7 5

Anthropoid Evolution: The Molecular Evidence

John E. Cronin

It has become apparent, in recent years, that mac-
romolecules can provide phylogenetic information
elucidating the branching order of lineages (cladis-
tics), the amount of evolution occurring along lineages
(patristics), and the approximate times of divergence
of these lineages (chronistics). Within the primates,
particularly with respect to the Anthropoidea, a
number of different cladograms have been con-
structed using macromolecular comparisons includ-
ing DNA (Kohne et al. 1972; Hoyer et al. 1972),
hemoglobin (Wilson and Sarich 1969; Goodman and
Moore 1973), fibrinopeptides (Doolittle et al. 1971;
Wooding et al. 1972), myoglobin (Romero-Herrera et
al. 1973), mixed immunology (Goodman and Moore
1971), albumin (Sarich and Wilson 1967; Wilson and
Sarich 1969; Sarich 1970; Cronin and Sarich 1975;
Sarich and Cronin 1977), transferrin (Cronin and
Sarich 1975; Sarich and Cronin 1977).

The serum protein albumin has been studied in a
number of vertebrate groups, other than primates,
including carnivores (Sarich 1969a, 1969b, 1973),
ranid frogs (Wallace, King and Wilson 1973) and hylid
frogs (Maxson and Wilson 1975).

In this paper we construct two independent
phylogenies of the Anthropoidea using the two serum
proteins, albumin and transferrin. The groups
examined consist taxonomically of three super-
families, Hominoidea, Cercopithecoidea and
Ceboidea (Napier and Napier 1967). We find that the
two molecular cladograms are in excellent agreement
with each other and furthermore, that both are in
accord with other molecular data and macromolecular
cladograms. Finally, the total macromolecular evi-
dence supports the use of "molecular clocks."

Experimental Procedures

Serum albumin and serum transferrin were
purified for immunological comparisons from those
species listed in Tables 1, 2 and 3. In every case except
that of Tupaia the proteins were purified from a single
individual.

Transferrin purification

Transferrin was purified by a modification of the
Stratil and Spooner (1971) procedure (Sarich 1974;
Cronin and Sarich 1975). To a given serum was added
FeCl3 (0.1 mg/ml of serum). Rivanol (2-ethoxy 6-9
diaminoacridine lactate; 1 mg/ml serum) was dissolved
in 0.25 M tris (pH 8.8; 0.25 mg/ml of serum) and then

diluted 4 to 5 fold with distilled water. The serum was
then added to the rivanol solution and cooled to 5C.
The solution was centrifuged (12,000g) and
(NH4)2 S04 was added to 40% saturations, then sub-
sequently to 70% saturation. The final solution was
centrifuged at 12,000g and the precipitate, usually a
salmon pink, redissolved into a minimal volume of
isotris buffer dialyzed against isotris containing citric
acid and ferrous ammonium sulfate to promote sat-
uration.

The solution was then subjected to polyacrylamide
disc gel electrophoresis (Wallace 1971; Maxson and
Wilson 1975), alternatively. An 8% cyanogun (95%
acrylamide, 5% bis) solution was used in the gel mix-
ture with a tris-glycine buffer (0.1 M tris, 0.05M glycine
pH 8.8) as both well and gel buffer. In this case,
equilibration was prolonged to 1 hour at 5 ma/tube
(Palmour and Sutton, 1971).

Albumin purification

Albumin was usually purified from serum by
polyacrylamide disc gel electrophoresis (Wallace and
Wilson 1972). In the single case of Cacajao, albumin
was purified by the heat caprylate method of Hoch
and Chanutin (Sarich 1967).

Antisera Production

Antisera were prepared in rabbits according to the
method of Sarich (1967). Individual antisera were
pooled in inverse proportion to their microcomple-
ment fixation (MC'F). Titres and all reported results
are from pooled antisera (Prager and Wilson 1971a
and b). Purity tests were conducted according to pub-
lished techniques (Wallace, King and Wilson 1971). In
some cases, minor components were observed in the
gels, indicating that antibodies had been induced to
proteins other than albumin or transferrin. However,
these minor components do not interfere with mic-
rocomplement fixation experiment results as identical
results are obtained whether one uses the purified
protein or unmodified primate serum as the antigen
source (Sarich 1966; Cronin 1975).

Microcomplement Fixation

Quantitative microcomplement fixation experi-
ments as devised by Wasserman and Levine (1961)
were modified by Sarich and Wilson (1966) and exten-
sively detailed by Champion et al. (1974). Immunolog-
ical distance (y) is generally related to the percent

difference in amino acid sequence (x) between two
molecules by the equation y=5x (Prager and Wilson
1971a and b; Champion et al. 1974). Evidence
suggests that one unit of immunological distance is
approximately equivalent to one amino acid differ-
ence between two species albumin or transferrin
(Maxson and Wilson 1974; Sarich 1974).

Transferrin Saturation

The possibility exists that the binding of iron to
transferrin alters the configuration of the molecule in
such a way as to affect immunological recognition in
microcomplement fixation experiments. Normally,
transferrin is about one third saturated with apo trans-
ferrin (Tf), monoferro, and diferro molecules present
in plasma (Palmour and Sutton 1971). Presumably
upon injection into the rabbit, molecules of all three
types would be present.

To test the effect of saturation on immunological
recognition human transferrin was deprived of iron,
resaturated and tested immunologically with purified
human transferrin and serum as controls. Microcom-
plement fixation tests were performed using anti-
human transferrin (genotype CC) with apo Tf,
purified Tf, serum and 80% saturated Tf. No differ-
ence was detected in any of these duplicated experi-
ments. In addition similar procedures were carried
out on chicken ovo transferrin and identical results
obtained (Prager et al. 1974).

Results

Construction of Phylogenetic Trees

The phylogenetic trees were constructed using the
method of Sarich (Sarich 1969; Cronin and Sarich
1975). The immunological distance values used were
the average of the values determined by reciprocal
tests. The phylogenetic trees presented have a stated
percent deviation (Fitch and Margoliash 1967) as a
measure of "goodness of fit" where input data is com-
pared with the output data. The latter is calculated by
adding up the immunological distance units separat-
ing any pair of species in the cladogram.

The method of tree construction is similar to that of
Fitch and Margoliash (1967). The difference lies in the
fact that to apportion lineage lengths between two
species A and B, Fitch and Margoliash average the
distances from A to every other species and from B to
every other species. The method as described in Cro-
nin and Sarich (1975) does not use every other species
as an outside reference point but uses the next closest
group to align branch lengths in the group under
study. For example, anti-cercopithecoid antisera
would serve as outside reference points for the
Hominoidea and similarly the anti-ceboid antisera

would serve as outside reference points for aligning
the cercopithecoid lineages. The different methods of
construction yield similar although not identical re-

sults (Farris 1972; Wallace et al. 1973; Maxson and
Wilson 1975).

Antisera to the purified transferrins from 5
hominoid species, 5 cercopithecoid species, and 5
ceboid species were used to investigate the phyletic
relationships of the suborder Anthropoidea. Using
the microcomplement fixation technique each an-
tiserum was tested with transferrin of every species
listed in Table 1. Table 1 gives all immunological dis-
tances between all 105 possible pairs of species. Thus,
between every pair of species 2 immunological dis-
tances were obtained.

Similarly antisera to the purified albumins from 5
hominoid, 5 cercopithecoid and 10 ceboid species
were used to investigate the phyletic relationships
within the Anthropoidea. Table 2 gives the matrix of
immunological distance units for all 45 pairs of catar-
rhine species. Table 3 gives the matrix of immunolog-
ical distance units for all 45 pairs of platyrrhine
species.

The trees constructed are shown in Figures 1 and 2.

The percent standard deviation of the transferrin
cladogram is 8.9%. The lowest tree for 20 eukaryotic
cytochrome c sequences constructed by Fitch and
Margoliash (1967) was 8.7%. No tree this large has
been published previously using immunological data.
For a tree of 12 hylid frog species Maxson and Wilson
(1975) report 10% standard deviation. For a tree of 15
cercopithecoid species (Cronin and Sarich in prep.) a
12.5% standard deviation was obtained. The an-
thropoid albumin tree has a 8.0% standard deviation.

Correlation of Albumin and Transferrin Distances
Albumin immunological distances are correlated
with transferrin immunological distances. To illus-
trate this one may consider all the pairs of catarrhine
species that have been subjected to both albumin (Ta-
ble 2) and transferrin (Table 1) comparison. For these
45 pairs a product moment correlation coefficient of
0.86 was observed (p <<O.O1). Similarly when the aver-
age albumin distances given in Table 4 for primates as
a whole are compared with the corresponding trans-
ferrin distances r = 0.83 (p<<0.01).

The equation for the regression lines were as fol-
lows:

Catarrhine A= mT+b
Primates A= mT+b

Where A is the albumin immunological distance and T
is the transferrin immunological distance. Transferrin
immunological distances are on the average 1.5 fold
greater than albumin distances between the same
species. Similar results have been found in comparison
of carnivore, rodent and chiroptera albumins and
transferrins (Sarich 1973). In birds transferrin dis-

tances average 2.2 times the albumin distances.
(Prager et al. 1974).

7 6

7 7

TABLE 1

Immunological Distances Among Anthropoid Transferrins

ANTISERA
T

p
A
N

13

10
18
22
33
41
36
33
33
28
78
71
83
82
80

G
0
R
I
L
L
A

5
5

24
23
37
51
37
31
31
36
61
60
70
70
68

p
0
N
G
0

33
36
37

35
43
66
61
44
40
39
83
82
101

92
95

H
y
L
0
B
A
T
E
S

27
27
26
28

43
63
45
39
37
42
84
80
86
89
87

p
R
E
S
B
y
T
I
S

64
54
59
52
55

43
34
34
34
31
95
92
90
97
97

H
E
R
0
p
I
T
H
E
C
U
S

55
52
50
50
47
20

5
13
16
13
85
89
95
85
92

p
A
p
I
0

C
y
N
0
p
I
T
H
E
C
U
S

61    49
56    46
52    45
50    42
45    44
15   19

3    20
-   18
16     -
20    19
20     7
68    82
110   77
108   72

99   71
106   78

E
R
y
T
H
R
0
C
E
B
U
S

52
46
49
47
45
21
24
21
12

9
75
79
81
76
75

C
E
R
C
0
C
E
B
U
S

A
T
E
L
E
S

42    73
41    71
39    75
38    85
58    78
21    79
24   100
23    98
12   86
23    86

-   82
79     -
81   31
79    32
81    28
81    28

A   C
0   E
T   B
U   U
S   S

73    75
73    74
73   62
83   71
78    82
89    96
100   112
100   109
89   100
86   91
94    89
28    30

-   32
30     -
30    29
30    35

C
A
L
L
I
T
H
R
I
x

S
A
G
U
I
N
U
S

70    86
70   74
68    76
74   88
74    85
90   104
103   113
101   110
85   101
81   94
75   92
26   25
35   28
36   32

-    8
6     -

TABLE 2

Immunological Distances Among Catarrhine Albumins.

ANTISERA

C.
C.

H                 P    G
Y           A    R     A

G           L          E     E     L     P.
0           0    M.    T     S     E

R     P     B          H     B     R     P
H           I    0     A     I     I     Y     I    A
0     P     L    N     T     R     0     T    T      P
M     A     L     G    E     U     P     I     U     I
ANTIGEN                  0     N    A     0     S     S     S     S    S     0
Homo                      -    4     4    11    10   37    49    42    37   34
Pan                      7     -     9    11    14   37    49    46    38   31
Gorilla                  5     7     -    11   10    32    62    51   37    36
Pongo                   12     8    10     -    11   42    57    48    38   40
Hylobates               13    13    12    11     -   36    49    44    35   43
M. irus                 37    30    21   38    36      -   12    24     6    0
C. aethiops             44    42    21   33    37    10     -    22     9    4
Presbytis               42    40    22    38   38    24    34     -    28    16
C. galeritus            39    32    20   33    33     7     0    19     -    4
P. papio                40    30    18    34   36     2     7    25     8     -

H
0
M
0

ANTIGEN

Homo
Pan

Gorilla
Pongo

Hylobates
Presbytis

Theropithecus
Papio

Cynopithecus
Erythrocebus
Cercocebus
Ateles
Aotus
Cebus

Callithrix
Saguinus

7
8
16
20
34
41
37
34
34
35
64
67
66
67
64

7 8

TABLE 3

Intraplatyrrhine Albumin Immunological Distances

ANTISERUM

C                                  C(
A                                  A

L     A     P                 S    1,   R
L     L     I     S    C     A     L,   E

A           I     0     r     A    A     G      I   L, Ra
A     T     C     C     U     H     I    C,    U     T    A  A
0     E     E     E     A     E    M     A      I    H    T F
T     L     B     B     T     C     I    J     N     R     I E
U     E     U     U     T     I     R    A     U      I   V
AN'FIGEN                  S     S     S     S    A     A      I    0     S     X    E

Aotus trivirgatus           0    31    29    26    28    18    32    22    32    27   -12
Ateles geoJfroyi           33     0    31    19    21    27    29    17    43    42     0
Cebus capucinus            37    43     0    30    36    25    38    37    59   56   +12
Callicebus cupreus         36    33    29     0    36    28    48    24    50   41     -2
Alouatta palliata          38     12   30    22     0    31    42    31    56    55   +6
Pithecia monachus          29    29    18    12    34     0    24     4    50    30    -4
Saimiri scuirea            40    42    3 1   34    36    39     0    43    60   46    +8
Cacajao rubicundus         33    36    29    24    33     9    45     0    41   34    +2
Saguinus leucopus          36    27    17    16    25    26    27    14     0   29    +3
Callithrix jaccus          29    39    37    22    45    37    40    22    29    0    +2

aFhese numbers indicate the mean amount by which the given albumin species is more distant (+) or less

distant (-) than Ateles from a series of nonplatyrrhine albumins (Homo, Pan, Macaca, Cercopithecus, Presbytis,
Nycticebus) from Sarich and Wilson 1967; Sarich 1970.

TABLE 4

Immunological Distances Among Primate Albumins and Transferrinsa

Non-

H     C     P     Ta   Tu     Cy    Le    Lo primatesb

Hominoidea

C(ercopithecoidea
Platyrrhini
Tarsius
Tupaia

Cynocephalus

Lemuriformes
Lorisiformes

TRANSFERRI N

0    33    58   123    98   117   112    94   +19

45     0    56   128   102    114   110   107  + 19  A
76    91     0   109    89    98    112   107  + 12  L
172   164   156     0    93   109   122    95    +5   B
170   190   180   159     0   103   117    108    -5  U
175   156   177   161   183     0   109    113   +5   M
164   171   164   156   178   162     0    88      0  I
196   195   193   178   183   185   118     0     -4  N

aThese data have been gathered using antisera to the albumins and transferrins of Homo,
Pan, Gorilla, Pongo, Hylobates, Papio, Cercocebus, Cercopithecus, Presbytis, Aotus, Ateles, Cebus,
Callithrix, Saguinus, Lemur, Avahi, Lepilemur, Propithecus, Phaner, Daubentonia, Nycticebus,
Galago, Tupaia, and Cynocephalus; to the transferrin of Macaca nigra; and to the albumins of
Ursus, Genetta, Hyaena, Felis, Hipposideros, Pteropus, Tadarida, Antrozous, Bradypus, Cabassous,
Tamandua, Scapanus, and Solenodon.

bThese numbers represent the amount by which those albumins are more (+) or less (-)
distant from the various nonprimates than the lemuriform mean. A similar analysis is not
possible for the transferrins as antisera to nonprimate transferrins do not generally react
with those of primates.

TRANSFERRIN PHYLOGENY

I       ~  ~I       I           I          I          I

50         40          30         20          10          0

IMMUNOLOGICAL    DISTANCE

Figure 1. A transferrin cladogram of the Anthropoidea constructed according to
the method of Sarich (1969). The immunological distances on which this cladog-
ram is based are those of Table 1. The percent standard deviation is 8.9%. The
number in parentheses following the species names represent the amount of
transferrin immunological distance accumulated on that species lineage from the
node separating the catarrhines and platyrrhines.

ANTHROPOID ALBUMIN PHYLOGENY

s         . PONGO

3,    HOMO
13                 3      3  .PAN

/                      ~~~~~~~3 GORILLA

/  6      . HYLOBATES
/4           PRESBYTIS

;   7              r      4  , Cercopithecus aethiops

/; ^ _^ ~~~~~~~~~Mocaca irus

0     Papio papio
\                             ~~~~~~~~~~~2  .  C- gobritus
\6

\  4      . AOTUS

12%Sguinus/Calfithrix
Is     Ateles /Alouatta

CALLICEBUS
<   21     , CEBUS
\   24     . SAIMIRI

10           Pithecia /Cacaiao

Figure 2. An albumin cladogram of the Anthropoidea constructed as in Figure 1.
The immunological distances on which this cladogram is based are those of Tables
2 and 3. The percent standard deviation is 8.0%.

7 9

Anthropoid Evolution

The agreement, in terms of cladistics, between the
albumin and transferrin cladograms is evident. No
major differences are readily apparent. There is one
difference and that is in terms of the placement of the
ceboid lineage Aotus. Previous albumin studies placed
it diverging before the major ceboid radiation (Sarich
1967). The transferrin data, however, suggest that it is
one of the 7 major ceboid lineages which radiate from
a common node (Cronin and Sarich 1975) as do the
hemoglobin data (Boyer et al. 1971).

Beyond that specific discrepancy (discussed in
depth in Cronin and Sarich 1975) no other specific
disagreements are noted. It is clear that the species
within Anthropoidea comprise a group with a shared
lineage of some considerable time depth. The initial
divergence is between the Old and New World forms,
the Catarrhini and Platyrrhini. The former group sub-
sequently diverges into the two constituent subunits,
the Hominoidea and the Cercopithecoidea. Thus, the
major groupings apparent at the molecular level are
in good agreement with the classification of the An-
thropoidea (Napier 1967).

The fact that the two molecular cladograms are in
such close agreement supports the using of im-
munological data in order to elucidate the phylogene-
tic relationship of taxa.

Given the data, it is readily apparent that the An-
thropoidea is composed of three clusters of species,
including the Hominoidea, the Cercopithecoidea and
the Ceboidea. The former are closer to each other
from almost every molecular measure, except for the
primate lysozyme data, than they are to the latter
group the Ceboidea. Reciprocity problems are inhe-
rent in the lysozyme - anti lysozyme system as has
been pointed out (Prager et al. 1974). The overwhelm-
ing evidence is that the Catarrhini are a unit relative to
the Platyrrhini.

All the molecular data support the conclusion that
the Anthropoidea share a common lineage subsequent
to the divergence of any prosimian stock. Along this
lineage accumulated some 70 units of combined albu-
min and transferrin units of change, three myoglobin
sequence changes, seven fibrinopeptide changes, five
beta hemoglobin changes and two units of antigenic
distance, before the divergence of the New World
monkeys.

The Ceboidea form a fairly tight unit (evolutionary
radiation) with a division into units, the Callithricidae
plus Callimico, and a number of other fairly indepen-
dent ceboid lineages (Cronin and Sarich 1975; Baba et
al. 1975; Boyer et al. 1971).

The Catarrhini form a group relative to the Platyr-
rhini, with respect to serum proteins, albumin and
transferrin (Sarich 1970; Sarich and Cronin 1975),

prealbumin, albumin, alpha 2 macroglobulin, thry-
globulin, transferrin, ceruloplasmin and gamma
globulin (Goodman et al. 1971), DNA (Kohne 1972),

myoglobin (Romero-Herrera 1973), hemoglobin
(Goodman et al. 1973, 1974), fibrinopeptides (Wood-
ing et al. 1972) and carbonic anhydrase (Nonno et al.
1969).

The Catarrhini are traditionally broken into two
taxonomic units, the Hominoidea and the Cer-
copithecoidea on morphological grounds. This is also
true with respect to the molecular division (all refer-
ences above and figures this paper). The Cer-
copithecoidea have had little detailed genetic work.Yet
it is apparent that the basic division into leaf eaters
(Colobinae) and non leaf eaters (Cercopithicine) is
recognized at the molecular level. (Sarich 1970;
Barnicot and Wade 1970; Goodman and Moore 1971).
A detailed presentation of the evolution of the Cer-
copithecoidea as determined by the microcomplement
fixation technique is in preparation (Cronin and
Sarich) arid a brief preliminary version is in press
(Sarich and Cronin 1977).

Albumin and Transferrin Molecular Clocks

It appears that albumin accumulates approximately
50 units of change per 60 million years per lineage or
100 units per 60 million years of separation. These
results are found for a large variety of vertebrates
besides primates (Sarich 1970) including carnivores
(Sarich 1969), ranid frogs (Wallace, et al. 1973), hylid
frogs (Maxson and Wilson 1975) and iguanid lizards
(Gorman et al. 1971).

Transferrin accumulates about 150 units per 60
m.y. of separation in mammalian and snake lineages
(Mao et al. 1971; Cronin and Sarich 1975; Sarich

1974). In birds, the rate of transferrin evolution is ?4 to

'/2 that found in mammals and snakes. Similar results
are observed in serum albumin, with a rate of evolu-
tion '3 that found in mammals (Prager et al. 1974).
This phenomen of an evolutionary slowdown is not
unique to these two molecules. Lysozyme and cyto-
chrome c seem to have evolved about twice as slowly in
birds as in mammals (Prager et al. 1974). This problem
points to the fact that rates of evolution cannot be as-
sumed; they must be calculated for every group under
study.

In the equation: Immunological Distance = K X
Time (106 year), K approximates 1.67 for albumin
and 2.5 for transferrin per million years for mamma-
lian lineages. Thus, 250 units of combined albumin
plus transferrin immunological distance calibrates to
approximately 60 million years. Using this calibration
point for albumin and transferrin it can be seen in
Table 5 what the approximate times of separation of
various taxa are for the two molecules independently
and averaged.

The dates of divergence calculated from the im-

munological data presented in this study are in good
agreement with dates proposed by Sarich (1970);
Sarich and Wilson (1967a and b); and Wilson and
Sarich (1969). Certain differences are, however, appa-

8 0

8 1

TABLE 5

Times of Divergence Among Various Primate Taxa Predicted by

Albumin and Transferrin Immunological Distances

Time of Divergence (106 yrs.)a

Taxa                               Albumin    TIransferrin    Average

Homo vs Pan vs Gorillab               3.6            4          3.8
Homo vs Pongo vs Hylobatesb            7.2         9.6          8.4
Colobinae vs Cercopithecinae          14.4        10.8         12.6
Hominoidea vs Cercopithecoidea       22.2           18         20.1
Platyrrhini vs Catarrhini            34.2         33.6         33.9
Anthropoidea vs Prosimii             70.2           64         67.1

aTimes of divergence are calculated on the basis that approximately 100 ablumin and

150 transferrin units accumulate between two taxa over sixty million years.

bThe taxa listed are cladistically equally divergent from a common node so that it is a

three way comparison. Since all three lineages separated cladistically from the same
node, they have existed for equal evolutionary time.

rent. It was postulated that prosimians and an-
thropoids differed by about 100 albumin units. This
figure has had to be adjusted upward by about 10 units
given the additon of five new lemurid albumin anti-
sera.

A second point of difference is the date of 30 million
years for the hominoid-cercopithecoid divergence
used to calculate the man-chimpanzee divergence time
of approximately 5 m.y. BP (Sarich and Wilson
1967b). In subsequent publications a divergence time
of approximately 22 million years has been estimated
for the hominoid-cercopithecoid separation (Sarich
1970, 1971). The date estimated from the transferrin
data of 18 m.y. BP is in close agreement with this latter
figure. The initial publication (Sarich and Wilson
1967b) assumed a hominoid-cercopithecoid divergence
time of approximately 30 m.y. BP. From this date,
given the ratio of the man-chimpanzee albumin dis-
tance of 6 units versus the hominoid-cercopithecoid
distance of 35 immunological distance (I.D.) units; and
given the hypothesis of the regularity of protein
evolution a time of divergence of 5 m.y. BP was calcu-
lated for man and chimpanzee. Additional albumin
data have now been gathered. Given the data from
Table 3, a mean hominoid-cercopithecoid distance of
37 I.D. units is obtained. A cladistic analysis of the
hominoids indicates that man, chimpanzee and gorilla
diverged from a common node into three separate
lineages. Each lineage is a sample of the amount of
change given the time of divergence for those lineages.
A mean distance using all descendent lineages from a
node may give a better representation of the actual
distance between species descending from that node
than only one measurement. This consideration gives
a mean man-chimpanzee-gorilla albumin distance of 6

units. The same analysis on the transferrin data reveal
a mean distance of 8 units. The hominoid-
cercopithecoid mean is 45 transferrin units. The ra-
tio is 1:6.1 for albumin and 1:5.6 for tranferrih.
Given the small distances involved between the great
apes and man, the two ratios are consistent and predict
that the human-chimpanzee divergence is approxi-
mately 5 m.y. BP.

Agreement Among Genetic Distances

Fortunately, with Anthropoidea we do not merely
need to rely upon only a comparison of albumin and
transferrin genetic distances. The reliability of these
phylogenies may be tested with other molecules and
there is good agreement with the cladograms pro-
duced by studies of hemoglobin (Wilson and Sarich
1969; Goodman and Moore 1973) fibrinopeptides
(Doolittle et al. 1971; Wooding et al. 1972), DNA
(Kohne et al. 1972; Hoyer et al. 1972) myoglobin
(Romero-Herrera et al. 1973) transferrin (Cronin and
Sarich 1975) and mixed immunology (Goodman and
Moore 1971).

Table 6 lists the relative genetic distances between 5
primate taxa. The molecular data when considered as
a whole support the hypothesis that the immunologi-
cal data are not unique or unrepresentative of such
macromolecular evidence. The correlation coefficient
for the nine molecular systems among the 5 taxa com-
pared is 0.96 (p <<(.01). The rank correlation coeffi-
cient is 1.0 (Cronin, Sarich and Dempster in prep.). It
appears that data on any one of the molecules yields
substantially the same conclusions as data from any
other molecular system.

8 2

TABLE 6

Relative Molecular Differences Among Various Primate Groups Referred to a

Catarrhine-Platyrrhine Distance of 1.0

Trans-               Mixed       Myo-      Fibrino-      Beta      Carbonic     Lyso-
Groups Compared             Albumina     ferrinb   DNA' Immunologyd      globine   peptidef Haemoglobinh Anhydrase'     zym'
Homo vs Pan                   0.12        0.13      0.15       0.12      0.007          0            0        .05          0
Homo vs Pongo vs Hylobate.s9  0.25       0.30       0.33       0.29       .143        0.3          .43        NA         NA
Hominoidea vs Cercopithecoidea 0.58      0.53       0.61      0.53        0.47        0.5         .85         .65        .90
Catarrhini vs Platyrrhini      1.0         1.0       1.0        1.0        1.0        1.0          1.0        1.(         1.0
Anthropoidea vs Prosimii       2.1         1.8       2.6       1.67        1.5        1.8         1.57       1.73        1.45

aFrom Sarich (1970) and subsequent work in this laboratory. The number is immunological distance between the groups divided by the catarrhine-

platyrrhine albumin immunological distance.

bData from this laboratory and expressed as in a.

'Calculated from the data in Hoyer et al. (1972) and Kohne, Chiscon, and Hoyer (1972) by dividing the ATm for the groups by the corresponding
catarrhine-platyrrhine value.

dFrom Goodman and Lasker (1973).

eFrom Romero Herrera and Lehmann (1973). The calculations are based on amino acid differences.

fFrom Wooding and Doolittle (1972); Tables I and 2). The calculations are based on amino acid differences.
gThe taxa listed are cladistically equally divergent from a common node so that it is a three-way comparison.

hFrorn Goodman and Lasker (1973). The data presented is based on amino acid differences found between the taxa.
'From Nonno et al. (1969).

JFrom Hanke et al. (1973; Table 11). Only the data gathered with the anti human lysozyme is included.
NA: Data Not Available

Implications for Primate Evolution

The albumin-transferrin "molecular clock" suggests
that the Anthropoidea diverged from other primate
lineages approximately 70 m.y. BP. This date is very
close to the date postulated for the beginning of the
Tertiary. Furthermore, if correct, the separation of
various major lineages of the primates predates the
postulated first known primnatePurgatorius (Van Valen
and Sloan 1965; Clemens 1974). Anthropoids may
have arisen in North America in the late Cretaceous.
Subsequently, the ancestral anthropoids migrated to
the Old World along with other ancestral primate
forms before the North American-European land
connection was broken in the Eocene (Walker 1972).

The initial divergence among the Anthropoidea,
that between the Old World forms (man, apes and Old
World Monkeys) and the New World forms (New
World monkeys) occurred about 35 m.y. BP in the Old
World, according to "molecular clock" considerations.
This argues that the introduction of anthropoids into
South America was via Africa. Anthropoid grade

forms are found in deposits in Africa at a slightly
younger date (Simons 1965, 1967, 1972). Similar con-
clusions have been reached from interpretations of
morphological and fossil data. (Hershkovitz 1972;
Hoffstetter 1972, 1974). There are however, objec-
tions to this interpretation (Simons 1972).

The next major division within the Anthropoidea,
the separation of the lineages leading to the
Hominoidea and the Cercopithecoidea occurred ap-
proximately 20 m.y. BP and was most likely African in
location.

Alternative interpretations exist as to the origin and
divergence of the major lineages of the Catarrhini
(Delson 1974; Simons 1969, 1972); a more detailed
picture of within-hominoid and within-cercopithecoid
relationships will be presented elsewhere. Yet, it is
clear that the basic picture of anthropoid cladistics has
been documented through the use of macromolecular
comparisons and is in substantial agreement with in-
terpretations based on anatomical data. However, the
problem as to the time of the origin of these major

anthropoid lineages remains controversial (Simons
1972; Delson 1974; Read 1975; Delson and Andrews
1976).

In conclusion the fact that the macromolecules are
in agreement as to the relationships among the
Anthropoidea is an important observation. Mac-
romolecules may be as unambiguous a source of
phylogenetic information as any body of data. It is
interesting that within each life form a history of that
form is encoded. We need only to know how to read
the enclosed message.

Acknowledgement

The author wishes to thank V.M. Sarich for help in
both the research and in the writing of this article. Dr.
A. Wilson generously provided laboratory space.
Thanks go to A. Hill and P. Hornbeck for technical
assistance. L. Brunker generously aided in the prep-
aration of the technical illustrations. Thanks also go to
E. Meikle, N. Boaz and L. Brunker in editing and
commenting on the drafts of the manuscripts. T. Cord
generously provided her time in typing the manus-
cript.

REFERENCES CITED

Barnicot, N.A., and P.T. Wade 1970 Protein Structure

and the Systematics of Old World Monkeys. In Old
World Monkeys: Evolution, Systematics and Be-
havior, J.R. Napier and P.H. Napier, eds. New
York: Academic Press.

Boyer, S.H. et al. 1971 Primate Haemoglobins: Some

Sequence and Some Proposals Concerning the
Character of Evolution and Mutation. Biochem.
Genetics 5:405-448.

Champion, A.C., E.M. Prager, D. Wachter, and A.C.

Wilson 1974 Microcomplement Fixation. In
Biochemical and Immunological Taxonomy of
Animals, C.A. Wright, ed. London: Academic
Press.

Clemens, W.A. 1974 Purgatorius, An Early

Paromomyid Primate (Mammalia). Science
184:903-905.

Cronin, J.E. 1975 Molecular Systematics of the Order

Primates. Ph.D. Dissertation, University of Califor-
nia, Berkeley.

Cronin, J.E., and V.M. Sarich 1975 Molecular Sys-

tematics of the New World Monkeys. J. Human
Evol. 4:357-375.

Cronin,J.E., V.M. Sarich and E. Dempster 1977 Rates

of Molecular Evolution in the Primates. Unpub-
lished Manuscript.

Delson, E. 1975 Evolutionary History of the Cer-

copithecidae. In Approaches to Primate Paleobiol-
ogy, F. Szalay, ed. Contrib. Primate. 5:167-217.

Delson, E., and P. Andrews 1975 Evolution and Inter-

relationships of the Catarrhine Primates. In
Phylogeny of the Primates, W.P. Luckett and F.S.
Szalay, eds. New York: Plenum.

Doolittle, R.F., G.C. Wooding, Y. Sen, and H. Riley

1971 Hominoid Evolution as Judged by Fib-
rinopeptide Structures. J. Molec. Evol. 1:74-83.

Farris, J.S. 1970 Methods for Computing Wagner

Trees. Syst. Zool. 19:83-92.

Fitch, W.H. and E. Margoliash 1967 Construction of

Phylogenetic Trees. Science 155:279-284.

Goodman, M. and W. Moore 1971 Immunodiffusion

Systematics of the Primates. I. The Catarrhines.
Syst. Zool. 20:19-62.

Goodman, M., and G.W. Moore 1974 Phylogeny of

Haemoglobin. Syst. Zool. 32:508-532.

Gorman, G.C., A.C. Wilson and M. Nakanishi 1971 A

Biochemical Approach the Study of Reptili?n
Phylogeny: Evolution of Serum Albumin and Lactic
Dehydrogenase. Syst. Zool. 20:167-185.

Hershkovitz, P. 1970 Notes on Tertiary Platyrrhine

Monkeys and Description of a New Genus from the
Late Miocene of Columbia. Folia Primat. 12:1-37.
1972 The Recent Mammals of the Neotropical Reg-

ion: A Zoogeographic and Ecological Review. In
Evolution, Mammals and Southern Continents, A.
Keast, F.C. Erk and B. Glass, eds. Albany: State
University of New York Press.

8 3

Hoffstetter, R. 1972 Relationship, Origins and History

of the Ceboid Monkeys and Caviomorph Rodents:
A Modern Reintrepretation. In Evolutionary Biol-
ogy, Vol. 6, T. Dobzhansky, M.K. Hecht and W.C.
Steere, eds. New York: Appleton, Century Crofts.
Hoyer, B.H., N.W. van de Velde and M. Goodman

1972 Examination of Hominoid Evolution by DNA
Sequence Homology. J. Human Evol. 1:645-649.

Kohne, D.E., J.A. Chiscon, and B.H. Hoyer 1972

Evolution of Primate DNA Sequences. J. Human
Evol. 1:627-644.

Mao, Shao-hsien, and H.C. Dessauer 1971 Selectively

Neutral Mutations, Transferrins and the Evolution
of Natricine Snakes. Comp. Biochem. Phys.
40:669-680.

Maxson, L., and A.C. Wilson 1974 Convergent Mor-

phological Evolution Detected by Studying Proteins
of Tree Frogs in the Hyla exemia Species Group.
Science 185:66-68.

1975 Albumin Evolution and Organismal Evolution
in Tree Frogs (Hylidae). Syst. Zool. 24:1-15.

Napier, J.R., and P.H. Napier 1967 A Handbook of

Living Primates. New York: Academic Press.

Palmour, R.M., and H.E. Sutton 1971 Vertebrate

Transferrins, Molecular Weights, Chemical Com-
position, and Iron Binding Studies. Biochem.
10:4026-4032.

Prager, E.M., and A.C. Wilson 1971a The Depen-

dence of Immunological Cross-Reactivity Upon
Sequence Resemblances Among Lysozymes. J. Biol.
Chem. 246:5978-5989.

1971b The Dependence of Immunological Cross-
Reactivity Upon Sequence Resemblance Among
Lysozymes. II. Comparison of Precipitin and Mic-
rocomplement Fixation Results. J. Biol. Chem.
246:7010-7017.

Prager, E.M., A.H. Brush, R.A. Nolan, M. Nakanishi

and A.C. Wilson 1974 Slow Evolution of Transfer-
rin and Albumin in Birds According to Microcom-
plement Fixation Analysis.J. Mol. Biol. 3:243-262.

Read, D. 1975 Primate Phylogeny, Neutral Mutations

and Molecular Clocks. Syst. Zool. 24:209-221.

Romero-Herrera, A.E., H. Lehmann, K.A.Joysey and

A.E. Friday 1973 Molecular Evolution of Myoglo-
bin and the Fossil Record: A Phylogenetic Synthe-
sis. Nature 246:389-395.

Sarich, V.M. 1967 A Quantitative Immunological

Study of Evolution of Primate Albumins. Ph.D. Dis-
sertation. University of California, Berkeley.

1968 Human Origins: An Immunological View. In
Perspectives on Human Evolution, S.L. Washburn
and P.C. Jay, eds. New York: Holt, Rinehart and
Winston.

1969a Pinneped Origins and the Rate of Evolution
of Carnivore Albumins. Syst. Zool. 18:286-295.

1969b Pinneped Phylogeny. Syst. Zool. 18:416-422.

1970 Primate Systematics with Special Reference to
Old World Monkeys: A Protein Perspective. In Old
World Monkeys: Evolution, Systematics and Be-
havior,J.R. Napier and P.H. Napier eds. New York:
Academic Press.

1973 The Grand Panda is a Bear. Nature 245:218-
219.

1974 Just How Old is the Hominid Line? Yearbook
of Physical Anthropology 17:98-112.

Sarich, V.M., and A.C. Wilson 1966 Quantitative Im-

munochemistry and the Evolution of Primate Al-
bumins. Microcomplement Fixation. Science
154:1563-1566.

1967a Rates of Albumin Evolution in Primates.
Proc. Nat. Acad. Sci. 58:142-148.

1967b Immunological Time Scale for Hominid
Evolution. Science 158: 1200-1203.

Sarich, 'V.M., and J.E. Cronin 1977 Molecular Sys-

tematics of the Primates. In Molecular Anthropol-
ogy, M. Goodman and R. Tashian, eds. New York:
Plenum (In Press.)

Simons, E.L. 1965 New Fossil Apes from Egypt and

the Initial Differentiation of Hominoidea. Nature
205: 135-139.

1967 The Earliest Apes. Sci. Amer. 217:28-35.

1972 Primate Evolution. New York: Macmillan Co.

Stratil, A., and R.L. Spooner 1971 Isolation and Prop-

erties of Individual Components of Cattle Trans-
ferrin: The Role of Sialic Acid. Biochem. Gen.
5:347-365.

Walker, A. 1972 The Dissemination and Segregation

of Early Primates in Relation to Continental Con-
figurations. In Calibration of Hominoid Evolution,
W.W. Bishop and J.A. Miller, eds. New York: Scot-
tish Academic Press.

Wallace, D.G. 1971 A Comparison of Amphibian Al-
bumins. Taxonomic and Evolutionary Significance.

Ph.D. Dissertation, University of California, Ber-
keley.

Wallace, D.G., M.C. King and A.C. Wilson 1972 Al-

bumin Differences Among Ranid Frogs:
Taxonomic and Phylogenetic Implications. Syst.
Zool. 22:1-14.

Wallace, D.G., and A.C. Wilson 1972 Comparison of

Frog Albumins with That of Other Vertebrates. J.
Mol. Evol. 2:72-86.

Wooding, G., and R. Doolittle 1972 Primate Fib-

rinopeptides: Evolutionary Significance. J. Hum.
Evol. 1:553-563.

Van Valen, L., and R.E. Sloan 1965 The Earliest Pri-

mates. Science 150:743-745.

Wasserman, E. and L. Levine 1961 Quantitative Mic-

rocomplement Fixation and Its Use in the Study of
Antigenic Structure by Specific Antigen-Antibody
Inhibition. J. Immunology 87:290-295.

Wilson, A.C., and V.M. Sarich 1969 A Molecular Time

Scale for Human Evolution. Proc. Nat. Acad. Sci.
63:1088-1093.

8 4