236 
PACIFIC SCIENCE, Vol. 1, October, 1947 
study is there a gap between ranges of com¬ 
parable measurements that would make this 
segregation feasible. In all elements where 
there is a difference in average size, am plus 
is the larger, except for the distal widths of 
the ulna and tibiotarsus. Extremely large 
or small specimens can be allocated in most 
instances to am plus and calijornianus, re¬ 
spectively. 
Ratios designed to show the proportions 
of the bones cannot be used as a means of 
differentiation for individual bones. In gen¬ 
eral, the bones are sturdier in the fossil form, 
but the ranges of comparable ratios in the 
two series consistently overlap. 
Despite this inability to separate the two 
series, the significance tests made on the 
means of 34 different measurements and 
ratios demonstrate conclusively that we are 
dealing with two distinct forms which, in 
most measurements, have significantly dif- 
TABLE 16 
Results of Significance (t) Tests on Means 
t P 
Sternal length . 2.84 <.02 
Keel length . 2.54 >.02 
Keel height .. 2.68 <.02 
Keel, anterior width . 2.55 >.02 
Keel length : total length . 1.21 >.10 
Keel height: keel length. 0.58 >.10 
Width scapular blade . 2.58 >.02 
Spread of furculum . 1.10 >.10 
ferent means. Only 5 of the 34 tests failed 
to indicate significant differences between 
the means and the two populations repre¬ 
sented by them. These 5 tests were on length 
of digit III of the wing, furcular spread, and 
two ratios on sternal proportions. However, 
the significance test on both the means and 
variances shows also that length of ulna, 
length of digit I of wing, and length of 
phalanx 1 of digit II are not significantly 
different. 
Gymnogyps am plus was found to be no 
more variable than G. calijornianus. The co¬ 
efficient of variation was calculated for 33 
identical measurements and ratios in each 
species. In 18 ratios and measurements the 
V was greater in calijornianus than in 
am plus; in 15 instances the reverse was true. 
The mean V for am plus was 3.56 and that 
for calijornianus was 4.30. The relatively 
small samples of several elements for cali¬ 
jornianus in all probability reduced the co¬ 
efficient of variation disproportionately in 
this species, although the V for two ele¬ 
ments was abnormally high. In one series 
(phalanges of the foot) where it seemed im¬ 
practical to calculate the coefficient of varia¬ 
tion for calijornianus, all the coefficients in 
the large series for am plus were similar to 
those usually found in zoological series. 
Because coefficients of variation of linear 
measurements and of ratios are not com¬ 
parable, the Vs were separated. In the 
Recent species the average V for linear 
measurements was 4.00, compared to 3.69 
for the fossil. The Vs for the ratios were 
5.45 for calijornianus and 2.98 for amplus. 
To make the comparison even more reliable, 
only those pairs of Vs based on samples of 
10 or more linear measurements were in¬ 
cluded in the last examination of these co¬ 
efficients; the average V was 4.32 in cali¬ 
jornianus and 3.78 in amplus. Further, in 
only three of the pairs in this analysis was 
the V higher in the Pleistocene species. 
Therefore, for Gymnogyps at least, Pleisto¬ 
cene and Recent specids show no major dif¬ 
ferences in total variability. 
REFERENCES 
Fisher, Harvey I. The skulls of cathartid vul¬ 
tures. Condor 46: 272-296, 1944. 
- Adaptations and comparative anatomy 
of the locomotor apparatus of New World 
vultures. Amer. Midland Nat. 35: 545-727, 
1946. 
Simpson, George G. Tempo and mode in evo¬ 
lution. xviii + 237 p. Columbia Univ. Press, 
New York, 1944. 
-—, and Anna Roe. Quantitative zoology. 
xvii + 414 p. McGraw-Hill, New York, 1939. 
