REPRINTED FROM:

LIFE SCIENCES AND >
: SPACE RESEARCH II

A SESSION
OF THE FOURTH INTERNATIONAL SPACE SCIENCE SYMPOSIUM
WARSAW, JUNE 3-12, 1963

COSPAR
(COMMITTEE ON SPACE RESEARCH)

EDITORS

M.FLORKIN A.DOLLFUS

MULTIVATOR - A BIOCHEMICAL LABORATORY FOR
MARTIAN EXPERIMENTS

E. LevINTHAL, L. HUNDLEY AND J. LEDERBERG

Instrumentation Research Laboratory, Genetics Department, Stanford University

 

1964
NORTH-HOLLAND PUBLISHING COMPANY — AMSTERDAM
MULTIVATOR —- A BIOCHEMICAL LABORATORY
FOR MARTIAN EXPERIMENTS *

E. LEVINTHAL, L. HUNDLEY and J. LEDERBERG

Instrumentation Research Laboratory, Genetics Department, Stanford University

Abstract: The Multivator represents a particular choice of a detection system to serve in
the quest for signs of life on Mars. We will discuss the establishment of a critique for
making such a selection, the present restrictive limitations, and other choices that might
be made when less restrictive requirements are imposed.

The details of Multivator as a device will be discussed. The design, or minor modifica-
tions of the design, permit many different kinds of biochemical experiments to be carried
out, particularly those which can be arranged to use a photomultiplier as an output
transducer. Such experiments include fluorometry, nephelometry and scintillometry.
Our efforts have largely centered on photometric assays of fluorescence, particularly
those to determine a widely distributed enzyme, phosphatase.

Of this abstract no Russian translation has been received.

1. Introduction

We wish to discuss the philosophical concepts underlying the activities
of the exobiology group of the Instrumentation Research Laboratory and
their embodiment in the Multivator as an instrument. A modification of
the instrument has also been designed by Professor John E. Arnoldt of the
Department of Mechanical Design at Stanford University. While we clarify
those types of assays which the Multivator can carry out and give a little

* This research was supported by the National Aeronautics and Space Administration
under grant NsG 81-60.

+ Since Professor Arnold’s death in September, 1963, Professor Peter Bulkeley has assumed
the responsibility for the continued development of Multivator mechanical design.
11.5 MULTIVATOR — A BIOCHEMICAL LABORATORY 113

detail on some that have been investigated in our laboratory, a complete
presentation of these experiments is left to future publication by Dr. Elie
Shneour and others in our laboratory carrying out this research.

2. Signs of life

The field of exobiology is growing both in the scope of its problem and in
the number of scientists devoting attention to them. It therefore becomes
a matter of importance to attempt to state what these concerns are in an
organized way. This should serve to indicate where new experimental and
theoretical effort is needed and to provide a framework for the evaluation
of specific experimental possibilities as they arise.

There are an increasing number of reasons for accepting the possibility
that life in the form of reasonably complex replicating molecules is prevalent
throughout the universe. This same kind of reasoning leads one to further
suppose that intelligent forms of life are perhaps an order of magnitude less
prevalent. The quest for signs of life is the overriding immediate concern
of exobiology and leads to different experimental criteria than the concern
which follows; namely, understanding the nature of the life where signs have
been discovered. Obviously the successes and even the failures of the search
give some insights pertinent to the later analytical concern, but the instru-
mentation for search can be different than for analysis.

The difference between criteria for search and analysis is illustrated
particularly clearly in connection with the question of intelligent life in
other solar systems. At first one has to explore the complete domain of
space and possible communication links, asking only for evidence of non-
random phenomena unexplainable by known physical laws. A positive
response to such a query is not at all convincing as far as the question of the
existence of intelligent life elsewhere and says nothing of its nature. It,
however, rationalizes an enormous narrowing of the region of search and
thereby allows a corresponding increase in the analytical possibilities.

Besides the distinction between search and analysis there is another
meaningful classification. Efforts to find simple forms of life by direct
contact will, for at least the next decade, be limited to our solar system,
where there is no evidence for the existence of intelligent forms of life* on
any planet other than the earth. Thus, for the present, the search outside

* There may be other manifestations of well developed intelligence on earth than human
intelligence (i.e. Cetacea), If this is so, attempts to communicate with these species permit
the development of analytical techniques independently of search techniques, and require

no development of technology and manipulative skills on the part of the species with whom
we are attempting to communicate.
114 E. LEVINTHAL, L. HUNDLEY AND J. LEDERBERG 11.5

the solar system is necessarily limited to intelligent forms of life capable
of communication, and that within the solar system for simple forms.

The dichotomy between search and analysis enters into the search for
signs of simple forms of life within our planetary system. This is sometimes
obscured by the desire to avoid a two-step process and combine both func-
tions in one experiment. This demands an experiment that is at the same
time completely general and yet sufficiently rich in analytical details to give
an unambiguously interpretable answer. This combined goal is sought not
only because the prize of definitive proof of the existence of life elsewhere
is so great, but also because of the problems associated with contamination.
This latter concern puts a high premium on the first landed capsule giving
an unequivocal answer. It also supports the conclusion that the first capsule
should not be landed until fly-by search missions have served to narrow
either or both the geometrical volume and the volume of a matrix of possible
chemical and biological properties of living systems that must be analyzed.

It is such a matrix that needs to be established in order to have a basis
for choosing experiments for either search or analytical missions.

A possible format for such a matrix is shown in fig. 1. The concept of this
matrix will be clarified by the following comments.

(a) Here we list the elements common to terrestrial living systems.

(b) In this column the typical chemical structure and function remain the
same, but there are substitutions within a class. For example, arsenic
rather than phosphorus is used for formation of energy storage bonds; or
selenium or sulphur is used to play the role of oxygen; alpha aminobutyric
acid might be an amino acid.

{c) Under remote analogues we simply specify inorganic, organic and
macro-molecules that play functional roles similar to those found in the
terrestrial system but do not ask that they have the same structure, for
example, polyesters might be the basis of enzymatically active polymers in
place of proteins (polypeptides).

(d) Included here is DNA (as well as protein, polysaccharides, etc.) but
in the corresponding box under remote analogues we look for a more general
description of information-storing and -replicating molecules.

(e) None of these molecules affords a sufficient criteria for the presence
of life, but the accumulation of some of these molecules as components of
organisms in high local concentrations is a necessary condition.

(f) An important observation can be made concerning a property of
organic compounds in carbon-based living systems that might be extended
to all living systems. If, in a living system, there are organic compounds of
even moderate complexity, and the production of these compounds involved
11.5

MULTIVATOR — A BIOCHEMICAL LABORATORY 115

First priority scope of

<— High temperature systems —>

 

 

 

Mars investigation <——__ Interstellar space ———>
Jupiter —_>
(b) (¢) (g)
The Close Remote Non- Non-carbon Non-
earth analogues . Non- os
ay analogues aqueous liquid
system with class aqueous
generalization
Atoms C, N, O, H, Si, N, B,
P,S, Na, (a) PB...
Ix.
Molecules
in environ- H,0O, CO,,
ment in Og, CH, = (e)
organisms NH,
Organic amino acids

molecules

nucleotides ({)
metabolites

Macro- DNA, RNA
molecules proteins — (d)
poly-
saccharides
Cells Intclligent
organisms communi-
societies cation

 

 

Fig. 1.

a template which preserves their steric form, it must follow that organisms
exhibit concentrations of compounds containing asymmetric carbon atoms
and net optical activity of particular molecular species.

(g) Biochemical criteria for non-carbon, non-aqueous or certainly non-
solvated systems of life require a philosophical generalization of biological
theory not attempted at this point. The temperature of the habitat is
perhaps the single, most important parameter that defines the range of
research possibilities of biochemical constructions. Most of our thinking
is of course conditioned by the biospheric range of 250-400 °K, defined
primarily by the properties of water. For our first considerations we have
decided to concentrate our efforts on this familar territory before planning
distracting expeditions into even less accessible regions outside this range.

Having constructed such a matrix, the next task is to give probability
values to its dimensions and adjust the scales so that equal areas are equally
probable. A rational experiment or series of experiments minimizes the
116 E. LEVINTHAL, L. HUNDLEY AND J. LEDERBERG II.5

effort required to explore this surface. Our experiments are based on the
expectation that the maximum probability in our dimension lies in the
region of close analogues. While we recognize the existence of a region of
high temperature systems we have presently assigned it a negligible prob-
ability for Martian investigations.

For this procedure to be useful it is neither necessary nor presently
possible for it to be carried out with rigor: the uncertainties of the solutions
will, however, become more deductively ascertainable functions of the in-
precisions of the stated premises.

The usefulness of a matrix of the chemical and biological properties of
living system versus premises of different order of generalities would be
enhanced if there also existed an exhaustive categorization of methods of
experimental observation. Then one could be sure that one had arrived at
a logical choice of experimental objective and at the same time the optimum
technique. It is not possible at present to make such an exhaustive cate-
gorization. To be able to do so would imply that there existed a complete
theory for chemical behavior. At present there is hardly a complete descrip-
tion of the behavior of large molecules, let alone a theory. It still may be
worthwhile to attempt such an enumeration even though it be limited to
a specific region of possible experiments, For example, a listing of all the
possible ways of detecting hydrogen and its isotopes would demonstrate the
enormous range of techniques the exobiologists should keep in mind in
arriving at his research decisions. It is obvious that this is a matter of con-
tinuing concern. The means of satisfying the parameters such as incident
flux and detectivity associated with any technique are constantly changing
so that one must continually reexamine the experimental choices made. This
also makes it a matter of importance to have a framework into which one
can put new data pertinent to a particular experimental approach. That is
to say that an exhaustive organization of experimental methods coupled
with a system of scientific information storage and retrieval would be a
powerful tool in making experimental decisions and might even bring to
light completely new methods.

3. Multivator

Multivator represents an attempt to generalize the instrumental require-
ments of exobiology experiments. We are designing an instrument to meet
the constraints of Mariner-type missions and which will be able to carry
out a wide range of biochemical experiments. A simple set of performance
characteristics of this instrument becomes the interface the biologists and
II.5 MULTIVATOR ~ A BIOCHEMICAL LABORATORY 117

MULTIVATOR PARAMETERS

 

Reaction chamber

Volume w 0.1 ml
Number 15; 12 receive a soil sample of = 5 mg; 3 are blanks
Mode of operation Reagents stored dry in vessel; soil insufflated; solvent injected

(water), incubated; measurements made. Note that some reagents
may be stored as time-release capsules.

Temperature w 30°C (uncertain over what period of Martian diurnal cycle
this would be maintained).
Measurements

Types Fluorometry, colorimetry, nephelometry. Wavelength range avail-
able 300-600 mu. Lamp power may be assumed to be about 10
microwatts/mu. Detectivity is about 1] part in 10® of input light
for fluorometry or nepholometry and 1! part in 100 for colorimetry.
Different input and output filters may be chosen for each reaction

chamber.
Frequency One reading on all chambers every 15 min.
Data form Each reading quantitized into binary code of 6 bits (64 levels),

May be linear or log.

 

Fig. 2.

biochemists in our laboratory have to cope with, not the total complexities
of space missions. We hope, in time, to broaden the range of experiments
that Multivators can carry out and to design other instruments with similar
breadth of purpose.

3.1. PRESENT MULTIVATOR PARAMETERS

The interface characteristics of the current design are shown in fig, 2.
The following possible changes would require only minor modifications:
(a} non-aqueous solvents; (b) different solvents in each chamber; (c) measure-
ment of conductivity, pH, and of radioactive reaction products; and (d)
increase reading frequency to every five minutes and number of quantitiza-
tion levels to seven bits (128 levels).

It should be remembered that the system will be in dormant storage for
the six-month interval of space flight, and that all components of the ex-
periment must withstand this rest period. Temperature will be maintained
above freezing.

According to present sterilization criteria, the component of an experi-
ment must be amenable to heating at 135 °C for 24 hours. We have not yet
met this requirement for the assays we are investigating.
118 E, LEVINTHAL, L. HUNDLEY AND J. LEDERBERG II.5

3.2. Mark I MULTIVATOR DESIGN

The design and fabrication of a device of the Multivator concept has been
undertaken. Limitation on size, weight and power consumption as well as
operation independent of orientation have led to the configuration of the
present model, designated Mark I.

This particular design is based on the use of substrates tagged with a
fluorogenic indicator for the detection of enzymes characteristically pro-
duced by soil microbes. The design is in accord with the parameters indicated
in the proceeding section. It lends itself to the use of radioactive tagging
techniques with only minor modification of the existing instrument.

Mark I is designed to analyse a dust sample, collected from the planet
surface and delivered to the reaction chambers as an aerosol. A number
of sample collection methods are being investigated including vacuum
cleaning and electrostatic pick up.

The photographs depict the bench model of the device. The complete
Multivator capsule is about 25 cm long, 7 cm in diameter and weighs
380 g.

Fig. 3 shows the device as it is removed from the capsule. From the upper
left the components are: base plate, solvent chamber piston spring, solvent
chamber, lamp connector ring, photomultiplier tube shield and photomulti-
plier.

In fig. 4 the aerosol inlet tube is visible running through the center of
the solvent liquid chamber piston spring. The reaction chamber exhaust
ports can be seen cut into the lower end of the solvent chamber. The ports
in the photomultiplier housing are for the fluorescent exitation lamps, one
for each reaction chamber. These lamps will be turned on sequently to obtain
a reading from each reaction chamber.

Fig. 5 affords a view of the solvent chamber piston and the reaction
chamber cover. The guide rod for the aerosol inlet tube shut-off valve may
be seen protruding from the center of the reaction chamber cover. This
valve is spring loaded, fuse-wire actuated and closes after filling is com-
plete.

Fig. 6 shows the individual reaction chambers, the aerosol distribution
channels, the solvent loading ports in the channels, the end of the aerosol
loading tube and the air exit ports in the bottom of the reaction chamber.
This configuration with 15 reaction chambers is designed for three or four
experiments, each with several controls. The reliability and interpretibility
of any assay is obviously dependent on adequate control. As an example of
one of the controls, those chambers which do not have inlet channels to
Base plate
Lamp connector ring

Photamultipher tube shield

   

Photomultipler
Solvent cylinder
Solvent chamber piston spring

Fig. 3.

Exhaust port

we

Saivent chamber

 

Lamp cert ,
indet tube

Fig. 4.

Guide rod for aerasal inlet valve
I

 

Solvent chamber piston
Reaction chamber cover

Fig. 5.
Reaction chamber
NE Boe aay Blank chamber

a

 
   

Solvent loading,
part ,

<<

     
   

  

Aerosol Aerosol distribution
loading tube channel

Blank chamber

 
 
  

Acroso!l distribution channel

   
 

Aerosol loading tube

Reaction chamber
“Lamp pork

Fig. 6.
II.5 MULTIVATOR — A BIOCHEMICAL LABORATORY 119

the aerosol inlet tube are designed to test the stability of the fluorescent
substrate after the long space voyage.

The operation is as follows. Dust carrying the microbial sample is filtered
to ensure that there are no oversize particles and is then blown through the
inlet tube and into the reaction chambers. The dust is collected by a sticky
coating on the chamber walls, and the dust-free air exhausts out the port
in the rear of the chamber. This method of collection has been experimentally
shown to collect about 96% of the dust in the aerosol. The aerosol inlet
valve then closes, sealing off the individual reaction chamber. A small
rotation of the solvent chamber closes off the air exhaust ports and aligns
solvent filling ports in the solvent chamber with those in the inlet channel.
The solvent chamber piston then injects the solvent into the reaction cham-
bers. The substrates which have been stored dry in the chambers are dis-
solved and the reaction begins. After a preset reaction time, the exitation
lamps are then turned on sequentially, and the light level in each chamber
is detected by the photomultiplier tube. This information is then reduced
to digital form and transmitted to earth.

3.3. Mark II MuLtivaror

Further refinements of the mechanical design are being carried out by a
separate program under the direction of Professor J. Arnold in the Mechani-
cal Design Department. We are working closely with Professor Arnold’s
group in this endeavor.

For the sake of completeness we will briefly review the present state of
design of the Mark II Multivator and point out some respects in which it
differs from Mark I.

The Mark II Multivator is mechanically different but is electrically and
optically the same. Seals are replaced by bellows, and explosive charges
are used to supply the energy for mechanical operations. The overall size,
weight and shape remain about the same. The light source system, photo-
multiplier detector and assay techniques remain essentially the same. Fig. 7
gives an overall diagramatic view. The Multivator cylinder is ejected from
the capsule and erects itself by means of the adjustable legs. The collection
system is being designed as a separate element and is not shown. Dust from
this system is blown into the test chambers. The chambers and the aerosol
flow path is shown in fig. 8. The firing of a squib charge seals the air passage
as shown in the upper part of fig. 9. A second squib is fired resulting in a
piercing of membrane containing the solvent in its chambers and the release
of solvent to the assay chambers. Bottoming of the solvent chamber piston
seals the fluid injection ports. This is illustrated in the lower part of fig. 9.
120 E. LEVINTHAL, L. HUNDLEY AND J. LEDERBERG II.5

The reactions and measurements then proceed as in the previous model.
The Multivator is attached to the capsule which contains the telemetry
equipment.

~

 

 

 

 

 

 

 

 

oN
/ \
fees ~—4---—
NY
-_ _ Electronics q switching
i
Reaction chamber jt” \
sub-assembly ~~ reine ae
nt
Mi
| |
I Photo-rnultipticn
i
i li
t i
i |
iM i
.
aa -—>
fol oO}
ANG Ko .
yo t + jb Centrifugal blower
Stators AS
é Sea 44 \ 5,
peop OS
~ ~“ Ro or’ :
Collection ROTOR
aa Acrosol CN
/ ancration \\
jf mechanism \

/

oa

\

 

Fig. 7.

3.4. Marx III] MuLtivatTor

We have set as a goal the ability to measure abundances as low as 100
to 1000 bacteria in samples of 1 to 10 mg of Martian soil. We have not yet
reached this goals with the present instrument. The limitations do not seem
at present to be the sensitivity of the fluorescent assay itself but rather the
chemical noise introduced by the rate of hydrolysis of the substrate. We
need to find methods of improving this stability or of finding substrates
with similar fluorescent cross sections and improved stability. We are now
working on a Multivator operating in a manner which gets around these
difficulties, as well as on improved substrates. To the extent that the chemical
noise is distributed throughout the volume of the sample being observed,
while the fluorescent signal of biological origin is not, we can improve the
sensitivity in proportion to the reduction of the observed volume. We are
thus taking advantage of the fact that the great majority of soil bacteria
grow in colonies which are rather firmly bound to the soil particles, giving
11.5

MULTIVATOR — A BIOCHEMICAL LABORATORY

Notes ~
I. System shown open
to aerosol flow
stages [-4
(arrows indicate
path of aerosol
thru one chamber)

cs

   

 

 

 

 

 

 

7

 

Photo - multiplier tube ‘} |

 

 

     

  
   

 

 

 

Sequence of events

 

Stage Description

 

 

t

. 2
Stage *5 3
Aerosol passages sealed 4

j

5
qe 6
h SY! ~ 7

goa | ee A\8

 

 

}

Eject from capsule
rect on legs

Start collection pump

Pump air ¢ dust thru
chambers

Fire valve squib ¢ seal
dir passages

Fire soWent ¢ fill chambers}

incubate

Turn on lights
Sequentially ¢

 

note results

 

 

 

 

 

 

  

 

 

 

 

 

Stoges *6-8
Aerosol passages sealed~ solvent injected

Fig. 9.

121
122 E. LEVINTHAL, L. HUNDLEY AND J. LEDERBERG I1.5

much higher local metabolic activity in the vicinity of a particular soil
particle than the average activity of a complete sample.

With the foregoing thought in mind, the following system is being con-
sidered. The soil particles are to be more or less evenly distributed over the
surface of a substance, such asa gel, which contains the fluorescent substrate.
If the diffusion rate of the product is low compared to the rate of enzymatic
activity, there will then be found concentrations of fluorescent product in
the vicinity of bacterial colonies. This local concentration may then be
detected above the background by a suitable scanning technique. If these
local concentrations occupy a volume equal to or less than the resolution
of the system, the improvement of signal over background is in the ratio
of total enzymatically active points to total points scanned. It should be
possible to make this improvement 10? or better. As any device which is
to be sent on a mission must of necessity be power-limited, the best possible
instrumental signal-to-noise ratio for fixed power input must be determined.
It can be shown that for a given light source flux, the signal-to-noise ratio
becomes dependent only on the absolute noise of the detector, regardless
of any other considerations. This immediately indicates that the flying
spot scanning system with the low-noise photomultiplier detector is an
appropriate choice. A laboratory flying spot scanning system has been built
to test the proposed techniques. Preliminary experiments with the scanner
have been encouraging. With this laboratory instrument, further studies
of this technique will be pursued. If they are successful, design parameters of
a Mark III Multivator will be determined and a model constructed. In
addition we will consider what other new dimensions this adds to the ex-
perimental possibilities of Multivator.

4, Multivator assays

The assays we are investigating fall into two general categories. The
first detect the presence of hydrolytic enzymes by fluorometry, testing for
the catalysis of RX + H,O— RH + XOH. The bases of the test is the
release of XOH which differs from XR in being highly colored, or, better,
fluorescent. For example, R = phosphate; X = alpha naphthol to consti-
tute a fluorometric assay of phosphatase. The second uses the techniques of
membrane separation to measure the production of molecules with membrane
transmission properties differing widely from those of the substrate.

The specific permeability of membranes could thus furnish a simple
approach to biochemical analysis especially suited to the constraints of
exobiological experiments. At present we have been principally concerned
I1.5 MULTIVATOR — A BIOCHEMICAL LABORATORY 123

with the separation of gaseous or monomeric reaction products released
from larger molecules under the action of specific enzymes or of whole
organisms. The identification of the end-product molecules could be depen-
dent on the use of radioactive isotopes, for example, the detection of meta-
bolic CO, from C1 labelled nutrient substrates. Fluorometry, however,
could still be the detection mechanism if the polymer substrate were labelled
with a fluorescent molecule. Our biochemical studies have been directed
mainly towards evaluation of the phosphatase assay and achieving an under-
standing of membrane permeability which would permit exploitation of
membrane operation techniques. Surprisingly little previous work has been
done on the permeance of membranes on a wide enough dynamic range to
be pertinent to the present applications.

Phosphatase activity has been given particular attention for the following
reasons:

(a) It is widespread among terrestrial organisms.

(b) It catalyses a wide range of reactions with moderate specificity.

(c) It is involved with the unique role of phosphorus in metabolism and
energy transfer, which may very well be a universal characteristic of carbon-
based aqueous living systems.

(d) It is capable of being detected with relatively high sensitivity.
Because its design philosophy is not tied to a particular assay, Multivator
opens the possibility that scientists in laboratories other than our own could
design experiments using Multivator as an instrument, and take advantage of
the simplification of the interface problem. This is important because there can
never be a simple, unique, definitive detection experiment that can cover
all possible manifestations of life. It becomes more important after the first
sign of life has been found and a large community of scientists wish to
examine a wide range of characteristics of this life. Our laboratory looks
forwards to these co-operative possibilities and intends to explore them.
 

 

 

COSPAR BOOKS
(Committee on Space Research)

LIFE SCIENCES AND SPACE RESEARCH
VOLUME I

A Session of the Third International Space Science Symposium
Washington, D.C., 1962

Editors: R. B. LIVINGSTON, Washington
A. A. IMSHENETSKY, Moscow
G. A. DERBYSHIRE, Washington

64 x 94” xvi + 184 pages 14 plates

SPACE RESEARCH I
Proceedings of the First International Space Science Symposium
Nice, 1960
Editor: H. KALLMANN-BIJL, University of California
6} x 93” xvi + 1195 pages 62 plates

SPACE RESEARCH II
Proceedings of the Second International Space Science Symposium
Florence, 1961

Editors: H.C. VAN DE HULST, University of Leyden
Cc. DE JAGER, University of Utrecht
A. F. MOORE, The Royal Society, London

64 x 9)” xvi + 1242 pages 94 plates

SPACE RESEARCH III
Proceedings of the Third International Space Science Symposium
Washington, D.C., 1962
Editor: W. PRIESTER, University of Bonn
64 x 94” xvi -+ 1276 pages 72 plates

SPACE RESEARCH IV
Proceedings of the Fourth International Space Science Symposium
Warsaw, 1963
Editor: P. MULLER, Observatoire de Meudon
64 x 94" xvi + 998 pages 36 plates

THE USE OF ARTIFICIAL SATELLITES
FOR GEODESY

Proceedings of the International Symposium on the Use
of Artificial Satellites for Geodesy

Washington, D.C., 1962

Editor: G. VEIS, Smithsonian Institution, Astrophysical Observatory,
Cambridge, Mass.

64 x 94” xii + 424 pages 30 plates

NORTH-HOLLAND PUBLISHING COMPANY
AMSTERDAM P.O. BOX 103 THE NETHERLANDS

 

 

Sole distributors for U.S.A.: Interscience Publishers,
a division of John Wiley & Sons, Inc. — New York

Printed in the Netherlands