U-iv.vL4.i-' 
. \ ... <ln 
Measurement of Variability in Species 
of Fossil Invertebrates 
John Imbrie, Columbia University 
The coefficient of variation ( V ) has been widely em¬ 
ployed as an objective measure of variability. If sam¬ 
ples are reasonably homogeneous with respect to age, sex, 
environment, stratigraphic position, and locality, this pro¬ 
cedure is normally justified. It frequently happens, how¬ 
ever, that individuals in a collection of fossil inverte¬ 
brates vary widely in age and that criterions are lacking 
to distinguish age groups. In these circumstances the 
form of thfik size-frequency distribution of any linear 
measurement i^ determined primarily by geologic factors 
extrinsic to the organisms, and V is unrelated to biolog¬ 
ical variability. TM3 difficulty can be largely avoided by 
considering pairs of logically related linear measurements 
and defining variability in terms of the dispersion of 
values about the curve or\relative growth calculated by 
standard regression methodsS 
Using the standard error ofsestimate as a measure of 
absolute dispersion, a coefficients of relative dispersion 
analogous to V can be obtained. Th<? use of this measure 
has been justified empirically by Klanber on populations 
of modern rattlesnakes and by the present author on 
populations of modern and fossil brachiopods. 
Purification of Human Antidextran 
Elvin A. Rabat, Columbia University College of 
Physicians and Surgeons and Presbyterian Hospital 
Antibody to dextran has been produced in human 
beings by the injection of small quantities of dextran. 
These antibodies, as formed in most individuals, have 
specific combining sites with dimensions complementary 
to terminal nonreducing chains averaging at least three 
and probably four 1 —> 6 anliydroglucopyranose units. 
Using dextran NRRL B512 containing 96 percent 1 —» 6 
and 4 percent 1 —> 3 like units as established by periodate 
oxidation [J. C. Rankin and A. Jeanes, J. Am. Clnem. Soc. 
76, 4435 (1954)], the antidextran is precipitated and 
washed with saline to remove extraneous serum protein, 
the specific precipitate is suspended in acetate buffer at 
pH 5.1 and a small quantity of dextranase from P. funi- 
culosum [kindly supplied by E. J. Hehre; see H. M. 
Tsuchiya et al., J. Bad. 64, 513 (1952)] is added. On 
warming at 37°C for 1 hr the precipitate goes almost 
completely into solution as a consequence of the digestion 
of the dextran. The resulting oligosaccharides are dialyzed 
out, and a small amount of insoluble material is removed, 
giving the final antibody solution. 
Antibody solutions contained about 90 to 95 percent of 
the nitrogen in the original specific precipitate, 80 to 90 
per cent of which was reprecipitable bv dextran. The hu¬ 
man antidextran showed a single homogeneous boundary 
in the ultracentrifuge of sedimentation constant of about 
7.0. Availability of purified antibody to an antigen com¬ 
posed of only a single sugar constituent together with the 
already described homologous oligosaccharides makes pos¬ 
sible direct physiochemical measurements on the nature 
of oligosaccharide antibody complexes. 
These studies were carried out under the William J. 
Matheson Commission and the Office of Naval Research 
[contract Nonr 266 (13)], Navy Department, Washing¬ 
ton, D.C. 
8 
Relationship of Wound Sap to the Formation 
of Crown-Gall Tumor Cells 
Richard M. Klein, Neiv York Botanical Garden 
Tap roots of carrots differ in their ability to form 
crown-gall tumor cells independent of the strain of viru¬ 
lent crown-gall bacteria used in the quantitative assays. 
The tumorous alteration of the cells is a function of their 
cell and tissue type and of their position in the root. It is 
also affected by the juices expressed from the various 
tissue areas. This wound sap is involved in the activation 
or conditioning of normal plant cells and in the synthesis, 
by crown-gall bacteria, of a tumor-inducing principle that 
alters conditioned cells into incipient tumor cells. The 
bacterial synthesis of tumor-inducing principle has been 
demonstrated in vitro. 
General Mechanisms for Enzyme Action 
Daniel E. Koshland, Jr., Brookhaven National Laboratory 
A wide variety of enzymatic reactions that perform 
diverse biological functions and present apparently dis¬ 
similar chemical reactions, nevertheless, conform to the 
stoichiometry BX + Y BY + X. Included in this group 
are the phosphatases, phosphorylases, esterases, lipases, 
proteases, glycosidases, branching enzymes, and so forth. 
It is postulated that this large group of enzymes acts 
catalytically through a few basic mechanisms. These have 
been named the single-displacement mechanism, the front¬ 
side-displacement mechanism, and the double-displace¬ 
ment mechanism. The first two of these require that both 
substrates be present simultaneously on the enzyme sur¬ 
face, whereas the latter allows a primary electron-sharing 
^ttack by the enzyme on one substrate ( BX ) even in the 
jence of the second substrate. The fundamental charac¬ 
teristic of all of these mechanisms is that they proceed by 
steps \i which a single bond is broken as a single bond 
is formed in direct analogy to the well-known displace¬ 
ment mechanisms of organic chemistry. 
With tliesfc assumptions it is possible to explain many 
observed phenomena such as the stereochemistry of the 
enzymatic reactions, the correlation of bond breaking with 
specificity, the occurrence or absence of exchange between 
BX and radioactive A, the simultaneous transferase and 
hydrolytic properties of some enzymes, and a rough ex¬ 
planation of the pH dependence of enzymatic activity. 
Specific illustrations havk been obtained with 5' nucleo¬ 
tidase, beta-glucosidase, h^etylcholinesterase, invertase, 
the phosphorylases, and glutstaiine synthetase. 
Research carried out undeX the auspices of the U.S. 
Atomic Energy Commission. \ 
\ 
Effect of Some Iron Complexes on Hydrogenase 
Alvin Krasna and David Rittenberg, 
Columbia University College of Physicians and Surgeons 
We have previously shown that the enzyme hydrogenase 
is inhibited by very low concentrations of nitric oxide 
[ Proc. Natl. Acad. Sci. 40, 225 (1954)]. In a search for 
other inhibitors, we have tested the effect of nitroprusside 
and several other pentacyano and hexacyano iron com¬ 
plexes on the hydrogenase of Proteus vulgaris. The en¬ 
zyme activity was measured by determining its catalytic 
effect on the rate either of the orthopara hydrogen conver¬ 
sion or of the exchange reaction between HDO and H 2 
o n it 1 1 
* ' f 
