Mammal-Like Reptiles
As previously stated, a succession of transitional fossils exists that link reptiles (Class Reptilia) and mammals (Class Mammalia). These particular reptiles are classifie as Subclass Synapsida. Presently, this is the best example of th e transformation of one major higher taxon into another. The morphologic changes that took place are well documented by fossils, beginning with animals essentially 100% reptilian and resulting in animals essentially 100% mammalian. See the chart below...
Modern reptiles and mammals are very distinctive, easily diagnosable, and do not intergrade. Reptiles are covered by scales, mammals by hair; reptiles are cold-blooded, mammals warm-blooded; reptiles do not suckle their young, mammals have mammary glands; reptiles have sprawling posture, mammals have upright posture. Most of these features are soft part anatomy or physiology that very rarely fossilize (although dinosaur skin impressions are known from Cretaceous sediments, and imprints of mammal hair are known from Eocene bats from Germany; Franzen, 1990). In the fossil record, we must look to skeletal features.
There are many skeletal features which allow us to distinguish the reptiles from the mammals (Carroll, 1988; Table 1, rows A, M). The single most important defining characteristic is the nature of the articulation of the lower jaw to the skull (Simpson, 1959). In reptiles, multiple bones comprise the lower jaw. A small bone at the posterior end of the lower jaw, the articular, articulates with the quadrate bone of the skull (Simpson, 1959; Carroll, 198
. In mammals, one large bone, the dentary, comprises the lower jaw. It articulates with the squamosal bone of the skull (Simpson, 1959; Carroll, 198
.
From comparative anatomy studies, it is certain that most of the bones of the reptiles and mammals are homologous (Crompton & Parker, 1978; Carroll, 198
. Of greatest importance, the middle ear bones of mammals (stapes, incus, malleus, and tympanic) are homologous with several of the skull and jaw bones of reptiles (stapes, quadrate, articular, and angular, respectively; Romer, 1956, p. 33-38, 1970a; Allin, 1975, 1986; Allin & Hopson, 1992; Crompton & Parker, 1978; Hopso n, 1987, 1994; Carroll, 198
. One group of reptiles, the synapsids (Subclass Synapsida), share with the mammals an additional homologous structure: the lateral temporal fenestra, which is an opening in the skull behind the eye socket at the triple junction between the squamosal, jugal , and post orbital bones (Broom, 1932; Frazetta, 1968; Kemp, 1982; Carroll, 198
. A band of bone composed of the jugal and the squamosal is adjacent to the lateral temporal fenestra (Broom, 1932; Kemp, 1982; Carroll, 198
. This is the cheek arch so characteristic of mammal skulls (Broom, 1932; Kemp, 1982; Carroll, 198
. Therefore, synapsids are commonly named the “mammal-like reptiles.”
The presence of diagnosable morphologic differences between reptiles (including the oldest reptiles and the oldest synapsids) and mammals distinguishes them as distinct taxa. This allows us to test evolution by looking for transitional forms between the two. Because many of the bones are homologous, we should find evidence illustrating how these bones were modified over time to become the new bones. Furthermore, these morphologic changes should happen in parallel and in geochronologic succession.
Synapsid reptiles inhabited Pangea from the Middle Pennsylvanian through the Early Jurassic (Kemp, 1982, 1985; Sloan, 1983; Carroll, 1988; Hopson, 1969, 1987, 1994; Hopson & Crompton, 1969; Hotton, et al., 1986; Crompton & Jenkins, 1973; Sidor & Hopson, 1998; Romer & Price, 1940; Broom, 1932; Boonstra, 1963, 1969, 1971; Tchudinov, 1983; Olson, 1944; Tatarinov, 1974; Vyushkov, 1955; Efremov, 1954). From the Early Permian through the Early Triassic, they were the largest and most abundant land animals (Sloan, 1983; Colbert, 1965). Though much less well known to the general public than dinosaurs, one of the “cereal box dinosaurs,” Dimetrodon (the sail-backed reptile), is a synapsid, not a dinosaur (Romer & Price, 1940; Carroll, 198
. The oldest mammals are Late Triassic (Kemp, 1982; Carroll, 198
. Below is a discussion of the geochronologic succession linking synapsids and mammals. The oldest reptiles (named protorothyrids; Carroll, 1964, 1988, p. 192-199) are from the lower Middle Pennsylvanian, and the oldest synapsids (Reisz, 1972) are from the upper Middle Pennsylvanian, both of Nova Scotia. Upper Pennsylvanian and Lower Permian forms are known primarily from the midcontinent and Permian Basin region of the United States (Romer & Price, 1940; Currie, 1977, 1979; Kemp, 1982; Sloan, 1983). The basal Upper Permian forms are known from Russia (Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972; Ivakhnenko et al., 1997). Most of the Upper Permian and Lower Triassic succession is known from southern Africa, especially the Great Karoo of South Africa (Broom, 1932; Boonstra, 1963, 1969, 1971; Hopson & Kitching, 1972; Kemp, 1982; Sloan, 1983). The Middle Triassic forms are from South America (Romer, 1969a, 1969b, 1970b, 1973; Romer & Lewis, 1973; Bonaparte & Barbarena, 1975), and the Upper Triassic and Lower Jurassic mammals are known from Eurasia (Kermack, Mussett, & Rigney, 1973, 1981; Kemp, 1982). Subsequent Mesozoic mammals are known from all over the world (Simpson, 1928; Lillegraven et al., 1979).
When placed in proper geochronologic succession, the synapsids naturally form a succession of taxa (genera and families) that progressively become more mammal-like and less reptile-like (Kemp, 1982, 1985; Sloan, 1983; Sidor & Hopson, 1998; Hopson, 1987, 1994). Morphologic changes, summarized in Table 1 and Figure 1, affect the entire skeletal anatomy of these animals, but are most clearly displayed in their skulls.
The lateral temporal fenestra increased in size from a tiny opening smaller than the eye socket to a giant opening occupying nearly half the length of the skull. Ultimately, it merged with the eye socket, thus producing the full development of the cheek arch so characteristic of mammals (Broom, 1932; Frazetta, 1968; Kemp, 1982; Sloan, 1983; Hopson, 1987, 1994; Carroll, 198
.
Successively, the relative proportion of the lower jaw comprised of the dentary bone (teeth-bearing bone) gradually increased until the entire lower jaw consisted of the dentary (Kemp, 1982; Sloan, 1983; Carroll, 1988; Hopson, 1987, 1994). In Pennsylvanian and Lower and basal Upper Permian synapsids, the postero-dorsal edge of the lower jaw rose broadly but only slightly above the level of the tooth row (Romer & Price, 1940; Currie, 1977, 1979; Ivakhnenko et al., 1997; Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972; Hopson, 1987, 1994). In succeeding forms, the posterior part of the dentary expanded dorsally and posteriorly as a blade-like process, and progressively became larger (Broom, 1932; Boonstra, 1963, 1969, 1971; Sigogneau, 1970; Brink, 1963; Kemp, 1979; Hopson, 1987, 1994), forming the coronoid process (Parrington, 1946; Fourie, 1974; Romer, 1969b, 1970b, 1973; Hopson, 1987, 1994) to which the mammalian-type jaw musculature is attached (Barghusen, 1968; Bramble, 1978; Crompton, 1972; Crompton & Parker, 1978; Kemp, 1982; Sloan, 1983; Carroll, 198
. Concomitantly, the post-dentary bones progressively reduced in size (Allin, 1975; Crompton, 1972; Crompton & Parker, 1978; Kemp, 1982; Sloan, 1983; Carroll, 1988; Hopson, 1987, 1994).
Beginning with the Upper Pennsylvanian sphenacodonts, a notch developed in the angular bone that offsets a projection, the reflected lamina (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Romer & Price, 1940; Currie, 1977, 1979; Kemp, 1982; Sloan, 1983; Carroll, 198
. The reflected lamina first became a large blade-like flange (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Ivakhnenko et al., 1997; Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972; Broom, 1932; Sigogneau, 1970; Boonstra, 1963, 1969, 1971), and then was progressively reduced to a delicate horseshoe-shaped bone (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Brink, 1963; Parrington, 1946; Fourie, 1974; Romer, 1969b, 1970b, 1973; Kermack, Mussett, & Rigney, 1973, 1981; Kemp, 1979, 1982; Sloan, 1983; Carroll, 198
.
Simultaneously, the quadrate progressively decreased in size (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Kemp, 1982; Sloan, 1983; Carroll, 198
. The articular did not decrease in size much, being small initially, but developed a downward-pointing prong (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Kemp, 1982; Sloan, 1983; Carroll, 198
. In the synapsids, the lower jaw was hinged to the skull by the articular and quadrate bones (Crompton, 1972; Crompton & Parker, 1978; Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994). Thus they are classified as reptiles (Simpson, 1959; Kemp, 1982; Sloan, 1983; Carroll, 198
. As the quadrate and articular became smaller, they were relieved of their solid suture to the dentary and skull (Crompton, 1972; Allin, 1975, 1986; Allin & Hopson, 1992; Hopson, 1987, 1994; Crompton & Parker, 1978; Kemp, 1982; Sloan, 1983; Carroll, 198
. A projection of the dentary extended posteriorly and made contact with the squamosal. Morganucodon possessed the mammalian dentary-squamosal jaw joint adjacent to the reptilian articular-quadrate jaw joint (Kermack, Mussett, & Rigney, 1973, 1981; Carroll, 198
. It is classified as the first mammal, but it is a perfect intermediate. Now that a new jaw joint was established, the quadrate and articular were subsequently relieved of that function (Crompton, 1972; Allin, 1975, 1986; Allin & Hopson, 1992; Hopson, 1987, 1994; Crompton & Parker, 1978; Kemp, 1982; Sloan, 1983; Carroll, 198
. Ultimately, in Middle and Upper Jurassic mammals, the tiny quadrate, articular, and ring-like angular migrated as a unit to the middle ear where they joined the stapes and became the incus, malleus, and tympanic bones (Allin, 197 5, 1986; Allin & Hopson, 1992; Hopson, 1987, 1994; Kemp, 1982; Sloan, 1983; Carroll, 198
.
Progressively, the teeth became differentiated. The large canines developed first, followed by the development of multicusped cheek teeth, reduced tooth replacement (Osborn & Crompton, 1973; Crompton & Parker, 197
, and finally full y differentiated incisors, canines, premolars, and molars with one tooth replacement during life (Kemp, 1982; Hopson, 1994).
Many other morphologic changes are documented in the fossil record. These demonstrate the morphologic and geochronologic succession from sprawling limb posture to upright limb posture of mammals (Jenkins, 1971; Romer & Lewis, 197 3; Kemp, 1982; Carroll, 1988; Hopson, 1994). As Jenkins (1971, p. 210) stated, “In details of morphology and function, the cynodont post-cranial skeleton should be regarded as neither ‘reptilian’ nor ‘mammalian’ but as transitional between the two classes .” Other changes have been adequately summarized elsewhere (Kemp, 1982; Sloan, 1983; Carroll, 1988; Hopson, 1994). Obviously, fundamental physiologic changes must have taken place as well, many of which are not directly preserved in the fossil record, though some can be inferred from the skeletal anatomy (Findlay, 1968; Kemp, 1982; Sloan, 1983, Carroll, 1988; Hopson, 1994).
This is well documented in the fossil record by a massive volume of incontrovertible data that cannot be explained away. Such large-scale, progressive, continuous, gradual, and geochronologically successive morphologic change (Sidor & Hopson, 199
is descent with modification, and provides compelling evidence for evolution on a grand scale.
Scientific Creation
There are a variety of different forms of creationism (Scott, 1999b). In this article, I am limiting my discussion to “scientific creation” (= young earth creation; Scott, 1999b) because its proponents were instrumental in the Kansas dec ision. This is clearly indicated by the removal of the earth’s age from the new curriculum (Scott, 1999a, p. 21).
Gish (1985, p. 35) defined the “creation model” as follows: “By creation we mean the bringing into being of the basic kinds of plants and animals by the process of sudden, or fiat, creation described in the first two chapters of Gene sis. Here we find the creation by God of the plants and animals, each commanded to reproduce after its own kind using processes which were essentially instantaneous. We do not know how God created, what processes he used, for God used processes which are not now operating anywhere in the natural universe. This is why we refer to divine creation as special creation. We cannot discover by scientific investigations anything about the creative processes used by God.” “Scientific creationists” consider the earth to be approximately 6,000 to 10,000 years old (Gish, 1995, p. 4
, and the creation event to have lasted six 24-hour days (Whitcomb, 1986, p. 28, 32; Gish, 1995, p. 4
.
It is important to note that, “Creationists do not deny...the origin of variations within kinds, but they do deny...the evolutionary origin of basically different types of plants and animals from common ancestors” (Gish, 1995, p. 30) . Thus, the “creation model” presents a picture of all of the “kinds” created during the creation week. Since then, no new “kinds” have emerged. However, there has been diversification within each “kind” to produce, for example, the different breeds of do gs. This is the “creationist orchard” (Sarfati, 1999, p. 38, 39). This “variation within kinds” results from each of the “kinds” being created with a range of genetic material.
Because all the major “kinds” of organisms were created during the initial creation event, “...the organisms represented in the fossil record would all have been living contemporaneously, rather than scattered in separate time-frames over hundreds of millions of years...The only reason to think that all should not have been living contemporaneously in the past is the assumption of evolution. Apart from this premise, there is no reason to doubt that man lived at the same time as the d inosaurs and trilobites” (Morris, 1985, p. 112). The “kinds” of organisms presumably lived in different ecologic zones just as they do today (Morris, 1985, p. 118, 119). Consequently, “creation scientists” propose an alternative model to uniformitarian historical geology, in whic h a world-wide “great flood” formed the rock record, especially the fossiliferous Phanerozoic sediments (Whitcomb & Morris, 1961, p. 258, 265, 327; Morris, 1985, p. 117, 118, 123, 129; Gish, 1995, p. 49; Brown, 1996, p. 84-86). During the “great flood,” organisms were hydrodynamically sorted according to size and shape (Morris, 1985, p. 118, 119), and the ecologic zone they lived in.
In synthesis, the “creation model” includes a number of critical tenets that are drastically different from evolution, geology, biology, and paleontology: 1) gaps between created kinds, i.e., no transitional fossils; 2) “great flood;” 3) instantaneous creation and contemporaneity of faunas; and 4) young age of earth. The following sections review these tenets.
Missing Links
“Missing links” are probably the most widely known argument against evolution. “Creation scientists” claim that transitional fossils do not exist and that systematic gaps between taxa (especially higher taxa) are ubiquitous in the fossi l record (Morris, 1985, p. 78, 79; Morris & Parker, 1987, p. 11, 12; Gish, 1995, p. 80, 81, 100-103, 109, 186, 187). Concerning the lack of transitional fossils connecting ichthyosaurs with other reptiles, Gish (1995, p. 109) stated, “What we have is undoubted proof of special creation, if ever such proof is possible.” It is certainly true that transitional forms have not been found to bridge the gap between some taxa. Transitional fossils connecting the ichthyosaurs or pterosaurs with other reptiles, or the bats with other mammals are not known (Carroll, 1988, p. 251-254, 331-337, 463, 464). However, this is absence of data, and the absence of data by itself is ambiguous. It is certainly not proof of anything. That is a serious logical fallacy and a fatal flaw in “creation science” arguments against evolution. What is relevant is that are many other transitional fossils, which corroborate evolution but contradict “creation science.”
The absence of data can be interpreted in a number of different ways. First, some transitional forms may have been soft-bodied organisms that were never fossilized. A few exceptional fossil deposits (such as the Burgess Shale; Dott & amp; Prothero, 1994, p. 216-219) demonstrate that significant numbers of soft-bodied organisms lived with skeletal organisms, but that for the most part they are not preserved. Second, transitional forms may not have been preserved because of geographic a ccidents: some organisms lived in actively eroding, instead of depositing areas, or they were destroyed in an orogeny (or major erosional event). Third, transitional forms may not have been found yet. As discussed earlier, prior to the early 1980’s no tra nsitional forms were known to connect whales with their most closely related fossil relatives, the mesonychids. Since then, an impressive succession of intermediate forms has been found clearly documenting their evolution. Science progresses by finding ne w data. Fourth, transitional forms may have existed for such a short duration of geologic time that they were not preserved, resulting in evolution by punctuated equilibrium. Fifth, transitional forms may have lived elsewhere and the organisms subsequentl y migrated to other locales. In the upper Upper Cretaceous (Campanian and Maastrichtian) of North America, we find abundant ceratopsian dinosaurs with no obvious local ancestors (Carroll, 1988, p. 309-311). However, in eastern Asia, we find two forms in t he upper Lower and lower Upper Cretaceous, Psittacosaurus and Protoceratops, which bridge the gap between generalized bipedal ornithopods and classical quadrupedal ceratopsians (Carroll, 1988, p. 309-311; Norman, 1998, p. 128-133). Apparentl y, the protoceratopsians migrated to North America during the middle of the Upper Cretaceous and an adaptive radiation ensued resulting in a plethora of genera, including the most famous, Triceratops (Edwords, 1982; Carroll, 1988, p. 309-311).
Critique of the Fossil Evidence
Not only do “creation scientists” misinterpret the ambiguity of missing links, they also do not accept the evidence for transitional fossils that does exist. An instructive lesson is a careful examination of their arguments against the evolution of mammals from synapsid reptiles (Gish, 1995). First, concerning amphibians, reptiles, and mammals, Morris (1985, p. 83) stated, “All of them are four-legged vertebrates with similar skeletal s tructures and thus their fossilized remains provide little basis for distinguishing between them.” That statement is partially incorrect; there are, in fact, many skeletal features which distinguish these classes (Carroll, 198
.
Second, as is typical of “creation science” books, Gish (1995) presented a series of quotes that have been taken out of context from various articles about synapsid evolution and fitted together to form a story that misrepresents the meaning of the original authors’ works. That is not corroborative evidence for “creation science,” and it does not disprove evolution.
Third, Gish (1995, p. 149) claimed that, “In their attempts to establish an evolutionary tree or phylogeny for the mammal-like reptiles and the mammals, evolutionists rely almost entirely on similarities to link these creatures in an evolutionary scenario. They are forced to do this because of the lack of transitional forms required for their hypothetical evolutionary ladder.” Further, Gish claimed that large and systematic gaps separate the major groups of synapsids from each other and from mammals (Gish, 1995, p. 151, 159-163). Gish (1995, p. 161) claimed, “The fossilized remains of each of these stages appear fully-formed, with no transitional forms documenting the gradual transition of one stage into the next, and very little, if any, further change occurs until this stage or level is abruptly replaced by the next.” All of these claims are directly contradicted by fossil evidence. Virtually none of the groups appear fully formed nor exhibit very little internal evolution. Within the sphenacodonts, from the Upper Pennsylvanian to Lower Permian, the angular notch is noticeably deepened and the reflected lamina became a noticeable flange (Currie, 1977, 1979; Romer & Price, 1940; Hopson, 1994). Within the cynodonts, significant e volution took place. The oldest cynodonts from the upper Upper Permian have an incomplete bony secondary palate (Parrington & Westoll, 1940; Hopson, 1987, 1994; Kemp, 1979). Subsequently, in Triassic cynodonts, the palate was completed. The bony palat e was progressively formed and sutured, not instantly formed (Parrington & Westoll, 1940; Parrington, 1946; Hopson, 1987, 1994; Fourie, 1974; Romer, 1969b, 1970b, 1973; Carroll, 198
. Also, the reflected lamina was progressively and gradually transfo rmed into a tiny horseshoe-shaped bone (Allin, 1975; Allin & Hopson, 1992; Hopson, 1987, 1994; Brink, 1963; Kemp, 1979; Fourie, 1974; Romer, 1969b, 1970b, 1973; Carroll, 198
.
Fourth, Gish (1995, p. 164) stated that, “It is apparent that the argument for linking pelycosaurs to the therapsids is extremely weak and is based solely on certain similarities. There are no transitional forms that would provide ac tual evolutionary links between pelycosaurs and therapsids.” Here Gish quoted Romer & Price (1940) out of context and altered the meaning of their conclusions. In fact, the fossil evidence contradicts Gish’s claim: the oldest therapsids and sphenacodo nts are very similar in nearly every aspect of their morphology, especially the skulls (Romer & Price, 1940; Currie, 1977, 1979; Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972). This was demonstrated 90 years ago (B room, 1910)!
Fifth, Gish (1995, p. 165) stated that, “It must be emphasized that the non-cynodont therapsids appeared abruptly, that is, with all their basic characteristics complete ...There are no transitional forms, no intermediates, that link these...to some hypothetical pelycosaur ancestor.” That statement is contradicted by the fossil record. The therapsids do not appear fully formed; the oldest are more similar to the pelycosaurs than to the cynodonts in their morphology (Romer & Price , 1940; Currie, 1977, 1979; Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972). The eotitanosuchians are in fact intermediates, as is clearly indicated by the enlargement of the lateral temporal fenestra and reflected lami na compared with the pelycosaurs but not to the degree of the other younger therapsids, including the therocephalians. Moreover, the eotitanosuchians do not have a coronoid process (like the pelycosaurs) but do have well developed canines (like subsequent therapsids; Romer & Price, 1940; Currie, 1977, 1979; Tchudinov, 1960, 1983; Efremov, 1954; Olson, 1962; Sigogneau & Tchudinov, 1972).
Sixth, Gish (1995, p. 166) stated that, “the cynodonts...are found at the earliest levels in rocks of the Late Permian.” That is incorrect; the oldest cynodonts are in fact from the late part of the Late Permian (Kemp, 1979, 1982).
Seventh, concerning Morganucodon, Gish (1995, p. 169) stated, “What is the evidence for a squamosal-dentary joint in these creatures? This evidence consists of an alleged condyle on the dentary.” That statement misrepresents t he data. The condyle is definitely present; it is not “alleged” (Kermack, Mussett, & Rigney, 1973, 1981; Carroll, 198
. The corresponding shallow depression in the squamosal into which the condyle fitted in life corroborates this. Hopson (1987) state d that articulated material has been found with the two bones in contact.
Eighth, Gish (1995, p. 169, 170) stated that, “The anatomy required for (the reptilian jaw-joint), including the arrangement and mode of attachment of musculature, the arrangement and location of blood vessels and nerves, etc., must be quite different from that required for a mammalian jaw-joint. How then could a powerful, fully functional reptilian jaw-joint be accommodated along with a mammalian jaw-joint?” That statement is irrelevant because fossils of Morganucodon conclus ively demonstrate that the two jaw joints did exist side-by-side (Kermack, Mussett, & Rigney, 1973, 1981; Carroll, 198
. That is incontrovertible evidence that cannot be explained away. Moreover, the new mammalian jaw musculature developed progressiv ely throughout the cynodonts, beginning with the Upper Permian procynosuchids (Barghusen, 1968; Crompton, 1972; Crompton & Parker, 1978; Hopson, 1987, 1994). The new muscles progressively expanded onto more and more of the coronoid process and posteri or part of the dentary (Barghusen, 1968; Crompton, 1972; Crompton & Parker, 1978; Hopson, 1987, 1994). Thus, the mammalian jaw musculature evolved while the lower jaw was still hinged to the skull only by a reptilian jaw joint, and was already present in Morganucodon .
Ninth, Gish (1995, p. 171) stated that, “Finally, and this is conclusive, not a single intermediate between an animal with a powerful, fully functional reptilian jaw-joint and a powerful, fully functional mammalian jaw-joint has been found.” That statement is incorrect. Morganucodon contains both jaw joints side-by-side (Kermack, Mussett, & Rigney, 1973, 1981; Carroll, 198
. Tenth, concerning the evolution of the middle ear, Gish (1995, p. 167, 16
stated that, “Another difficulty with the above notion is the fact that while thousands of fossils have been found which possess a single ear bone and multiple jaw bones, and thou sands of fossil mammals have been found which possess three ear bones and a single bone in the jaw, not a single fossil creature has ever been found which represents an intermediate stage, such as one possessing three bones in the jaw and two bones in the ear.” That misrepresents how the mammalian middle ear evolved. As demonstrated by the fossil record, the quadrate and articular became part of the middle ear as a unit, along with the angular (Allin, 1975, 1986; Allin & Hopson, 1992; Hopson, 1966, 1987, 1994; Ro sowski, 1992; Crompton, 1972; Crompton & Parker, 197
. Gish (1995, p. 171) continued by stating, “This would have required that the stapes (columella) of the reptile become free from its attachment to the tympanum (ear drum), and the retroarticular p rocess of the articular gain an attachment to the tympanum...Somehow, while all of this was going on, the quadrate bone of the reptilian ancestor must gain its freedom, move into the middle ear, and insert itself between the stapes and malleus...” That is false.
First, in the oldest reptiles, including synapsids, the stapes extended from the inner ear to the quadrate; a lizard-like tympanum was never present (Carroll, 1964, 1988; Romer & Price, 1940; Allin, 1975, 1986; Allin & Hopson, 1992).
Second, in therapsids, especially cynodonts, the tympanum was suspended across the angular cleft supported in part by the reflected lamina and the other post-dentary bones including the articular (Allin, 1975; Allin & Hopson, 1992).
Third, the retroarticular process of the articular already had contact with the tympanum (Allin, 1975; Allin && Hopson, 1992).
Fourth, the quadrate was already present between the stapes and the articular, not necessitating any such movement (Allin, 1975; Allin & Hopson, 1992).
All of this is documented by the fossil record. Concerning his version of origin of the mammalian middle ear, Gish (1995, p. 171) stated, “There is absolutely no fossil evidence whatsoever to support such an incredible scenario.” That’s because his versi on of the origin of the mammalian middle ear is wrong.
Eleventh, continuing about the soft part anatomy of the mammalian inner ear, Gish (1995, p. 172) stated that, “The organ of Corti...has no homologue in reptiles. There is no possible structure in the reptile from which it could have been derived. It would have had to have been created de novo, since it was entirely new and novel.” That statement is incorrect. In fact, the basilar papilla of the reptilian inner ear is homologous with the organ of Corti (Romer, 1956, p. 33-38, 1970a; S loan, 1983). The basilar papilla, or organ of Corti, is contained in one portion of the inner ear known as the cochlear canal (Romer, 1956, p. 33-38, 1970a; Allin & Hopson, 1992). The cochlear canal is encased in bone and endocasts can be made that deline ate its shape (Allin & Hopson, 1992). In cynodonts, the cochlear canal is a short blunt feature (Allin & Hopson, 1992). In Morganucodon, it is elongate and slightly curved (Allin & Hopson, 1992). In modern mammals it is long and coiled (Allin & Hopson, 1992). Indeed, even the inner ear did not appear fully formed but progressively evolved.
Twelfth, Gish (1995, p. 152) stated that the succession of synapsid taxa “...must be juxtaposed according to an imagined evolutionary sequence, or at least according to some scheme determined by assumptions based on indirect evidence .” That is incorrect. In fact, the evidence for the geochronologic order of the various synapsid-bearing strata is quite good and independent of any evolutionary assumptions (Kemp, 1982; Frenzel, et al., 1988; Jones & Hentz, 1988; Olson, 1957, 1962; Tchudinov 1965; Efremov & Vyushkov, 1955; DuToit, 1954; Keyser & Smith, 1977-78; Bonaparte, 1966; Romer, 1966; Romer & Jensen, 1966; Waterhouse, 1978; Ross, 1979; Harland et al., 1989). In conclusion, everything that is known abou t the fossil record of synapsid reptiles and early mammals contradicts not only the basic predictions of the “creation science” model, but also contradicts point-by-point most of the detailed discussion of the topic by Gish (1995). The same conclusion can be said for “creationist scientists’” interpretations of the remainder of the fossil record (Gish, 1978, 1985, 1995). One paleontologist’s critique of Gish (197
is: “On 67 of the 97 text pages I found at least one error of fact, logical error, or quota tion out of context, all chosen carefully to mislead the reader. On checking a standard college logic text with a list of logical fallacies, I found that Gish did not manage to miss a single one! Their works have the appearance of scholarship, but not the substance” (Sloan, 1983, p. 263).
Created Kinds
“Scientific creationists” classify organisms not by standard Linnean taxonomic procedures, but rather group them into basic “kinds” (the terms “type” and “kind” are apparently synonymous; Gish, 1995, p. 29-31). As Morris & Parker (1987, p. 137, 13
stated, “For creationists, it’s the created type that is the real unit in nature.” Gish’s (1995, p. 29) definition is that, “A basic animal or plant type would include all animals or plants which were truly derived from a single stock.” As e xamples of “basic kinds,” Gish (1995, p. 30) offered the following: “Among the vertebrates, the fishes, amphibians, reptiles, birds, and mammals are obviously different basic types. Among the reptiles, the turtles, crocodiles, dinosaurs, pterosaurs..., an d ichthyosaurs...would be placed in different kinds. Each one of these major groups of reptiles could be further subdivided into the basic kinds within each. Within the mammalian class, duckbilled platypuses, opossums, bats, hedgehogs, rats, rabbits, dogs , cats, lemurs, monkeys, apes, and men are easily assignable to different basic types. Among the apes, the gibbons, orangutans, chimpanzees, and gorillas would each be included in a different basic kind.”
That is curious: “kinds” are identified to exist as a hierarchy, “kinds” within other “kinds.” Yet Gish (1985, p. 34; 1995, p. 35) stated both that God separately created all of these basic animal and plant “kinds,” and that a “kind” includes those variants which have been derived by genetic variation from a single stock. If a “kind” consists of all those variants derived from a single stock, then how can some of the variants also have been created separately? At what level did God r eally create? This is both internally inconsistent and a major logical fallacy. Furthermore, it renders “creation science” neither falsifiable nor scientific. Scientists, and to some extent “creation scientists,” both agree that finding transitional forms between taxa (or “kinds”) would falsify “scientific creationism” (Cuffey, 1984; Gish, 1995, p. 40, 41). But, given the internally inconsistent definition of “basic kind,” what should be looked for? If intermediates between two species of the brachiopod < i>Eocoelia are found, it can be explained as “variation within the Eocoelia kind.” If intermediates connecting Hyracotherium with Equus are found, it can be explained as “variation within the horse kind.” If intermediates between mesonychid ungulates and whales are found, it can be explained as “variation within the mammal kind.” If intermediates between reptiles and mammals are found, it can be explained as “variation within the vertebrate kind.” One could also argue that such in termediates are separate “created kinds.” For example, Morris & Parker (1987, p. 137) stated that, “Because of its unique combination of complete, functionally integrated traits, Archaeopteryx would qualify as a created type.” That is an illogi cal semantic game that renders “creation science” unscientific.
“Scientific creationists” equate “variation within a kind” with microevolution and origin of new “kinds” with macroevolution (Brown, 1986; Morris, 1994). That misrepresents both concepts. As previously stated, microevolution results in the origin of new species. As defined by Gish, some “kinds” are species (such as humans; Gish, 1995, p. 29) and so their origin would be by microevolution, not macroevolution, and thus could be considered “variation within a kind.” That internal incons istency thus implies that humans originated by microevolution and were not specially created. Furthermore, some “kinds” as defined by Gish include not just many species, but also more than one higher taxon, and so their origins include both macroevolution and microevolution.
Faunal Succession and Correlation
“Scientific creationists” reject faunal succession and the geologic time scale (Whitcomb & Morris, 1961; Morris, 1985; Bliss, 1988; Gish, 1995). That is an attempt to discredit evolution. If one can show that superposition, faunal s uccession, biostratigraphy, correlation, and the geologic time scale are invalid, then the geochronological successions of transitional fossils would be rendered invalid. Thus, “creation scientists” propose that the successive appearance of taxa in the fo ssil record is the result of ecologic zonation (Morris & Parker, 1987, p. 163-165), mobility of vertebrates (Whitcomb & Morris, 1961, p. 275-277; 279-286), and hydrodynamic (Whitcomb & Morris, 1961 p. 273, 274; Morris, 1985) or liquefaction so rting (Brown, 1996) during the “great flood.” Significant evidence (e.g., C. Cuffey, 1999) indicates that the rock record is definitely not the result of a “great flood,” and therefore any hypothesis about the origin of faunal succession involving the “great flood” is also false and irrelevant.
“Scientific creationists” assert that relative age dating of rocks uses circular reasoning based on the assumption of evolution (Whitcomb & Morris, 1961, p. 134, 136, 203, 205; Morris, 1978; Morris, 1985, p. 94-96, 134-137, 229, 232; Morris & Parker, 1987, p. 239-242; Bliss, 1988, p. 36). For example, Morris (1985, p. 136) stated, “Here is obviously a powerful system of circular reasoning. Fossils are used as the only key for placing rocks in chronological order. The criterio n for assigning fossils to specific places in that chronology is the assumed evolutionary progression of life; the assumed evolutionary progression is based on the fossil record so constructed. The main evidence for evolution is the assumption of evolutio n!”
That is false. Rocks are placed in proper chronological order by superposition (Van de Fliert, 1968; Dott & Prothero, 1994, p. 16-41, 73-91). Therefore, the succession of rock layers accurately represents the history of the earth . Thus, the fossil record contained within accurately represents the history of life on earth. If we collect fossils from stratigraphic sections where superpositional relationships can be easily determined solely on physical criteria, we find that there i s a definable succession of fossil taxa, each occurring in a limited stratigraphic interval. This pattern is an empirical observation that is both verifiable and repeatable. Within a local area, where the correlations between stratigraphic sections can be determined on physical criteria (geologic mapping, well log correlation), we find very similar successions of taxa at each of the sections (Lochman-Balk, 1971; Palmer, 1971; Key, 1990; Stitt, 1971, 1977). Such physical correlations are spectacularly illustrated in the Lower Ordovician of the Arbuckle Mountains of south-central Oklahoma and the Upper Pennsylvanian and Lower Permian of north-central Oklahoma and east-central Kansas, where one can literally walk out lithostratigraphic units for miles across the prairies. If we expand our area of investigation, we fin d that the same succession of fossils repeatedly occurs elsewhere. It can be demonstrated by regional geologic mapping and subsurface correlation that the rock layers containing each specific assemblage of fossils are physically correlative. This is spect acularly illustrated on the Colorado Plateau, where one can clearly trace the rock units by physical correlation.
This is the principle of faunal succession. It does not rely on untestable evolutionary assumptions. Faunal succession was first demonstrated by Smith in England, and by Cuvier and Brogniart in France, in the late 1700s and earliest 1800s (Dott & Prothero, 1994 p. 23-25). Evolution was not assumed; in fact, Cuvier believed that all life was created early, and d’Orbigny believed that different faunas were repeatedly created and wiped out by God (Dott & Prothero, 1994, p. 25, 2 6). Such observations preceded Darwin’s Origin of Species by more than 50 years.
Faunal succession can be readily applied in our own experiences. While growing up and attending college in Pennsylvania and Ohio, I collected fossils from all over the western Appalachian Valley and Ridge, the Appalachian Plateau, th e Great Lakes region, and the Ohio River valley. Throughout this region, rocks containing the trilobite Cryptolithus (Shimer & Shrock, 1944; Hoskins, 1969; Feldmann & Hackathorn, 1996) are always overlain by rocks with the brachiopods St egerhynchus, Whitfieldella, and Eospirifer, and halysitid corals (Shimer & Shrock, 1944; Hoskins, 1969; Feldmann & Hackathorn, 1996); which are always overlain by rocks with the brachiopods Paraspirifer and Mucrospirife r and the trilobite Phacops (Shimer & Shrock, 1944; Kesling & Chilman, 1975; Hoskins, 1969; Feldmann & Hackathorn, 1996); which are always overlain by rocks containing the bryozoan Archimedes and the blastoid Pentremites (Shimer & Shrock, 1944; Galloway & Kaska, 1957; McKinney, 1999). This succession is invariant; I have never observed Archimedes stratigraphically below, nor with, Cryptolithus, for example. This succession can be independently dem onstrated by physical stratigraphy to be in the same superpositional order. Based on local and regional scale geologic mapping, and subsurface correlation, the rock intervals from different regions and containing the same fauna can be demonstrated to be p hysically correlative. These genera are not in a specific evolutionary ancestor-descendant relationship and so the succession is definable without any underlying assumption of evolution. Faunal succession is an empirical observation, not an evolutionary a ssumption.
At a more detailed level, the development of Ordovician graptolite biostratigraphy in North America provides a good case study of biostratigraphic methods based on faunal succession (Berry, 1977), and one that is independently testab le (Goldman et al., 1994; Mitchell et al., 1994). Fifteen graptolite biozones have been recognized, defined, and refined by nearly a century of detailed work. Based on superpositional order, the same succession of graptolite species and zone s is recognized in New York (Ruedemann, 1904, 1908, 1912, 1925, 1947; Berry, 1962, 1963, 1970; Mitchell et al., 1994; Goldman et al., 1994), Quebec (Riva, 1969, 1974), Newfoundland (Kindle & Whittington, 1958; Whittington & Kindle, 1 963), west Texas (King, 1937; Berry, 1960; Bergstrom, 197
, Yukon (Jackson, 1964; Jackson & Lenz, 1962), and east-central Alaska (Churkin & Brabb, 1965). Moreover, isolated localities with only short stratigraphic sections can be compared with po rtions of the zonation defined elsewhere (Ross & Berry, 1963). No assumption of evolution was made. The fact that this same succession occurs repeatedly in different regions all over North America, and that the succession can be independently verified by anyone willing to recollect the localities, leads to the conclusion that geochronologic correlation based on biostratigraphy is valid.
This conclusion can be independently tested. The Middle Ordovician Trenton Group and Utica Formation of New York contain three of the graptolite zones and also contain numerous K-bentonite beds (Goldman et al., 1994; Mitchell et al., 1994). Because the K-bentonite beds are volcanic ash-falls, they represent geologically instantaneous isochrons. Furthermore, trace element geochemistry of their contained volcanic glass distinguishes each K-bentonite bed, assures proper co rrelation, and establishes a geochronologic framework to which the graptolite zones can be compared (Goldman et al., 1994; Mitchell et al., 1994). Indeed, the graptolite zone boundaries are parallel with the K-bentonite beds, therefore indep endently corroborating graptolite zones as time-parallel (Goldman et al., 1994; Mitchell et al., 1994). This independently demonstrates the validity of biostratigraphy on a local scale, and corroborates its use as a method of geochronologic correlation.
Based upon similar detailed biostratigraphic correlations, augmented with physical stratigraphic correlations, mapping, and subsurface correlations, it has long been thought that the Decorah (Minnesota), Spechts Ferry (Iowa, Wisconsi n, Illinois), Tyrone (Kentucky), Eggleston (Virginia), and Carters (Tennessee) formations, and the Chickamauga/Stones River/Nashville (Alabama, Georgia) groups throughout the eastern United States are correlative with each other. These formations each con tain a pair of thick bentonite beds (Deicke and Millbrig bentonites), within the upper part of the Phragmodus undatus conodont zone (Huff & Kolata, 1990). Geochemical fingerprinting with rare earth elements independently demonstrates that in ea ch locale, it is the same pair of bentonite beds. This independently demonstrates both that these formations are correlative and the validity of biostratigraphy on a local and regional scale (Huff, 1983; Kolata, Frost, & Huff, 1987; Huff & Kolata, 1990; Kolata, Huff, & Bergstrom, 1996). Moreover, biostratigraphic studies indicated that these formations were equivalent to the middle Caradocian of Estonia. There, the Big Bentonite is found at the base of the Keila Formation (Huff, Bergstrom, &am p; Kolata, 1992). Both the Millbrig and the Big Bentonite contain the same phenocryst composition, geochemical fingerprint, and Ar-Ar radiometric age (Huff, Bergstrom, & Kolata, 1992). This independently demonstrates the validity of biostratigraphy on a world-wide scale. Similar independent tests have been done on Cretaceous biozones (Wise, 1998, p. 165) and Cenozoic biozones (Evernden et al., 1964). In both cases, the biostratigraphic framework was found to be consistent with radiometric ages.
Morris (1985, p. 95) quoted Evernden et al. (1964), as stating “Vertebrate paleontologists have relied upon ‘stage of evolution’ as the criterion for determining the chronologic relationships of faunas.” That appears to cast d oubt on the validity of the succession of Cenozoic land mammal faunas from western North America. That quote is, however, taken out of context and consequently misrepresents the conclusions of Evernden et al. (1964). In fact, the very next sentence in Evernden et al. (1964, p. 166), states that, “The physical dates presented in this paper demonstrate that temporal position of genera and species of fossil mammals in their accepted phylogenies is accurate at Mammal-Age degree of refinement.” Moreover, the overall conclusion of Evernden et al. (1964, p. 145), was that, “The K/A ages and the Mammal Age designations are in essentially perfect agreement, thus substantiating the usefulness of the K/A techni que throughout the Tertiary and supporting the conclusion that the defined Mammal Ages have true evolutionary significance. Correlations with European ages and Pacific Coast foraminiferal Ages through both K/A and fossil criteria are internally consistent .” And furthermore, “The correspondence of the two sets of data discussed in this paper is so close as to leave little doubt that the defined Land-Mammal Ages are time-sequential.”
Faunal succession directly contradicts the predictions of “creation science.” All taxa, including higher taxa from genera to phyla, do not appear simultaneously in the fossil record. The oldest occurrences of the animal phyla and cl asses range from the uppermost Precambrian (Phylum Cnidaria) through the Upper Jurassic (Class Aves; Benton, 1993). Likewise, the oldest occurrences of the classes of vascular land plants range from the Middle Silurian (possibly Upper Ordovician) through the Lower Cretaceous (Stewart, 1981). Furthermore, all taxa did not live contemporaneously as in a post-creation, pre-flood world. This is true not only if we examine species, but also higher taxa, from genera to phyla (Benton, 1993). Of special note here is the "Cambrian explosion." "Creation scientists" imply that the major taxa of invertebrates appeared suddenly, essentially contemporaneously, and fully formed in the Cambrian (Gish, 1995, p. 54-69, 75; Morris, 1985, p. 80, 81; Morr is & Parker, 1987, p. 126-129). That is incorrect. In fact, the oldest occurrences of the major taxa of invertebrates range from the uppermost Precambrian through the Upper Ordovician (Boardman, Cheetham, & Rowell, 1986; Taylor & Larwood, 19 90; Benton, 1993; Gehling, 1986; McMenamin, 1987; Ausich, 1997; Ausich & Babcock, 199
. Even considering just those major taxa that first occur in the Lower Cambrian, the first occurrences are still successive within the Lower Cambrian, not simultan eous (Narbonne et al., 1987; Smith, 1990; Signor & Lipps, 1992; Mount & Signor, 1992; Rozanov & Zhuravlev, 1992; Jiang, 1992; Briggs & Fortety, 1992; Popov, 1992; Debrenne, 1992; Crimes, 1992; Sprinkle, 1992; Landing, 1988, 1989, 19 92, 1994; Landing et al., 1989; Bowring et al., 1993; Isachsen et al., 1994). Moreover, the organization of at least some ecologic communities slowly developed throughout the Cambrian and Ordovician (Ausich & Bottjer, 1982, 1990; Bottjer & Ausich, 1986). When one examines Cambrian and Ordovician rocks, it is readily apparent that a typical, diverse "Paleozoic fauna," dominated by brachiopods, bryozoans, and crinoids, first occurs in Middle Ordovician rocks. Morris & Parker (1987, p. 127) and Bliss (1988, p. 39) presented a diagram of life on a Cambrian seafloor, including sea urchins and starfish. That is incorrect. The oldest starfish are Lower Ordovician (Benton, 1993, p.507) and the oldest sea urchins are Upper Ordovician (Benton, 1993, p.511). Bliss ( 1988, p. 41) stated, "Many complex invertebrates are found in Cambrian strata; the ones living today are similar." The second part of that statement is simply incorrect; Cambrian and modern invertebrates are easily distinguished.
Morris (1985, p. 227, 22
and Whitcomb (1986, p. 75, 76, 84), based upon their interpretations of Genesis, provided a list of predictions about the order of creation of organisms:
land plants were the first life forms created (Morris, 1985, p. 227, 228; Whitcomb, 1986, p. 75);
fruit trees before fishes (Morris, 1985, p. 227, 228; Whitcomb, 1986, p. 75);
birds before insects (Morris, 1985, p. 227, 22
;
birds before reptiles (Morris, 1985, p. 227, 228; Whitcomb, 1986, p. 76);
birds contemporaneous with fishes (Whitcomb, 1986, p. 76);
insects after flowering plants (Whitcomb, 1986, p. 76); and
whales before land mammals (Whitcomb, 1986, p. 84).
Because the fossil record accurately represents the history of life, we can test these predictions.
First, the oldest land plants (mosses and vascular plants) are from the Middle Silurian (possibly the Upper Ordovician) and are definitely not the oldest fossils (which are cyanobacteria from the Archaean; Benton, 1993, p. 779, 781; Stewart, 1981; Dot t & Prothero, 1994, p. 195, 286-28
.
Second, the oldest fishes (all classes of fishes) are definitely older than the oldest fruit trees (angiosperms); the oldest occurrences of the various fish classes ranges from Upper Cambrian to Lower Devonian (Benton, 1993, p. 574, 584, 589, 590, 594 , 611-613) whereas the oldest fruit trees are Lower Cretaceous (Stewart, 1981; Doyle, 1977).
Third, the oldest insects (middle Lower Devonian; Benton, 1993, p. 365) are definitely older than the oldest birds (upper Upper Jurassic; Benton, 1993, p. 717, 71
.
Fourth, the oldest reptiles (either Lower Mississippian or Middle Pennsylvanian; Benton, 1993, p. 681, 683) are definitely older than the oldest birds.
Fifth, birds and fish do not appear in the fossil record at the same time; the oldest fish are older than the oldest birds.
Sixth, the oldest insects are definitely older than the oldest angiosperms (“flowering plants”); even if “flowering plants” are considered to include the cycadeoids (Upper Triassic; Stewart, 1981, p. 289), the insects are still older.
Seventh, the oldest land mammals (Upper Triassic; Kermack, Mussett, & Rigney, 1973, 1981; Benton, 1993, p. 740) are definitely older than the oldest whales (Lower Eocene; Gingerich et al., 1983). In every case, the predictions of “creation scientists” are contradicted by the fossil record. Both Morris (1985) and Whitcomb (1986) concede that the paleontological evidence does indeed contradict their predictions.
“Creation scientists” attempt to discredit faunal succession and evolution by discussing examples of “out-of-order fossils.” “Creation scientists” claim fossil human footprints, body parts, and artifacts have been found in pre-Plioce ne rocks (Helfinstine & Roth, 1994; Baugh & Wilson, 1987, 1992; Brown, 1996, p. 22, 23, 43, 61, 62). A thorough discussion of each is beyond the scope of this article. In every case that I was able to check, these remains are demonstrably not huma n, but instead either dinosaur footprints (Weber, 1981; Cole et al., 1985; Hastings, 1985; Godfrey, 1981, 1985; Kuban, 1986), fish teeth (Hastings, 1995), inorganic rock features (Conrad, 1981), or hoaxes (Lippard, 1989; Weber, 1981; Godfrey, 1985) . Moreover, some of the artifacts are obviously modern tools dropped or buried in ancient sediment or rock (Lippard, 1989; Cole, 1985). There is no credible evidence of humans living prior to the Pliocene. Instead, there is abundant evidence indicating ho minids evolved during the Pliocene (Johansen & Edgar, 1996; Lewin, 1993).
“Creation scientists” claim that Precambrian rocks overlying Cretaceous rocks (by thrust faulting) in Glacier National Park of western Montana render superposition and faunal succession unreliable (Whitcomb & Morris, 1961, p. 180 -200). They stated “...we feel warranted in rejecting the whole concept of over-thrusting, at least when applied on the scale of the so-called Lewis and Hart Mountain Thrusts...” That is incorrect. Thrust faults, including the Lewis Overthrust, can be ide ntified by geologic mapping and other physical criteria (Van de Fliert, 1968; Allmendinger, 1992).
Bliss (1988, p. 49) stated that Eohippus (=Hyracotherium) has been found with Equus in surface strata, thereby discrediting the evolution of horses. No references are given, no localities are given, no museum specimen numbers are given, no illustrations are given, and no precise stratigraphic information is given (the term surface strata is ambiguous; does he mean soil or any of the Phanerozoic rock formations cropping out at th e surface?). Thus his claim is not credible. According to Simpson (1951) and MacFadden (1992), Hyracotherium is not found in the same strata with Equus. Gish (1992, p. 36) stated that, “The most recent fossil of an armored dinosaur (buried i n the ice in Antarctica), was found in 1988 by an expedition to the South Pole.” This statement is false; dinosaurs have been found in Cretaceous sedimentary rocks in Antarctica, but not ice (Wise, 1998, p. 171).
