CHAPTER IX Deciphering Primate Phylogeny from Macromolecular Specificities ’ MORRIS GOODMAN Department of Anatomy, W a y n e Stute University College of Medicine, Detroit, Michigan ABSTRACT A problem in deciphering primate phylogeny, morphological convergence between different evolutionary lines, can be overcome by species comparisons of proteins, macromolecules with specificities closely linked to the genetic code in DNA. Various chemical, electrophoretic, and immunological data on serum and tissue proteins in primates are reviewed with respect to their phylogenetic significance. Much of this data deals with protein specificities in the Hominoidea and depicts a particularly close genetic relationship between man and the African apes. Hominoidea, Cercopithecoidea, Ceboidea, and Lorisoidea are characterized by their proteins as monophyletic or natural taxa, even though the conventional subdivisions within several of these superfamilies are not in complete accord with the protein analyses. The protein evidence supports the conventional grouping of Cercopithecoidea with Hominoidea in the infraorder Catarrhini and the grouping of Catarrhini and Platyrrhini (Ceboidea) i n the suborder Anthropoidea. Lemuroidea and Lorisoidea appear to be closer to one another than to either Tupaioidea or Anthropoidea and closer to the Anthropoidea than to the Tupaioidea. Comparisons of primate DNA’s by Hoyer and coworkers are demonstrating genetic affinities among primates which agree with those deduced from the comparison of protein specificities. Species differences and similarities in the relative amounts of different protein macromolecules reflect the grade relationships of prjmates, but, unlike the comparisons of amino-acid sequences or antigenic specificities, are not reliable indicators of phyletic affinities. Data on the ratios of M(usc1e) to H(eart) type lactate dehydrogenase in a series of primate brains provides a biochemical example of the concept that there are “lower” (primitive) and “higher” (advanced) grades of evolutionary development among the extant primates. Several participants at this conference have shown how the species of Primates can be divided into groups on the basis of locomotor behavior. Classifying these species in such terms focuses attention on a n especially difficult problem in evolutionary systematics : deciding whether resemblances between species in features of functional morphology are due to nearness to a common ancestor or to parallel or convergent evolution. A clear example of convergence is provided by the adaptations for brachiation of spider monkeys and gibbons. In this case the particular resemblances in functional morphology in no way suggest close phylogenetic affinities. However, there are cases in which the taxonomic classification of certain species have been influenced by parallel or convergent resemblances. For instance, among the extant members of the Hominoidea, the large-sized ape species are conventionally placed i n the Pongidae while only m a n is placed in the Hominidae. A key diagnostic AM. J. PHYS. ANTHROP., 26: 255-276. feature from the standpoint of functional morphology for differentiating the two families is the presence of brachiating behavior in the various apes and the absence of such behavior in man. Yet the fossil record suggests that the recent nonhuman hominoids had independently acquired i n evolution their elongate forelimbs and other specializations for brachiation (Simons, ’62). Furthermore, as emphasized at this conference, marked differences i n locomotor behavior and ecological adaptations occur among the various apes. Thus, while the locomotor and other data on functional behavior provide a typological rationale for the respective placement of great apes in Pongidae and m a n in Hominidae, these data certainly do not provide a convincing phylogenetic rationale for this arrangement. Nevertheless, the view has been vigorously advocated in recent discussions of hominoid systematics ~ _ _ _ 1 Supported by N.S.F. grant GB-1241. 255 256 MORRIS GOODMAN that the differences among organisms in their structural-functional adaptations and ecological zones constitute an important basis for taxonomic classification. The literal adoption of this view in taxonomic practice would be a return, it seems to me, to one of the oldest ways of classifying animals, i.e., “according to their mode of life, environmental adaptations, and associations” (Simpson, ’45, p. 3). Such criteria, which favor the creation of polyphyletic taxa, have, according to Simpson, “long been abandoned for purposes of primary zoological classification or formal nomenclature.” However, in subtler form these criteria may have retained a hold on taxonomic practice. Even the authorities who believe it more probable than not that the chimpanzee and gorilla are less removed in descent from man than from orangutan still support the convention discussed above, which represents the chimpanzee and gorilla as closer to the orangutan. It can be reasoned in this case (see, e.g., Simpson, ’63) that the recent apes are termini of conservative lineages, whereas man is the product of a radically divergent lineage. In other words, one can look upon the Hominidae as a young family arising by extensive adaptive divergence from an older and more conservative family, the Pongidae. However, I wonder if this conception of man’s taxonomic status does not have a subjective bias to it. How can one be sure, for instance, that the orangutan’s arboreal ecological niche is less far removed from the chimpanzee’s combined terrestrial-arboreal niche than is man’s terrestrial habitat? Or, to consider social relationships in naturalistic settings, might not it be possible that chimpanzees and gorillas, which as a rule live in large groups, are actually closer in behavior to humans than to orangutans, which have been observed only as solitary travelers or in small family groups of no more than a mother, inCant, and father. Furthermore, the particular functionalmorphological features which indicate that man has radically diverged from other hominoids - such as his unusually large brain, his unrivaled manual dexterity, and his bipedal gait - do not in themselves demonstrate that man’s genotype is markedly different from that of chimpanzee or gorilla. Indeed, it is known that sometimes small genetic changes can cause marked morphological differences between organisms. Similarly, due to the dependence of gene expression on environmental factors, organisms with much the same genotype can, under the influence of different environments exhibit marked phenotypic variations in their morphologies. These problems, along with the opposite problem of convergent morphological similarities,raise formidable difficulties in unraveling the phylogenetic relationships of organisms. Although giant strides toward a phylogenetic classification of the primates have been achieved by the traditional methods of comparative morphology, many questions remain. In addition to the uncertainty about whether the African apes are genetically closer to man or to orangutan, there is uncertainty about the phylogenetic position of the gibbons. Some authorities (e.g., Simpson, ’45; Le Gros Clark, ’59) place the gibbons in a subfamily of the Pongidae. Other authorities (e.g., Fiedler, ’56; Straus, ’49) consider the gibbons a separate family, the Hylobatidae, distinct from Pongidae as well as from Hominidae but still within the Hominoidea. Recent chromosome analyses (Hammerton et al., ’63; Chiarelli, ’65) have even provoked the thought that the gibbons are closer to the Cercopithecoidea than to the Hominoidea. Another question in primate systematics pertains to the ceboids or platyrrhine primates. Although they are always placed in the same suborder as the catarrhine primates, not all authorities assume that this grouping represents the “clade” relationships of the two assemblages of higher primates. It is felt that placing these assemblages together in the Anthropoidea represents their “grade” relationships, however. A common view is that the simian or monkey-like grade, which can be defined by a complex of functional-morphological features, was attained independently in Old and New World continents by parallel or convergent advances in lineages which cladistically were as close to lines leading to modern prosimians as to each other. Such ambiguity about phyletic relationships increases on approaching the lower boundary of the order Primates, with systematists divided on whether or not this PRIMATE MACROMOLECULAR SPECIFICITIES order should include the Tupaioidea (tree shrews). For example, Van Valen (’65) argues from mounting paleontological data on early eutherian mammals for the exclusion of the tupaoids from the order Primates. However, the question is far from settled. It would seem that if we are eventually to gain a detailed, authentic picture of the phylogeny of the primates, there is need for new sources of data as well as for continued exploitation of the old sources. Data which closely reflect the genetics of organisms would be especially useful, for as Simpson (’45, p. 5) pointed out, complete genetic analysis would provide the most priceless data for mapping the stream of heredity which makes phylogeny. Although genetypic similarities and dissimilarities would have to be interpreted phylogenetically in much the same way as phenotypic likenesses and differences, “the advantage of genetics,” Simpson emphasized, “lies rather in the fact that the genes (and some other genetic factors) are the immediate physical continuants of phylogeny, while morphology is less direct, a result of these hereditary factors as modified by other influences.” In 1945, when Simpson set forth these germinal ideas on genes and phylogeny, most scientists were not yet convinced that the material substratum of the genes was deoxyribonucleic acid (DNA). Nor had it yet been proven that the genetic code in DNA determines with extremely high precision the detailed structures of proteins. Thus the potential significance for taxonomy of the precipitin work of Nuttall (’04), Boyden (’42), Wolfe (’39), and others, which demonstrated the “species-specificity” of proteins (Landsteiner, ’47), was not fully grasped. A different situation exists today, based on our growing understanding of the molecular basis of heredity. Efforts to map the genetic relationships of species by suitable serological and biochemical analyses are gaining momentum. In addition, promising methods have recently been developed for directly comparing DNA homologies among species. The purpose of my paper is to discuss the picture of evolution in the order Primates as furnished by the data already gathered on primate proteins and deoxy- 257 ribonucleic acids. When these molecular data agree with the functional-morphological data in grouping certain species together, we can have confidence that such assemblages are indeed phylogenetic ones. But when the molecular data do not agree with the groupings from the functionalmorphological data, we can suspect that the species under comparison represent convergent evolutionary lines stemming from different adaptive radiations. THE PHYLOGENETIC SPECIFICITY OF SEMANTIDES The evidence now seems overwhelming for the model of the molecular geneticists which depicts colinearity in primary structure between genes and proteins (Stretton, ’65; Khorana, ’65). According to this model, the genetic code resides in the sequence of nucleotide bases of long-chained DNA, with each consecutive triplet of bases coding for a correspondingly positioned amino acid in a polypeptide chain. The mechanisms by which the code in a gene is first transmitted into “messenger” RNA and then translated into the sequence of amino-acid residues of a polypeptide chain need not concern us in this discussion. What does concern us is the size and complexity of the genetic message reflected in the polypeptide chain. Typically such a chain or protein subunit as is found in the hemoglobin molecule consists of a sequence of about 150 amino acid residues of up to 20 different kinds. Since the code “words” for the different amino acids are different triplets drawn from the four kinds of nucleotide bases in DNA, the structural gene or section of chromosomal DNA coding the amino acid sequence of the polypeptide chain would consist of about 450 nucleotide residues. It is apparent that the theoretical permutations in structure for such a gene or its corresponding protein product are fantastically large. Thus the tiny fraction of the total coded information in a genome represented by a single structural gene is still of such complexity (and contains such a unique selection of information) that it could well exhibit properties of phylogenetic specificity. Pursuing this point, Zuckerkandl (’65) and others distinguish two classes of organic molecules. One class consists solely 258 MORRIS GOODMAN of nucleic acids and proteins, special substances which are termed semantides to emphasize their richness in coded information. The other class consists of organic molecules such as retinene which are synthesized under the control of proteinenzymes without the direct participation of nucleic acid templates and are relatively low in genetic information. The substances in this class are called episemantic molecules. It may be fairly common for longseparated lineages to produce the same members of this class by evolutionary convergence. Consequently, episemantic molecules are not nearly as useful as the semantides for phylogenetic studies. Given the almost innumerable choices €or mutations (changes in code letters) within even a single genetic locus and the random order in which mutations occur, it is highly unlikely that proteins of the same specificity can be produced by evolutionary convergence. However, during periods of phyletic evolution, any particular protein may have been more conservative (i.e., changed less in specificity) in certain lineages than in others. Thus, if lineage A diverged more from lineage B than from lineage C in terms of a particular protein, it would not necessarily mean that A has a more recent common ancestry with C than with B. On the other hand, this would be a reasonable assumption to make if B consistently showed the most divergence in a series of proteins examined. At the least, it would be fairly safe to conclude that A has more genetical affinities with C than with B. A priori reasoning and the trend of results already observed in studies of primate proteins suggest that when appropriate methods are employed, the data from a relatively small sampling of proteins can produce valid inferences about phylogenetic relationships. COMPARISONS OF PRIMATES FROM CHEMICAL DATA ON PROTEINS The chemical determination of the actual amino acid sequences in proteins can give the most precise description of protein homologies in different species. However, the methods of analysis are painstakingly slow and the data as yet are too meager to have any impact on primate classification. Perhaps the only relevant sequence data are for cytochrome c (Smith and Margoliash, '64) and fibrinopeptide A (Blomback, Doolittle and Blomback, '65). Even though the results were obtained on only a handful of species, they demonstrate that, in terms of cytochrome c and fibrinopeptide A, man and Old World monkey barely diverge from each other while showing sizable divergencies from nonprimate mammals. A broader but less precise set of data on hemoglobin structure in primates has been obtained by fingerprinting various primate hemoglobins and determining their amino acid compositions. In fingerprinting (Ingram, 'SS), peptide fragments with differing amino acid sequences are first produced by controlled tryptic digestion of a protein or protein subunit and are then separated into a complex pattern by chromatographic and electrophoretic procedures. Differences between proteins in their primary structure, i.e., in their amino acid sequences, can be spotted by differences in their fingerprint patterns. Amino acid composition data give the amounts but not the arrangements of the different amino acids in a protein substance. On the whole, the fingerprint and amino acid composition analyses provide minimum estimates of differences in primary structure between proteins. From these analyses, human hemoglobin appears to be almost identical with chimpanzee and gorilla hemoglobins (Zuckerkandl, '63) and to be more similar to orangutan (Zuckerkandl, '63) and gibbon (Hill and Buettner-Janusch, '64; Buettner-Janusch and Hill, '65) hemoglobins than to Old and New World monkey hemoglobins. The correspondence in hemoglobin structure appears to be much greater among these higher primates than among lemurs and other prosimians (Hill and BuettnerJanusch, '64; Buettner-Janusch and Hill, '65). The various chemical data on cytochrome c, fibrinopeptide A, and hemoglobin all suggest that genetic correspondence among higher primates may be relatively high. COMPARISONS O F PRIMATES FROM TWODIMENSIONAL ELECTROPHORETIC SERUM PATTERNS AND TISSUE ZYMOGRAMS By far the most extensive data on primate proteins are those being gathered by PRIMATE MACROMOLECULAR SPECIFICITIES electrophoretic and immunological procedures. A wide series of sera from primates and certain related mammalian types have been surveyed in my laboratory by the twodimensional electrophoretic method of Poulik and Smithies (’58). The proteins in a sample are separated by electrophoresis on filter paper followed by electrophoresis in starch gel. The proteins separate in the second dimension on the basis of molecular size as well as electrophoretic charge, forming a complex pattern of about 19 to 25 components. Such discrete properties as a protein’s electrophoretic charge or its polymer-forming tendencies can be sharply altered by gene mutations, even by a single-point mutation if the amino acid substitution in the polypeptide chain is a “radical” replacement (Zuckerkandl and Pauling, ’65): e.g., a positively charged amino acid for a negatively charged one, or an amino acid with an hydrophobic residue for one with an hydrophilic residue, and so on. It can be deduced that phylogenetic branching need not be extensive for striking differences to appear in the two-dimensional electrophoretic serum patterns of diverging species. The reasonableness of this proposition becomes apparent on considering that there are at least 25 and probably up to 100 or more gene loci which control the amino acid sequences of the different subunits of the proteins in serum. Furthermore the actual two-dimensional electrophoretic data gathered on a wide range of taxa (including 50 species from 31 primate genera) demonstrate marked divergencies in serum patterns between samples from different subfamilies and also in the large majority of cases from different genera within the same subfamily. The major exception is in the Cercopithecinae, a subfamily of Old World monkeys. Here fairly similar patterns are observed between genera (Goodman, ’63b; Goodman, Farris and Poulik, in press). In all groups examined so far, the samples from different species within a genus (such as Galago crassicaudatus and Galago senegalensis, or Pan troglodytes and Pan paniscus) invariably have very similar but usually not identical patterns (Goodman, ’62b, ’63b, ’65; Goodman, Farris and Poulik, in press ) . Thus the two-dimensional electrophoretic serum data can best be used to 259 assess phylogenetic relationships at the lower taxonomic levels. For example, these data (Goodman, ’63b) do not favor the proposal of Mayr (’50) and Simpson (’63) that the taxonomic distinction between gorilla and chimpanzee be only at the species level; instead the serum electrophoretic data suggest that there is sufficient genetic divergence between gorilla and chimpanzee to favor maintaining Gorilla as a separate genus from Pan. On the other hand, the serum patterns of Gol-illa, Pan, and Homo have more features in common with one another than with the patterns of Pongo, Hylobates, or Symphalangus. Conceivably, the conventional Ponginae, which groups Gorilla and Pan with Pongo while excluding Homo, is an artificial taxon not reflecting the true phyletic affinities of its members. This suggestion will be amplified later in the paper when we consider the more definitive immunological data on proteins. The two-dimensional electrophoretic serum patterns provide a small piece of evidence for the hypothesis that within the Cercopithecinae Erythrocebus is closer to Cercopithecus than to other genera ( M a caca, Cercocehus, Theropithecus, and Papio) . Although these various cercopithecine genera have fairly similar patterns, some obvious differences do occur among them, especially in the positioning of the protein components migrating both on filter paper and in starch gel between transferrin and albumin. In this area of the two-dimensional electrophoretic patterns, Erythrocebus least resembles Cercocebus, Papio, and Theropithecus and most resembles Cercopithecus (Goodman, Farris and Poulik, in press). The taxonomic significance of starch-gel (or other comparable) electrophoretic analyses can be enlarged by obtaining tissue patterns of the multiple molecular forms of particular types or classes of enzymes. These patterns are called zymograms (Hunter and Markert, ’57). Using starch gel electrophoresis and the appropriate histochemical reagents for demonstrating five distinct types of esterases, Tashian (’65) has compared the zymograms of erythrocyte esterases in 23 primate species. The patterns were similar in man, gorilla, chimpanzee, and orangutan but were 260 MORRIS GOODMAN sharply different in gibbon. On the other hand, Syner and Goodman (unpublished data), on examining the zymograms of brain esterases in 24 primate species, found that these esterase patterns were quite similar in human, chimpanzee, and gibbon brains (gorilla and orangutan brains were not available). In each study there was little variation among different cercopithecoid species in their zymogram patterns, which differed sharply from hominoid patterns and perhaps even more so from ceboid and prosimian patterns. Within the small limits of variation in erythrocyte esterase patterns among the four cercopithecine genera which were examined (Tashian, '65), Erythrocebus showed more similarity to Cercopithecus than to Papio or Macaca. Among both ceboids and prosimians, in contrast to cercopithecoids, the patterns varied sharply from species to species. The esterases are a broad class of enzymes and each tissue is likely to have its own specific assortment of members of the class. Thus more than a few gene loci are represented by the amino acid sequences of the various esterase components, though probably not as many loci as control the serum proteins. When more tissues and additional classes of enzymes are surveyed in a series of primates, the findings will almost certainly have significance for primate systematics. Even the limited erythrocyte and brain data already gathered add nicely to the serum electrophoretic data. One gains the impression from the sum of this electrophoretic data that the African apes and man are closer to the orangutan than to the gibbon but (taking into consideration the brain data) closer to the gibbon than to Old World monkeys and other primates. However, false impressions can easily be gained from electrophoretic data. In particular, estimations of the magnitude of genetic divergence between organisms from such data are likely to be fallacious. Electrophoretic mobilities of intact macromolecules cannot be used to measure the extent of the differences in amino acid sequence between proteins. All that can be assumed is that the more extensive the variation of amino acid sequences, the greater is the likelihood of detecting some difference between proteins in their electrophoretic mobilities. Genetic polymorphisms of various proteins (hemoglobin, transferrin, haptoglobin, carbonic anhydrase, lactic dehydrogenase, etc.) within species are often detected by electrophoretic analyses. Information obtained from population surveys of these electrophoretically detectable polymorphisms could perhaps be useful in unraveling the genealogical relationships of populations grouped together at quite low taxonomic levels and having only statistically minimal genetic differences between them. Such an approach, which would rely heavily on the techniques of population genetics, could be profitably used, for instance, to evaluate the systematics of the macaques. There is reason to suspect that among macaque groups, currently classified as distinct species, the degrees of speciation are actually not so complete as to have prevented all gene exchange or introgression between these groups in their zones of geographic contact and overlap (Goodman et al., '65). If this were the case, the frequencies of allelic genes in the local populations of different groups would tend to be more similar to one another in the zones of geographic overlap than elsewhere. In such studies the number of genetic loci that could come under inspection could be increased by using immunological procedures to type for the isoantigens on erythrocytes and on certain serum proteins. THE PICTURE OF PRIMATE PHYLOGENY FROM IMMUNOLOGICAL DATA ON PROTEINS To return now to questions of primate systematics which pertain to genera and higher taxonomic categories, the most informative data from primate proteins have been obtained by immunological procedures. Comparative analyses by agar-gel immunoprecipitin techniques are particularly effective. They allow attention to be focused on single proteins and thus permit a more accurate antigenic comparison of these proteins to their homologues in other animal sera. In contrast to an electrophoretic difference, the antigenic divergence between a protein and its homologue in a different species is related to the magnitude of code-word differences in the corresponding genes. A strong antiserum produced PRIMATE MACROMOLECULAR SPECIFICITIES in a vertebrate host to a foreign protein can detect a variety of different surface configurations or antigenic determinants on the foreign protein. Each such determinant would arise from the unique topographical features in a fairly small but still exceedingly intricate cluster of amino-acid residues. Due to the phenomenon of immune tolerance to self-substances, only those stereochemical configurations not found on the host's own circulating proteins would be antigenic. Tolerance would automatically exist to the less intricate configurations, which, being common to all kinds of diverse proteins, are found in almost all organisms. Thus, due to the confluence of two factors - the phenomenon of immune tolerance and the unrivaled architectural complexity of protein molecules - the proteins exhibit an exquisite degree of specificity at the antigenic level. The field of this specificity can be either broadened or sharpened in phylogenetic comparisons of proteins by the choice of antibody producer. The more closely related the host or antibody producer is to the donor of the immunizing proteins, the greater will be the similarity in the specificities of host and donor proteins and the more selective will be the antibody response of the host. Only the more species-specific (more recently evolved) configurations on the donor proteins will be antigenic to the host. For example, Old World monkey antisera to human proteins, being directed to the more recently evolved configuratons on the human proteins, would be able to distinguish sharply the sequence of relationships of man to other members of the Hominoidea but would be useless for studying the relationships of man to more distant species. On the other hand, antisera produced in rabbits or chickens, while not likely to draw as sharp distinctions within the Hominoidea as the monkey-produced antisera, could nevertheless provide valid information about the phylogenetic relationships of man to species throughout the order Primates and possibly even throughout the class Mammalia in the case of chicken antisera. Here it may be pointed out that only the precipitins to proteins are likely to provide such valid information. Other kinds of substances, e.g., the carbohydrates of the 26 1 blood-group isoantigens, often have heterogenetic properties. Such substances, usually made up in each case of only a few repeating units, are not nearly as complex in chemical structure as the proteins yet have an erratic appearance in nature. Although typically a heterogenetic substance is synthesized by organisms in diverse species of animals, plants, and bacteria, not all species or all organisms in a species synthesize the substance. Thus, unlike the case of the common (less intricate) configurations of proteins which are ubiquitous in organisms, immune tolerance would not be universal to a heterogenetic substance despite its relatively simple configurations. Consequently, the cross-reactions of antibodies to heterogenetic antigens could furnish misleading information about phylogenetic relationships, especially at the higher taxonomic levels. On the other hand, as indicated earlier, surveys of the distribution of such antigens could contribute to studies concerned with the divergencies among population groups at the lower taxonomic levels. As may be surmised, the proteins, with their particular properties as antigens, either lack heterogenetic properties or possess them to a weak degree. Thus, to reiterate an important point, the immunological comparisons that can furnish the most reliable information on phylogenetic relationships are those carried out on proteins. Of the immunoprecipitin data on primate proteins, the most extensive set has been gained with chicken, rabbit, and monkey antisera to the proteins of human serum. Many of the antiserawere produced to purified protein fractions (albumin, gamma globulin, transferrin, ceruloplasmin, etc.) and the rest to whole serum. In my published (Goodman, '62a,b, '63a,b, '64, '65) and as yet unpublished studies, a minimum of 1 4 different human serum proteins have been compared (some more intensively than others) to their homologues in various primate sera. With respect to at least 9 to 13 of these proteins, chimpanzee and gorilla showed less divergence from man than the orangutan and gibbons, and in no case did the chimpanzee and gorilla show more divergence than the Asiatic apes. Ever since the classic study of Nuttall ('04) this same 262 MORRIS GOODMAN trend of results has been found in the species comparisons with antihuman serum protein precipitins (Boyden, '58; Picard, Heremans and Vandebroek, '63; Williams, '65). As shown in figures 1 and 2, the cross-reactions of an Old World monkey antiserum to whole human serum very Fig. 1 Agar-gel immunoelectrophoresis on glass slides: the precipitin lines were developed by an Old World monkey (Cercopithecus aethiops) antiserum (M-68) to human serum. Note that human and chimpanzee sera gave much stronger precipitin reactions (number and intensity of lines) than orangutan and gibbon sera. clearly pictures man as being closer to the African apes than to the Asiatic apes. The immunological data which tend to demonstrate this relationship now include results obtained with rabbit-produced antiserum to human lense extract (Maisel and Goodman, '65; Maisel, '65) and other rabbit- and chicken-produced antisera to human thyroglobulin, the major protein found in the thyroid gland. The former antiserum detected a lens-specific protein whjch was not as conservative in evolution as the typical lens antigens. While its chimpanzee homologue was indistinguishable from it, its gibbon homologue diverged slightly. The antihuman thyroglobulin precipitins detected a small divergence of chimpanzee thyroglobulin and a further, more marked, divergence of gibbon thyroglobulin. Unfortunately the hominoid comparisons were incomplete since gorilla and orangutan lenses and thyroids were not available. In the additional comparisons carried out with the various antisera to human serum and tissue proteins, crossreactions decreased on going from gibbons to cercopithecoids to ceboids to prosimians, the decrease being steepest on leaving the Anthropoidea. A sizable amount of data has also been obtained with antisera to serum proteins and, in a few cases, lens proteins on nonhuman primate species. However, not as many antisera have been produced to proteins of each nonhuman primate as were produced to human proteins. Nevertheless, this additional precipitin data complement very nicely the antihuman precipitin data. So far no support is given the taxonomic convention of classifying man as the sole living representative of the Hominidae for chimpanzee, man, and gorilla barely diverge from one another in their serum antigens while clearly diverging from the orangutan and gibbons. The latter diverge from each other and from man, chimpanzee, and gorilla but show somewhat larger divergencies from Old World monkeys. Recent data with rabbit precipitins to Hylobates serum proteins demonstrate apparent identity in serum antigens of Hylobates and Symphalangus and some divergence of these gibbons from other hominoids. To amplify my earlier proposal (Goodman, '63a,b) : a serologically based Fig. 2 Comparative analysis (Ouchterlony) reactions of hominoid sera in agar gel developed by the monkey antiserum (M-68) to human serum. Man ( M ) formed small spurs against chimpanzee ( C ) and strong spurs against orangutan ( 0 ) and gibbon ( G i ) . Chimpanzee formed only a slight spur against gorilla ( G ) but formed moderate to strong spurs against orangutan and gibbon. In this type of testing (Goodman, '62a), a protein forms a spur against its homologue in another serum if it contains reactive antigenic sites that its homologue lacks. Thus the divergence from the human proteins was greater for the orangutan and gibbon homologues of these proteins than for the chimpanzee and gorilla homologues. 264 MORRIS GOODMAN classification would still place Pun, Homo, and Gorilla in the Hominidae; Pongo could remain in the Pongidae; and Hylobates and Symphalangus would be placed in the Hylobatidae. The three families would still be grouped together in the Hominoidea. The Old World monkeys (Cercopithecoidea) were found to be more similar in their serum and lens proteins to the hominoids than to New World monkeys and marmosets (Ceboidea) and the latter more similar to the Hominoidea and Cercopithecoidea than to any prosimian or comparable mammalian group (tarsiers, lorises, lemurs, tree shrews, etc.). These findings support the primate classifications (Fiedler, '56; Straus, '49; Hill, '57) which subdivide the suborder Anthropoidea into infraorders Platyrrhini (Ceboidea) and Catarrhini (Cercopithecoidea and Hominoidea). At the same time the protein findings are strong evidence that a true phyletic l(i.e., cladistic) relationship, not just a grade relationship, exists between the platyrrhines and catarrhines. The Cercopithecoidea, Ceboidea, and Lorisoidea, like the Hominoidea, appear from the precipitin data as true monophyletic assemblages or natural taxa. The subdivision of the Cercopithecoidea into Cercopithecinae and Colobinae is also in good agreement with the serum precipitin data. The three colobine genera examined (Presbytis, Colobus, and Nasalis) showed a high correspondence to one another in serum antigens while diverging from cercopithecines (Erythocebus, Cercopithecus, Cercocebus, Macaca, Papio, and Theropithecus). The latter in turn showed high antigenic correspondence among themselves while diverging from the colobines. Within the Cercopithecinae itself the clearest antigenic divergence - surprisingly, of serum albumin - was between the patas monkey ( E q t h r o c e b u s ) and the baboons (Papio and Theropithecus). Thus this serological analysis agrees with aspects of the functional-morphological analysis (Napier, '66) which suggest that the adaptations of the patas monkey and baboons for terrestrial existence were produced by parallel or convergent evolution. But at the same time the finding of serological divergence between these ground-living cercopithecines helps demonstrate that similarities in ecological adaptations and functional behavior are not necessarily indicative of genotypic similarities. Within the Ceboidea the subdivision of genera into subfamilies in the conventional taxonomic schemes has been largely an arbitrary matter in that the morphological differences between genera within a subfamily are often as marked as the differences between genera in different subfamilies. However, there is fairly strong evidence for the subfamily grouping of Ateles (spider monkey) and Lagothrix (woolly monkey) in Atelinae. These forms, which are noted for the abilities to brachiate and €or their prehensile tails, show a relatively high degree of similarity in functional morphology. Furthermore, the cross-reactions of rabbit precipitins to spider and woolly monkey albumins (Goodman, ' 6 5 ) and gamma globulins (figs. 3,4) demonstrate that Ateles and Lagothrix correspond more to one another than to other ceboids. The cross-reaction data also suggest that Ateles and Lagothrix are phyletically more closely related to the howler monkey Alouatta (another form with a prehensile tail and ability to brachiate) than to other ceboids. On the other hand, data obtained with anticapuchin and antisquirrel-monkey precipitins disagree with classifications (Simpson, '45; Hill, '60) which group Cebus and Saimiri in the subfamily Cebinae, since these two genera diverged as much from each other as from other ceboid genera. Nor is the grouping of Aotes and Callicebus in the Aotinae (Simpson, '45) supported by the cross-reactions of precipitins to night and titi monkey sera. These and other data suggest that the differences between genera in serum proteins are greater within the Ceboidea than within either the Cercopithecoidea or Hominoidea. Turning now to the Lorisoidea, the serum protein precipitin data (examples of which are given in figs. 5,6,7,8) show each genus of the four examined diverging from the others, with slightly less divergence between Nycticebus and Galago than between Nycticebus and Perodicticus. On the other hand, the divergence between Nycticebus and Galago may be slightly greater than the divergence between Nycticebus and Loris. These results would not appose the conventional grouping in the Lorisinae PRIMATE MACROMOLECULAR SPECIFICITIES 265 Fig. 3 Comparative analysis (Ouchterlony) reactions developed by anti-gamma globulin precipitins of chicken antiserum (C-939) to spider monkey serum. The test antigens were gamma globulins isolated (by paper electrophoresis and elution) from the sera of the following primates: spider monkey (Sp), woolly monkey ( W ) , howler monkey (H), night monkey ( N ) , squirrel monkey (Sq), man ( M ) , and slow loris (SI). With the exception of the slow loris and man, the other animals were all ceboids. Spider monkey formed a trace spur (not evident in the photograph) against woolly monkey, a moderate spur against howler monkey, and strong spurs against night and squirrel monkeys. In turn, squirrel monkey formed a medium to strong spur against man, and man a weak spur against slow loris. of Nycticebus and Loris but would appose that of Perodicticus, since Galago is conventionally excluded. Conceivably, parallel evolution may have played a n important part in bringing about the marked resemblance between Perodicticus and Nycticebus in their locomotor adaptations. The present serological results, while not yet extensive enough to encourage any categorical claims do suggest that the phyletic distance be- 266 MORRIS GOODMAN Fig. 4 Comparative analysis (Ouchterlony) reactions of gamma globulins developed by anti-gamma globulin precipitins of chicken antiserum (C-946) to woolly monkey serum. Woolly monkey ( W ) merged with spider monkey (Sp), formed a trace spur (perhaps not evident in the photograph) against howler monkey ( H ) , and strong spurs against night ( N ) and squirrel monkeys ( S q ) . Squirrel monkey formed a strong spur against man ( M ) , and man a moderate spur against potto (Po). tween Perodicticus and Nycticebus is every bit as great as the distance between Galago and Nycticebus. The serological results also apply to the problem of the systematic relationships of prosimian groups at the higher taxonomic levels. While Lorisoidea and Lemuroidea diverge markedly in serum antigens, each nevertheless appears to be closer to the other than to Tupaioidea, Tarsioidea, or Anthropoidea. This would support the removal of the tree shrews from the Lemuriformes and the grouping of Lemuroidea with Lorisoidea in the Strepsirhini (Hill '53). So far the immunoprecipitin data tend to affirm that the Anthropoidea is not quite as f a r from Strepsirhini as from Tupaioidea. For example, precipitins (Goodman, '63b) to human gamma globulin, ceruloplasmin, and alpha, macroglobulin, as well PRIMATE MACROMOLECULAR SPECIFICITIES as precipitins (Maisel and Goodman, '65; Maisel, '65 j to the least conservative fraction of the higher primate lens proteins, developed larger cross-reactions with lorisoids and lemuroids (lens could not be tested in this case) than with Tupaia and nonprimates. On the other hand, tupaiid albumin resembles higher primate albumins slightly more than do the strepsirhine albumins and distinctly more than do nonprimate mammalian albumins. However, as discussed elsewhere (Goodman, '63a j , a marked evolutionary conservatism of both tupaiid and higher primate albumins could account for this resemblance. In the initial study with antisera to galago lens proteins ( Maisel, '65) lorisoids 267 corresponded much more to the Anthropoidea than to tree shrew and nonprimate mammals. However, the lens results neither distinguished the different lorisoids (Galago, Perodicticus, and Nycticebus ) from one another, nor the lorisoids from the tarsier. The latter result (which requires confirmation) is especially interesting since the Lorisoidea and Tarsioidea diverge very markedly from each other in their serum proteins. Conceivably, since the lorisoids and tarsier have remained in nocturnal arboreal habitats, they may have retained lens proteins having the ancestral specificities of the early Tertiary primate lens. In contrast, certain lens proteins in the Anthropoidea clearly changed in speci- Fig. 5 Comparative analysis (Ouchterlony) reactions of lorisoid sera developed by a rabbit antiserum (R-85) to slow loris serum. Slow loris ( S l ) developed weak spurs against slender loris (Sle); slender loris (Sle) developed a weak spur against galago (Ga), weak to moderate spurs against potto (Po), and strong spurs against lemur (Le). 268 MORRIS GOODMAN Fig. 6 Comparative analysis (Ouchterlony) reactions developed by anti-albumin precipitins of the rabbit antiserum (R-85) to slow loris serum. The test antigens were albumins isolated (by paper electrophoresis and elution) from the sera of the following animals: slow loris (Sl), potto (Po), galago (Ga), lemur (Le), chimpanzee (C), and the tree shrew (Tr). Potto albumin diverged much more from slow loris albumin than did galago albumin, but did not diverge as much as lemur albumin. In turn, lemur formed a spur against chimpanzee and chimpanzee against tree shrew. ficity during the evolution of higher primates (Maisel and Goodman, '65; Maisel, '65). These results give credence from the standpoint of molecular affinities to Simpson's classification (Simpson, '45), which constructs two major grades of evolutionary development in the Primates and places the tarsier along with lorisoids and lemuroids in the lower grade, Prosimii. However, the serum data suggest that a determination at the molecular level of over-all genetic similarities and differences among extant primates may not necessarily support such a grade classification. GENETIC CORRESPONDENCE FROM DNA DATA A promising molecular approach for determining over-all genetic similarities and differences among organisms has been de- PRIMATE MACROMOLECULAR SPECIFICITIES veloped by Bolton and McCarthy ( ’ 6 2 ) ; (McCarthy and Bolton, ’63) and Hoyer (Hoyer, McCarthy and Bolton, ’64). Their procedure takes advantage of the tendency to helical formation between the complementary strands of DNA in a genome and measures the specific binding of radioactively tagged DNA fragments to long single-stranded DNA embedded in agar. The extent of “hybridization” obtained in this system between the DNA’s of different species can be considered a function of the genetic correspondence between the species. Hoyer and co-workers are applying their DNA method to primate materials. So far the DNA results (Hoyer et al., ’65; Hoyer, ’65) show extensive genetic similarities among higher primates and agree with the protein data which would subdivide the Anthropoidea into Platyrrhini and Catarrhini with man being closer to apes than 269 to Old World monkeys. In further agreement with the protein data, gibbon DNA is not quite as similar to human DNA as is chimpanzee DNA. In a series of comparisons which demonstrate the latter point, the homologous DNA was human and the percentages of correspondence of heterologous DNA’s to the human calculated in the test system were as follows: chimpanzee, 100; gibbon, 94; rhesus monkey, 88; capuchin monkey, 83; tarsier, 65; slow loris, 58; galago, 58; lemur, 48; tree shrew, 28; hedgehog, 21; mouse, 19; and chicken, 10 per cent (Hoyer, ’65). Such data whet one’s appetite to apply this approach much more extensively to primate and other mammalian materials. One would want to know the sequence of relationships to the tree shrew, to the tarsier, and so on, using a broad series of homologous and heterologous DNA’s. Undoubtedly the findings Fig. 7 Immunoelectrophoretic patterns developed by a chicken antiserum (C-858) to slow loris by the procedure of Weiner et al. (’64). Electrophoresis of the serum aliquot was carried out on a n agar-covered glass strip. The strip, after being placed in the bottom of a glass chamber was overlaid with agar, and the antiserum was then placed on top of the agar. Precipitin lines (the white ones) formed as the antigens diffusing up from the bottom and the antibodies diffusing down from the top met in the agar. The degree of cross-reaction (number and intensity of precipitin lines) decreased on going from slow loris ( S l ) to galago (Ga) to potto ( P o ) to the lemuroid Propithecus (Pr ) . 270 MORRIS GOODMAN Fig. 8 Comparative analysis (Ouchterlony) reactions of albumins developed by anti-albumin precipitins of chicken antiserum (C-796) to potto serum. Potto (Po) formed spurs against slow loris (Sl) and galago (Ga), and galago formed a spur against lemur (Le). In turn, lemur farmed a spur against baboon (B). would help solve many of the questions concerning the systematics of the Primates. ilar organisms are in the specificities of their semantides, the closer their proximity to a common ancestor. Comparisons by DIFFERENTIAL GENE EXPRESSION AND appropriate methods of either the proteins THE GRADE RELATIONSHIPS or deoxyribonucleic acids should yield the OF PRIMATES same sequence of probable phyletic relaThroughout this paper I have implied tionships. Furthermore, due to the inherthat protein and DNA data of the kinds dis- ent phylogenetic specificity of even single cussed can more consistently reflect the structural genes, the comparisons should genealogical relationships of organisms not have to involve the specificities of than can the data of comparative morph- very many proteins or structural genes to ology. My confidence in the molecular ap- reflect the same sequence of relationships proach rests on the deduction, made from as would be revealed by the complete speciconsiderations concerning the nature of ficities of the total genomes. On the other the genetic code, that evolutionary con- hand, there are numerous examples demonvergence in the detailed structural speci- strating that similar morphological features ficities of proteins is extremely unlikely. have been produced in widely separated The corollary follows that the more sim- phyletic lines by evolutionary convergence. PRIMATE MACROMOLECULAR SPECIFICITIES A morphological structure, of course, is shaped by the quantities, relative proportions, and spatial positions as well as by the intrinsic properties of the different macromolecules making up its substance. Thus it is not difficult to see that even if two organisms had many amino-acidsequence differences in their proteins, or code-word differences in their genes, a convergence in their locomotor or other functional adaptations could readily be achieved by the independent emergence of morphological structures containing similar quantities and spatial arrangements of the same sorts of macromolecules, thereby showing similar macroscopic properties. Of course, this situation would probably require a similarity between the two organisms in genetic regulatory mechanisms. But still there is no reason to suppose that a convergence in morphological features would increase the convergence between the organisms in the code words of either their “regulator” genes of their “structural” genes. Indeed, evolution, whether convergent or divergent at the morphological level, should still increase divergence of genetic code words at the molecular level. Organisms are said to show a “grade” relationship to one another if they possess a complex of similar morphological features irrespective of whether these features were produced by evolutionary convergence or inheritance from a common ancestor. At the molecular level, similarities in the differential expression of structural genes could preserve or create morphological resemblances even in widely separated phyletic lines and thus be responsible for a grade relationship between these lines. The concept of grades is also implicit in the ingrained view that there are “lower” and “higher” organisms, of which the highest organism is man. Although one should be wary of any tendency to pass anthropomorphic judgments on an animal’s status in nature, it is perfectly reasonable to believe that phyletic evolution has increased the complexity of organisms in certain lineages. Differences between organisms in their degree of complexity might be defined in the following genetic terms. Compared to a less complex - i.e., a ‘lower” or ‘primitive” organism - the “higher” or “advanced” organism would 271 have a more extensive genetic code and a more elaborate integration of gene actions during ontogeny. Several grades of Primates have been distinguished at the morphological level largely in relation to the state of evolutionary development of the central nervous system, with the neocortex being much more developed in the ‘%higher’ primates than in the ‘lower.” Since the marked expansion and increased function of the higher primate brain probably has a basis in the differential activity of particular genes and in the corresponding synthesis of particular macromolecules, the grade concept as applied to the Primates should be evident at the molecular level as well as at the morphological level. This point may be illustrated by data obtained on lactate dehydrogenase isozymes in lower and higher primate brains (Syner and Goodman, ’66; Goodman et al., ’66). The enzyme lactate dehydrogenase (LDH) was chosen for study due to its relevance to energy metabolism (an important parameter in brain function) and because much is known about the significance of the molecular forms or isozymes of this enzyme. Two basically different types of LDH and three hybrids of these types are found in many vertebrate tissues (Appella and Markert, ’61; Cahn et al., ’62). One type is a tetramer of a subunit termed A ( A p pella and Markert, ‘61) or, alternatively, M (Cahn et al., ’ 6 2 ) because this type predominates in adult skeletal muscle. The other type is a tetramer of a subunit termed B (Appella and Markert, ’61) or, alternatively, H (Cahn et al., ’62) since it predominates in adult heart. These two distinct subunits are encoded at separate genetic loci (Shaw and Barto, ’63) and have different electrophoretic mobilities, amino acid compositions, and antigenic structures (Vessel, ’65a; Kaplan, ’65). Although both LDH types catalyze the interconversion o f pyruvate and lactate, they have different sensitivities under certain in vitro conditions to inhibition by high pyruvate concentrations, with the heart enzyme (H4) being far more prone than the muscle enzyme (M1)to inhibition. The hybrids of these two enzymes (MIHs, M2H2, and M,HI) have intermediate catalytic behaviors which are proportionate to their 272 MORRIS GOODMAN H :M ratios (Dawson, Goodfriend and K a p ence of LDH isozymes. These theories may Ian,'64). Kaplan ('65) argues from a con- actually complement each other since the siderable body of evidence that M LDH nucleus lacks mitochondria (on which the suits conditions such as found in skeletal muscle which favor anaerobic glycolysis and sporadic energy production whereas H LDH suits conditions such as found in heart which favor respiration and sustained efficient energy production. This theory gives an attractive explanation for the finding (Markert, '63) that many more tissues have higher proportions of M to H LDH in the mammalian embryo than in the mammalian adult. The embryonic tissues, of course, are better capable of withstanding anaerobic conditions. Another piece of circumstantial evidence for Kaplan's theory is that the lamprey, one of the lowest vertebrates, has LDH of the M type only in all of its tissues. This correlates with the lamprey's relatively anaerobic environment and has possible significance with respect to the concept of grades of evolutionary development, since anaerobic metabolism is phylogenetically more ancient than respiration (Oparin, '57). Given the many facets of biological phenomena, it would be surprising if all the LDH data conformed to Kaplan's theory. In fact there are certain findings (Vessel, '65b) which do not readily conform, such as the reported lack of any difference between human M and H LDH with respect to pyruvate inhibition when the analyses were performed at 37". Vessel ('65a,c) believes that the main significance of the LDH isozymes has to do with differences in their intracellular locations, and he claims that LDH-5, the electrophoretic component which is the pure A or M tetramer, is localized predominantly in the cell nucleus. It should be pointed out that the two theories are concerned with different aspects of the problem poised by the exist- respiratory enzymes are organized) and probably has a lower oxygen tension than other cellular regions due to its deeper distance from the cell surface. I am still sufficiently impressed with the evidence for Kaplan's theory to use it to interpret in a provisional way the data obtained on M and H LDH in primate brains. These brain data, summarized in table 1, demonstrate a drastic change from a preponderance of M to a preponderance of H on ascending the phylogenetic scale of Primates, going from lorisoid to tarsier to ceboid to catarrhine brains. Of further interest are the LDH differences between particular brain regions in lower and higher primates. The relative amounts of M and H LDH in dissected regions of six lorisoid (three galagos, three slow lorises), six ceboid (three squirrel monkeys, one capuchin, one night monkey, one marmoset), and four catarrhine (three macaques, one human) brains were determined by zymogram and reflectance-densitometric analysis techniques (Goodman et al., '66). To briefly summarize the results: in the primitive primates (the Lorisoidea) three to four times more M than H was found in most cerebral cortical regions, whereas oneto-one amounts occurred in brain stem (pons and medulla) and spinal cord. In more advanced primates (the Ceboidea) the H/M ratios showed a marked increase in all central nervous system (CNS) regions, especially in the visual neocortical region represented by occipital pole and calcarine fissure. Three to four times more H than M was found in this visual region, a higher proportion of H to M than found in any other ceboid CNS region. The substitution of H for M was carried further in TABLE 1 Percentage values of M LDH and H LDH in whok brains of primates Superfamily Lorisoidea Tarsioidea Ceboidea Cercopithecoidea Hominoidea No. of species No. of animals M H 5 1 4 6 3 17 2 6 9 7 72.3 43.9 32.9 28.3 29.6 27.7 56.1 67.1 71.7 70.4 PRIMATE MACROMOLECULAR SPECIFICITIES the most advanced primates (the Catarrhini), for in addition to occipital pole and calcarine fissure, other neocortical areas, such as precentral gyrus, also had three times more H than M. Thus we may deduce that as an integral part of the extensive elaboration of the neocortex in the evolutionary development of higher primates, a shift from synthesis of M to H LDH occurred. In terms of Kaplan’s theory, which associates M LDH with sporadic energy production and H LDH with sustained, efficient energy production, the shift from M to H LDH suggests that the advanced primate neocortex operates continuously with a much richer supply of energy than does the primitive primate neocortex. To pursue this speculation further, it would seem that at the ceboid grade of higher primate development, biochemical evolution to a more efficent form of energy metabolism had been especially pronounced in the neocortical region concerned with the analysis of visual sensations. Such biochemical evolution, together with the marked morphological expansion of the visual neocortex, would constitute an important adaptation for arboreal existence. The data on the proportions of H and M LDH strongly indicate that the grade concept as applied to primate evolution is meaningful at the molecular level as well as at the morphological level. However, it must be emphasized that while such data on the proportions of macromolecules have a n evolutionary significance, they are not reliable indicators of phyletic affinities. For example, hedgehogs and soricoid shrews have rich amounts of M LDH in their brains, though not as much as lorisoids have. In terms of a grade classification based on these LDH result, it might be proposed that the Lorisoidea shows a closer relationship to certain groups in the Insectivora than to the Anthropoidea. Obviously such a classification would not reflect clade (phyletic) relationships. On the other hand, if the amino acid sequences or the antigenic specificities of either or both lactate dehydrogenases were compared in these various taxa, a closer relationship of the Lorisoidea to the Anthropoidea would be, I suspect, immediately apparent. 273 LITERATURE CITED Appella, E., and C. L. Markert 1961 Dissociation of lactate dehydrogenase into subunits with guanidine hydrochloride. Biochem. Biophys. Res. Commun., 6: 171-176. Blomback, B., R. F. Doolittle and M. Blomback 1965 Fibrinogen: structure and function. In: Protides of the Biological Fluids. Edited by H. Peeters. Elsevier, Amsterdam, 12: 87-94. Bolton, E. T., and B. J. McCarthy 1962 A general method for the isolation of RNA complementary to DNA. Proc. Natl. Acad. Sci., U.S., 48: 1390-1397. Boyden, A. 1942 Systematic serology: a critical appreciation. Physiol. Zool., 15: 109-145. 1958 Comparative serology: aims, methods and results. In: Serological and Biochemical Comparisons of Proteins. Edited by W. Cole. Rutgers University Press, New Brunswick, pp. 3-24. Buettner-Janusch, J., and R. L. Hill 1965 Molecules and monkeys. Science, 147: 836-842. Cahn, R. D., N. 0. Kaplan, L. Levine and E. Zwilling 1962 Nature and development of lactic dehydrogenases. Science, 136: 962-969. Chiarelli, B. 1965 Personal communication. Dawson, D. M., T. L. Goodfriend and N. 0. Kaplan 1964 Lactic dehydrogenases: functions of the two types. Science, 143: 929-933. Fiedler, W. 1956 ifbersicht uber das System der Primates. I n : Primatologia I. Systematik Phylogenie Ontogenie. Edited by H. Hofer, A. H. Schultz, and D. Stark. S. Karger, Basel, pp. 1-266. Goodman, M. 1962a Evolution of the immunologic species specificity of human serum proteins. Human Biol., 34: 104-150. 196213 Immunochemistry of the primates and primate evolution. Ann. N. Y. Acad. Sci., 102: 219-234. 1963a Man’s place in the phylogeny of the primates as reflected in serum proteins. In: Classification and Human Evolution. Edited by S. L. Washburn. Aldine Press, Chicago, pp. 203-234. 196313 Serological analysis of the systematics of recent hominoids. Human Biol., 35: 377-436. 1964 Problems of primate systematics attacked by the serologica1 study of proteins. In: Taxonomic Biochemistry and Serology. Edited by Charles A. Leone. Ronald Press, New York, pp. 4 6 7 4 8 6 . 1965 The specificity of proteins and the process of primate evolution. In: Protides of the Biological Fluids. Edited by H. Peeters. Elsevier, Amsterdam, 12: 70-86. Goodman, M., A. Kulkarni, E. Poulik and E. Reklys 1965 Species and geographic differences in the transferrin polymorphism of macaques. Science, 147: 884-886. Goodman, M., W. Farris and E. Poulik Immunodiffusion and electrophoretic investigation of the systematics and evolutionary genetics of old world monkeys with particular reference to baboons. In: The Baboon in Medical Research Research-11, University of Texas Press, Austin. ( I n press.) 274 MORRIS GOODMAN Goodman, M., F. N. Syner, C. W. Stimson and J. J. Rankin 1966 Differential levels of the two kinds of lactate dehydrogenase in primate brain regions. In preparation. Hammerton, J. L., H. P. Klinger, D. E. Mutton and E. M. Lang 1963 The somatic chromosomes of the Hominoidea. Cytogenetics, 2: 240-263. Hill, R. L., and J. Buettner-Janusch 1964 Evolution of hemoglobin. Fed. Proc., 23: 1236-1242. Hill, W. C. 0. 1953 Primates I. Strepsirhini. Edinburgh University Press, Edinburgh. Primates 111. Pithecoidea. Edinburgh University Press, Edinburgh. 1960 Primates IV. Cibidae. Part A. Edinburgh University Press, Edinburgh. Hoyer, B. H. 1965 Personal communication of unpublished data. Hoyer, B. H., E. T. Bolton, B. J. McCarthy and R. B. Roberts 1965 Evolution of polynucleotides. In: Evolving Genes and Proteins. Edited by V. Bryson and H. J. Vogel. Academic Press, New York. Hoyer, B. H., €3. J. McCarthy and E. T. Bolton 1964 A molecular approach in the systematics of higher organisms. Science, 144: 959967. Hunter, R. L., and C. L. Markert 1957 Histochemical demonstration of enzymes separated by zone electrophoresis on starch gels. Science, 125: 1294-1295. Ingram, V. M. 1958 Abnormal human haemoglobins. I. The comparison of normal human and sickle cell haemoglobins by “finger printing.’’ Biochemica et Biophysica acta, 28: 539545. Kaplan, N. 0. 1965 Evolution of dehydrogenases. I n : Evolving Genes and Proteins. V. Bryson and H. J. Vogel (eds.). New York, Academic Press. Khorana, H. G. 1965 Polynucleotide synthesis and the genetic code. Fed. Proc., 24: 1473-1487. Landsteiner, K. 1947 The Specificity of Serological Reactions. Harvard University Press, Cambridge, Mass. Le Gros Clark, W. E. 1959 The Antecedents of Man. Edinburgh University Press, Edinburgh. Maisel, H. 1965 Phylogenetic properties of primate lens antigens. In: Protides of the Biological Fluids. Edited by H. Peeters. Elsevier, Amsterdam, 12: 146-148. Maisel, H., and M. Goodman 1965 The ontogeny and specificity of human lens proteins. Invest. Ophth., 4: 129-137. Markert, C. L. 1963 Epigenetic control of specific protein synthesis in differentiating cells. In: Cytodifferentiation and Macromolecular Synthesis. M. Locke (ed.). New York, Academic Press. M a a , E. 1950 Taxonomic cateeories in fossil hominoids. Cold Spring Harbo; Symp. Quant. Biol., 25: 109-118. McCarthy, B. J., and E. T. Bolton 1963 A n approach to the measurement of genetic relatedness among organisms. Proc. Natl. Acad. Sci., US., 50: 156-164. Napier, J. R. 1966 The strucuture and function of the primate hand in relation to locomotion. Manuscript circulated among participants at the Primate Locomotion Symposium. Nuttall, G. H. F. 1904 Blood Immunity and Blood Relationship. Cambridge University Press, Cambridge, England. Oparin, A. I. 1957 The Origin of Life on the Earth. Academic Press, New York. Picard, T., J. Heremans and G. Vandebroek 1963 Serum proteins found in primates. Comparative analyses of the antigenic structure of several proteins. Mammalia, 27: 285-299. Poulik, M. D., and 0. Smithies 1958 Comparison and combination of the starch-gel and filterpaper electrophoretic methods applied to human sera : two-dimensional electrophoresis. Biochem. J., 68: 636643. Shaw, C. R., and E. Barto 1963 Genetic evidence for the subunit structure of lactate dehydrogenase isozymes. Proc. Nat. Acad. Sci. (U.S.A.), 50: 211-214. Simons, E. L. 1962 Fossil evidence relating to the early evolution of primate behavior. Ann. N. Y. Acad. Sci., 102: 282-294. Simpson, G. G. 1945 The principles of classif?cation and a classification of mammals. Bull. Amer. Mus. Nat. Hist., 85: 1-350. 1963 The meaning of taxonomic statements. In: Classification and Human Evolution. Edited by S. L. Washburn. Aldine Press, Chicago, pp. 1-31. Smith, E. L., and E. Margoliash 1964 Evolution of cytochrome c. Fed. Proc., 23: 12431247. Straus, W. L. 1949 The riddle of man’s ancestry. Quart. Rev. Biol., 24: 200-223. Stretton, A, 0. W. 1965 The genetic code. Brit. Med. Bull., 21: 229-235. Syner, F. N., and M. Goodman 1966 Differences i n the lactic dehydrogenases of primate brains. Nature, 209: 426427. Tashian, R. E. 1965 Genetic variation and evolution of the carboxylic esterases and carbonic anhydrases of primate erythrocytes. Am. J. Human Genetics, 17: 257-272. Van Valen, L. 1965 Tree shrews, Primates, and fossils. Evolution, 19: 137-151. Vessel, E. S. 1965a Genetic control of isozyme patterns in human tissues. In: Progress in Medical Genetics. H. G. Steinberg and H. G. Bearn (eds.). 196513 Lactate dehydrogenase isozymes: substrate inhibition in various human tissues. Science, 150: 1590-1593. 1965c Lactate dehydrogenase isozyme patterns of human platelets and bovine lens fibers. Science, 150: 1735-1737. Weiner, L. M., I. Macko, E. Poulik and M. Goodman 1964 Salt requirement for precipitation of chicken antisera in agar immunoelectrophoresis. J. Immunol., 93: 228-231. Williams, C. A. 1965 Immunochemical similarity as a n indicator of phylogenetic relationship of protein homologues. In: Protides of the Biological Fluids. Edited by H. Peeters. Elsevier, Amsterdam, 12: 62-69. Wolfe, H. R. 1939 Standardization of the precipitin technique and its application to studies of relationships in mammals, birds, and reptiles. Biol. Bull., 76: 108-120. PRIMATE MACROMOLECULAR SPECIFICITIES Zuckerkandl, E. 1963 Perspectives in molecular anthropology. In: Classification and Human Evolution. Edited by S. L. Washburn. Aldine Press, Chicago, pp. 243-272. 1965 Further principles of chemical paleogenetics as applied to the evolution of hemoglobin. In: Protides of the Biological 275 Fluids. Edited by H. Peeters, Elsevier, Amsterdam, 12: 102-109. Zuckerkandl, E., and L. Pauling 1965 Evolutionary divergence and convergenec in proteins. In: Evolving Genes and Proteins. V. Bryson and H. J. Vogel (eds.). New York, Academic Press.