Scenario: Demonstration of Mullerian Mimicry Between Florida Viceroys (Limenitis Archippus Floridensis) and Queens (Danaus Gilippus Berenice)." Evolution June 1991: 918.

Salvato, Mark. University of Florida Book of Insect Records Chapter 28 Most Spectacular Batesian Mimicry. 15 Apr. 1997. University of Florida. 15 Jan. 2003 <http://ufbir.ifas.ufl.edu/chap28.htm>.

"Tommy Hilfiger Sues Goody's Over Trademark Use." The New York Times 1 Aug. 2000: : C4.

Trivers, Robert. Social Evolution. Menlo Park, California: Benjamin/Cummings Company, Inc., 1985.

Works Consulted

Ayasse, Manfred, et al. "Evolution of Reproductive Strategies in the Sexually Deceptive Orchid Ophrys sphegodes: How Does Flower-Specific Variation of Odor Signals Influence Reproductive Success?" Evolution Dec. 2000: 1995-2006.

Blanchard, Tamsin. "Life: Investingation: Faking it: From the Street Trader to the Fashion Stylist, 10 Per Cent of the Clothing We Wear is Fake. But Where do these Compies Come From? And Who is Making Them? Tamsin Blanchard Reports From Leicester, the Counterfeit Capital of Europe." The Observer 19 Nov. 2000: : 19.

Corey, Gail, and William Sumner. Studying Living Organisms. Woodrow Wilson Biology Institute. 15 Jan. 2003  <http://www.accessexcellence.org/AE/AEPC/WWC/1995/mimicry.html>.

Grant, B S., and L L. Wiseman. "Recent History of Melanism in American Peppered Moths." The Journal of Heredity 93 (2002): 86-90.

Johnstone, Rufus A. "The Evolution of Inaccurate Mimics." Nature 1 Aug. 2002: 524-526.

Lorenzana, Janice C., and Spencer G. Sealy. "Did Blue Eggs of Black-Billed (Coccyzus erythropthalmus) and  Yellow-Billed (C. americanus) Cuckoos Evolve to Counter Host Discrimination?" The Auk July 2002: 851-854.

Mazer, Roslyn A. "From T-Shirts to Terrorism; That Fake Nike Swoosh May Be Helping to Fund Bin Laden's Network." The Washington Post 30 Sept. 2001: : B2.

Milkweeds. Weed Science Society of America. 4 Feb. 2003 <http://www.wssa.net/subpages/weed/larrymitich/milkweeds.html>.

Monarch Butterfly (Danaus Plexippus). 20 Oct. 2002. University of Wisconsin Green Bay. 15 Jan. 2003  <http://www.uwgb.edu/biodiversity/n...e_week/pow2002/pow20020920main.htm>.

Mumme, Ronald L., Noah T. Owen-Ashley, and Stephan J. Schoech. "Context-Specific Response of Florida Scrub-Jay Pairs to Northern Mockingbird Vocal Mimicry." The Condor Nov. 2002: 858.

Nei, Guo. "Campaign Targets Fake Products." China Daily 3 Jan. 2003: : 1.

"New Decoys Flap in the Breeze." Madison Capital Times 9 Oct. 2002: : D4.

Ritland, David B. "Mimicry-Related Predation on Two Viceroy Butterfly (Limenitis archippus) Phenotypes." The American Midland Naturalist July 1998: 1-20.

Shaul, Dvora B. "Orchid: This is One Sexy Flower." Jerusalem Post 25 Mar. 1994: : B11. "Suspect Sought in Mall Jewelry Scam." Dayton Daily News 13 Nov. 1996: : Z4.4.