The American Biology Teacher
Volume 55, No.2, February 1993, 78-83

Origin of Life & Evolution in Biology Textbooks - A Critique

Gordon C. Mills, Professor of Biochemistry, University of Texas Medical Branch, Galveston, TX
Malcolm Lancaster, Professor, Dept. of Family Practice, University of Texas Health Science Center, San Antonio, TX
Walter L. Bradley, Professor and Head, Dept. of Mechanical Engineering, Texas A&M University, College Station, TX

It has been noted by others that the states of Texas and California set textbook standards for the nation as a whole, primarily because of the large numbers of textbooks sold in those two states. The guidelines for textbook publishers in those states are Proclamation 66 (Texas) and the California Framework. Because of many differences in the two documents and because of the authors' participation in Texas adoptions, this critique will be limited to a comparison of Biology I textbooks with standards of Proclamation 66. However, it is believed that these criticisms should be pertinent to textbook evaluations nationwide. Criticisms are limited to chapters dealing with the Origin of Life and Evolution. It should be noted that we are not critical of all portions of these chapters. For example, descriptions of the experiments of Pasteur and others regarding spontaneous generation are generally well written. Also, portions on paleontology and classification of species are in most cases to be commended.

The books critiqued are limited to 1991 editions of Biology I textbooks adopted by the state of Texas as listed under References. The authors of this article recognize that textbook authors were given a mandate from the Texas State Board of Education to deal with the topics of Origin of Life and Evolution. Pertinent excerpts from Proclamation 66 that relate to these topics follow:

1. Scientific methods: under content: 1.4 scientific theories and laws based on existing evidence as well as new evidence; 1.6 problem solving (data collection and analysis, conclusion).

2. Important scientific discoveries and theories of the past. . .under content: 2.2 Pasteur's discoveries (non-spontaneous generation, rabies vaccine, experiments with anthrax); 2.6 Darwin's theory of evolution.

4. Specialization and functions of cells and cellular organelles: under content: 4.2 theory of chemical origin of life.

6. Drawing logical inferences, predicting outcomes and forming generalized statements: under process skills: 6.2 deducing a biological hypothesis from experimental data; 6.3 examining alternative scientific evidence and ideas to test, modify, verify or refute scientific theories.

9. Theories of evolution: under content: 9.1 scientific theories of evolution; 9.2 scientific evidence of evolution and other reliable scientific theories, if any; 9.3 mechanisms of evolution; 9.4 patterns of evolution.

Have the textbook authors and editors clearly followed the above guidelines such as item 6.3: "examining alternative scientific evidence and ideas to test, modify, verify or refute scientific theories"? This quotation is certainly an excellent expression of what constitutes valid science. Whether or not this guideline is followed is an important question for all biology teachers.

Origin Of Life Hypotheses: Credible Or Beyond Credibility?

Despite the abundant use of leading questions and tentative terminology in their origin of life discussions, the majority of textbooks exude confidence that confirmation of a naturalistic model of life's origin is inevitable. The treatment in these textbooks stands in marked contrast to a recent review article by Klaus Dose summarizing origin of life research. In this thorough review, a strikingly different picture emerges of the current state of affairs regarding the origin of life. Dose, one of the best known origin of life researchers for the past 20 years, in The Origin of Life: More Questions than Answers (Dose 1988, p. 348) provides the following summary:

More than 30 years of experimentation on the origin of life in the fields of chemical and molecular evolution have led to a better perception of the immensity of the problem of the origin of life on Earth rather than to its solution. At present all discussions on principal theories and experiments in the field either end in stalemate or in a confession of ignorance.

First, we will consider the validity of the atmospheric models used for origin of life experiments, followed by whether data from these experiments are properly evaluated and interpreted.

Clinging to Outdated Atmospheric Models

Comments like that quoted above and the objective tone of the entire review article by Dose stand in sharp contrast to the optimism that colors the treatment of life's origin in most of the biology textbooks. The latter generally give the impression that the origin of life problem is nearly solved, since amino acids and other small building blocks have been produced using simulated atmospheres. In regard to composition of the early atmosphere, the following statements illustrate inaccuracies or overstatements in some texts. "The atmosphere had no free oxygen as it does today. Instead, the air was probably made up of water vapor, hydrogen, methane and ammonia" (Biggs et al. 1991, p. 227). "The Earth's first atmosphere most likely contained water vapor (H20), carbon monoxide (CO) and carbon dioxide (CO2), nitrogen (N2 ), hydrogen sulfide (H25) and hydrogen cyanide (HCN)" (Miller & Levine 1991, p. 343). It is unfortunate that only a few of the books acknowledge that it is not likely that the earth ever contained an atmosphere comparable to those used in simulation experiments (Dose 1988, p. 351). The assumption that there was no oxygen in the early atmosphere is of crucial importance to the success of simulation experiments, yet there is no proof that oxygen was absent from that atmosphere.

Overstating the Experimental Results

In several of the textbooks, inconsistencies and overstatements regarding the nature of compounds produced in simulation experiments pose a second problem. In some cases false impressions are given because of what students are not told. Most texts fail to note that the compounds produced are markedly dependent upon starting materials and experimental conditions. Some quotes follow: "They found amino acids, sugars and other compounds just as Oparin had predicted" (Biggs et al. 1991, p. 228). "Nucleic acids and ATP also have been formed" (Biggs et al. 1991, p. 228). "Their experiments have produced a variety of compounds, including various amino acids, ATP and the nucleic acids in DNA" (Towle 1991, p. 210). "Similar mechanisms might have led to the formation of carbohydrates, lipids and nucleic acids." (Towle 1991, p. 210). "Thus, over the course of millions of years, at least some of the basic building blocks of life could have been produced in great quantities on early Earth" (Miller &: Levine 1991, p. 344). The texts fail to note that most of the compounds produced in Miller and Urey's original simulation experiment have no relevance to compounds found in living cells; that amino adds produced are always racemic (that is, D-, L-) mixtures; that carbohydrates and amino acids are never produced in the same experiment (they require different starting materials and different conditions); or that no one has produced any ATP or true nucleic acids using reasonable starting materials. As Dose (1988, p. 352) notes:

Substantial amounts of biologically relevant sugars, including D, L-ribose, have never been produced in realistic prebiotic simulation experiments.

They also neglect entirely the fact that compounds in cells have specific intramolecular bonds. Amino acids, carbohydrates, purines and pyrimidines all have many possible isomers, and in most cases only one, or at most very few of these isomers are found in living cells. In simulation experiments mixtures of isomers would usually be produced.

In regard to formation of proteins from amino acids, several quotations follow: "Other scientists have shown that amino acids will link up when heated in the absence of oxygen gas" (Towle 1991, p. 210). Also, ". . . amino acids tend to link together spontaneously to form short chains" (Miller & Levine 1991, p. 344). Neither of these texts notes that linkages occur only when amino acids are heated in the dry state; amino acids do not link together spontaneously in aqueous solution. Nor do these texts note that heating in the dry state produces some linkages that are not found in protein molecules, linkages that would prevent the formation of useful amino add sequences.

 Several quotations from the texts relating to membrane enclosures and/or cell formation are alive with expectation: "One process that must have occurred on the earth was the enclosure of nucleic acids in membranes. Once DNA was separated from the environment by some kind of boundary, it would be protected, and might be able to carry out the precise reactions of replication" (Towle 1991, p. 211). "Some of these droplets grow all by themselves, and others even reproduce" (Miller & Levine 1991, p. 344). These statements are pure speculation. Cell membranes usually contain lipids of various types, but they also contain proteins and carbohydrates. More importantly, membranes have very little to do with precise reactions of replication. Students are in no position to know it, but growth and division of coacervate droplets have no similarity to growth and reproduction of living cells.

The effect of the discussions in most of these texts is to make the emergence of life on Earth by chance appear to be highly probable. The following summary statement illustrates this:

If we just said that life did arise from nonlife billions of years ago, why couldn't it happen again? The answer is simple: Today's Earth is a very different planet from the one that existed billions of years ago. On primitive Earth, there were no bacteria to break down organic compounds. Nor was there any oxygen to react with the organic compounds.  As a result, organic compounds could accumulate over millions of years, forming that original organic soup. Today. However, such compounds cannot remain intact in the natural world for a long enough period of time to give life another start {Miller &: Levine 1991. p. 346).

It is not mentioned that degradation of organic compounds would occur in an early atmosphere as a result of electrical discharges, heat, ultraviolet light, etc.. opposing any accumulations of relevant organic compounds. Nor is it mentioned that no geological evidence of an organic soup has ever been found. Coal, oil and natural gas are all considered to be produced from ancient trees or organisms. For a critical evaluation of origin of life hypotheses, the reader is referred to two recent books that deal extensively with this topic [Thaxton et al. (1984) and Shapiro (1986)].

In closing this section, it should be noted that not all of the texts are equally careless in their statements regarding life's origin. Although all of the biology texts give the dear impression that the spontaneous origin of life on the early Earth is very plausible, the degree to which erroneous statements are made in support of that view varies widely.

Neglect of the Central Problem, Genetic Information

Although most of the texts deal with complex biochemical processes quite well in other chapters, none mention the problem of the origin and transfer of genetic information in dealing with origin of life studies. Moreover, the texts fail entirely to note that even if some complicated molecules were formed by chance, all of the machinery required to exactly reproduce these molecules must also be present in order for cells to survive and reproduce. Indeed, Harold Klein, chairman of a National Academy of Sciences committee which recently reviewed origin of life research, notes that the simplest bacterium is so complicated from the point of view of a chemist that it is almost impossible to imagine how it happened (Horgan 1991, p. 120).

Instead, the textbook's origin of life chapters uniformly disregard recent studies related to the complexity of origin of life requirements. Proteins in cells are made up of 20 different L-amino acids. The texts fail to note that unique linear sequences of these L-amino acids are required in protein molecules in order for those proteins to function. These unique amino acid sequences are required whether the protein is an enzyme, a structural component, or is used for some other function. The unique sequence, in turn, is responsible for the three-dimensional structure of the protein, which is also essential to its function. Even though there may be some variability in amino acid sequence in some positions of a protein molecule, calculations with cytochrome c, a protein 104 amino acids long, indicate that the probability of achieving the linear structure of this one protein by chance is 2 x 10-65 (Yockey 1977). Consequently, it is not surprising that the means of assembling such unique sequences during the process of protein synthesis in living cells is extremely complex. The genetic information for these unique linear sequences is initially carried in sequences of nucleotides in DNA of a gene in the nucleus of the cell. From there it is transferred to a nucleotide sequence in messenger RNA (a process called transcription) and from the mRNA to the sequence of amino acids in the final product, a protein molecule (a process called translation). The latter process is so complex that even in the simplest organisms, as many as 200 different protein molecules are required. Altogether, the result of these different processes is an amazingly accurate transfer of information from the nucleotide sequence in DNA to the amino acid sequence in the protein.

In addition, the texts fail to note that most of the more complex biochemical reactions of cells require not only a protein enzyme, they also require an additional component (coenzyme, prosthetic group, etc.). Examples of these groups are heme of various heme proteins and also the different vitamin coenzymes. These groups, which are often complex molecules, may be an integral part of the enzyme molecule (covalently bound), or they may freely dissociate from the protein. In the majority of cases, these organic components are absolutely essential to the catalytic function of the protein molecule. As a consequence, postulated scenarios for the origin of life must provide for the simultaneous formation of the essential coenzyme or prosthetic group and assembly of a specific linear amino acid sequence in the enzyme protein. They must, of course, also provide for the formation of many other complex macromolecules (nucleic acids, carbohydrates, lipids, etc.) that are essential to the function and reproduction of the living cell. The failure to address these requirements shows even more fully the implausibility of the origin of life scenarios presented in the texts.

Of the important problems for origin of life models, Dose (1988, p. 355) discusses the source of genetic information last, closing with a summary of few words: "The difficulties that must be overcome are at present beyond our imagination." In regard to the chance hypothesis for the origin of genetic information, Kuppers (1990, p. 60) notes:

The expectation probability for the nucleotide sequence of a bacterium is thus so slight that not even the entire space of the universe would be enough to make the random synthesis of a bacterial  genome probable.

Compare these statements with the easy confidence noted in the textbooks that a naturalistic explanation of life's origin is soon to be found. It is this confident tone, coupled with what students are not told, that makes origin of life chapters in the texts fall short of the guidelines "examining alternative scientific evidence and ideas to test, modify, verify or refute scientific theories."

Definitions of Evolution

It should be apparent that terms, such as the word "evolution" need to be clearly defined in high school biology textbooks. Such is not the case, however, as the books use the term in several senses without indication that the meaning is changed.  Keith Thomson (1982), professor of biology at Yale University, indicates three commonly employed meanings of evolution:

  1. Change over time
  2. Relationships of organisms by descent through common ancestry
  3. A particular explanatory mechanism for the pattern and process of (1.) and (2.), such as natural selection.

Thomson notes that factual patterns of change over time, particularly as seen in the fossil record, can be studied in the absence of theories of how these patterns came to be. Thomson also emphasizes that the second meaning, descent through common ancestry, is a hypothesis, not a fact, and that it is derived from the twin premises that life arose only once on Earth and that all life proceeds from preexisting life. Cladistic analysis, championed currently by a number of biologists, has sought to eva1uate relationships among organisms without regard to the twin premises cited above. In regard to the third meaning, a particular explanatory mechanism, there are currently many alternative hypotheses. Darwin insisted that changes had to be small and gradual. However, Gould and his associates (1980) have proposed static intervals (stasis), followed by periods of rapid change (punctuated equilibrium). The biology texts, in general, do a poor job of distinguishing between these three different meanings of evolution. They generally fail to note that it is possible to accept the factual evidence for change over time, while having a more restricted view of descent through common ancestry. For example, to speak of ancestral descent in regard to the relationship of an ancestral horse to a modern horse would be a very restricted use when compared to the relationship of an ancestral one-celled organism to a modern mammal. Likewise, accepting the factual evidence for change over time does not require the acceptance of a particular explanatory mechanism for these changes.

On another level, many scientists prefer to differentiate between microevolution and macroevolution: the former being the relatively small changes noted in the diversification of species, and the latter being the changes required in the development of new phyla, or possibly of new orders or classes. The term macroevolution has also been used in regard to development of new functions, such as vision or hearing. Many proponents of Darwinian natural selection have argued that processes demonstrated for microevolution may be extrapolated to account for macroevolution as well. When this type of extrapolation is used in an attempt to validate a theory, we have moved beyond the reasonable bounds of science. Scientifically, we should simply state that at present, there is no satisfactory scientific explanation for macroevolutionary events. Those explanations that have been presented lie in the realm of philosophy.

Arguments for Biological Evolution

When we examine the arguments for biological evolution in the different texts, we find that marked differences exist between them and mainstream medical and biologjcal science texts. The topics of structural homology (six texts), embryology (four texts) and vestigial organs (five texts) are treated with obsolete and erroneous discussions in the high school biology texts.

Structural Homology

All of the high school textbooks confidently offer classic examples of structural homology, such as the similarity of bony structures of the five-digit forelimb in a variety of animals, as evidence of common ancestry. Comments asserting or implying the common embryonic and genetic origin of homologous structures or their common ways of developing appear repeatedly in the discussions. Such an interpretation is clearly out of date and ignores a growing body of scientific data coming from prominent scientists. Sir Gavin de Beer (1971), for example, poses some important questions in his monograph titled Homology, an Unsolved Problem. For example, homologous structures do not necessarily derive from similar positions in the embryo or parts of the egg, nor do they share the same organizer-induction processes, nor are they even necessarily controlled by corresponding genes (de Beer 1971, pp. 13-15). The textbook authors should at least express the fact that this apparent argument for evolution, as attractive as it sounds, is not without very significant questions and problems that remain to be answered. At least one of the textbooks points to Darwin's explanation of homology as the best one. Yet de Beer and others (Goodwin 1982, p. 51; Webster 1984, p. 193) fault Darwin's concept of homology as "just what homology is not." Goodwin (p. 51) also adds that "... homological equivalence is independent of history."  It is clear that there are important questions about the very notion of homology, but there is no suggestion of these questions in the textbooks.

Vestigial Structures

One would think that knowledgeable scientists would be extremely cautious about referring to vestigial structures in view of the fact that dozens of them were once thought to be present, but time and new scientific knowledge have removed almost every one from the list. A vestigial structure can be defined as a part or organ which was well developed in ancestral forms, but the size and structure of which have diminished until it currently has no function. Identification of a genuine vestigial structure requires that the part in question serve no contemporary useful purpose. The textbooks cite the coccyx (four texts), appendix (five texts), muscles that move the ears (three texts), canine tooth root structure (one text), wisdom teeth (one text) and the remnant of the third eyelid (one text) as vestigial organs. Space will not permit us to consider all of these, but two, the coccyx or tailbone, and the appendix, will be examined in some detail to demonstrate the fallacy of the textbook arguments.

The coccyx. It is absolutely clear that the coccyx or tailbone is a functional unit and has been recognized as functional for many years. Examination of Gray's Anatomy (Goss 1948) will serve to indicate that the coccyx is one of four major points of attachment of the support of the perineal floor. Also attached to the coccyx is the coccygeus muscle and a portion of sacrotuberous ligament, thus forming a significant portion of the posterior perineal support and adding stability to the pelvis via the interossical ligaments. The sequence of ossification of the coccygeal segments permits mobility of the coccyx during child- bearing years and thus allows enlargement of the bony outlet of the birth canal during delivery of the baby. However, the expansion of the pelvic floor is minimized in other circumstances.

The appendix. Today there is little doubt that the appendix is involved as a contributor to human immune function. To its credit, one text (Biggs et al. 1991) gives at least a qualified acknowledgment of this role. A recent journal discussion by Bjerke et al. (1986, pp. 672-3) notes the abundant content of organized lymphoid tissue in the appendix. The authors add;

It seems justified to assume that the lymphoid follicles of the appendix are analogous to the Peyer's patches in having the capacity to generate IgA-cell precursors that migrate via lymph and blood to the distant gastrointestinal lamina propia. . . We have found that normal human appendix mucosa contains relatively more IgG-producing cells than the colonic counterpart. This difference can be ascribed to preferential accumulation of IgG immunocytes adjacent to the numerous lymphoid follicles in the appendix.

Kawanishi (1987, p. 19) shares the view of the above authors regarding functionality of the appendix when he writes:

The human appendix, long considered only an accessory rudimentary organ, could possess a similar antigen uptake role prior to replacement by fibrosed tissue after repetitive subclinical infections, or at least in early childhood when it is most prominent.

The appendix is clearly a functional organ in humans and therefore cannot be considered as vestigial.

Sequence Similarities & Ancestral Descent

One of the major points made in many of the texts is that similarities of protein sequence provide strong support for evolution. These similarities are often used to indicate lines of ancestral descent of organisms. Protein sequence similarities can be used to indicate relationships among organisms, but whether these relationships indicate molecular homology (that is, ancestral descent), or whether there may be some other cause of the relationship is not always clear. To give a specific example, rat and mouse cytochrome c molecules are identical in amino acid sequence and the nucleotide sequences in the coding region of the cytochrome c genes differ in nine positions. With these very minor differences, the evidence is strong that a common rodent cytochrome c gene of the past is ancestrally related to the cytochrome c genes found at present in rats and mice. However, if one compares mouse cytochrome c with cytochrome c of yeast (a unicellular eukaryotic organism), there are 37 differences in amino acid sequences of the two proteins and 118 differences in nucleotide sequences of the coding regions of the two genes (Mills 1991). There is clearly similarity between the mouse and yeast genes, but is there an ancestral relationship of the yeast cytochrome c (or more properly of an earlier eukaryotic cytochrome c) to the mouse cytochrome c gene? Scientifically, we must say that there is insufficient evidence at present to make a firm statement. The obstacles to a step by step (i.e. one nucleotide at a time conversion in 100 or more different nucleotide positions of an archetypal gene to form the present mouse gene, are very great, since every intermediate would have to code for a functional cytochrome c molecule. Is this possible? If one postulates only chance conversions as a consequence of mutations, this conversion would seem to be beyond the realm of possibility.  If one postulates that these nucleotide changes are under some type of control, the conversion might be more likely. But what is that control? Is it something built into the nature of molecules, or is it a consequence of an intelligent cause? In seeking an answer to this question, we have moved to the border of science and philosophy, where premises and presuppositions are of primary importance. Honest scientists and students should recognize that there is room for differing opinions at this point. 


The practice of teaching evolution to high school students is not disputed. However, it should be made clear that certain aspects of the theory of evolution are philosophical in nature. Evidence for the origin and evolution of life should be presented fairly and without distortion; but evidence that is not in accord with natural processes as an explanation should be clearly presented as well. When there are gaps or limitations in the data, these must be acknowledged. One of the outstanding biologists of the 19th century, Claude Bernard (1865, p. 40), noted:

...when we have put forward an idea or a theory in science, our object must not be to preserve it by seeking everything that may support it and setting aside everything that may weaken it. On the contrary, we ought to examine with greatest care the facts that would overthrow it. . .

From reading this critique, it should be apparent that the errors, overstatements and omissions that we have noted in these biology texts, all tend to enhance the plausibility of hypotheses that are presented. More importantly, the inclusion of outdated material and erroneous discussions is not trivial. The items noted mislead students and impede their acquisition of critical thinking skills. If we fail to teach students to examine data critically, looking for points both favoring and opposing hypotheses, we are selling our youth short and mortgaging the future of scientific inquiry itself. We concur with the requirement that biology texts examine "alternative scientific evidence and ideas to test, modify, verify or refute scientific theories," but we feel that the Origin of Life and Evolution chapters in most of these Biology I textbooks discussed here fall far short of meeting that requirement.