S-IT-A Supporting Affidavit of Daniel
OF NEW YORK
OF NEW YORK
P. NORMAN, being duly sworn, deposes and says:
February 9, 1952, I was consulted by Mr. Chester T. Lane,
attorney for Alger Hiss, with respect to the Woodstock typewriter,
N230099, placed in evidence in the Hiss trials. Mr. Lane explained
to me that he had reason to believe that the machine was not
the original machine owned by the Hisses in the early 1930s,
but a deliberately fabricated machine substituted in its place.
He said that experts who had examined specimens of typing
from the machine had expressed the view that there were definite
indications of forged typefaces on many of the letters; and
that one expert had confirmed this opinion by a microscopic
examination of the typefaces of the keys themselves. He asked
me whether my organization would be willing and able to examine
the machine in detail and advise him whether he had reasonable
grounds for his doubts as to its authenticity.
undertook to make the suggested examination, and make this
affidavit as a result of my study.
background for my conclusions, I state my qualifications,
and those of my organization:
am President of Skinner & Sherman, Inc., 246 Stuart Street,
Boston, Massachusetts, Consulting Industrial Chemists. Skinner
& Sherman, New England's oldest and largest firm in its
field, is engaged in the business of testing and analysis,
both physical and chemical, of metals, chemicals, paper, and
other materials, for the United States Armed Services, Federal,
State and Municipal Departments, and major industrial firms.
Skinner & Sherman, Inc., is a wholly owned subsidiary
of the New England Spectrochemical Laboratories, of Ipswich,
Massachusetts, a partnership, of which I am a member and the
Director of Chemical Research. The New England Spectrochemical
Laboratories are engaged in spectrographic analytical research
in industrial chemistry for a number of the major chemical
firms in the United States, and are noted for developing new
analytical methods, with respect to which they hold patents
and publish technical papers.
myself am a graduate of Boston University (A.B., A.M.) and
Harvard University (Ph.D.), and have studied spectroscopy
at Massachusetts Institute of Technology. I am a member of
the principal professional societies in my field, including
among others the American Chemical Society, the American Institute
of Chemists, and the American Optical Society. I am also a
member of the Spectrographic Analysis Committee of the American
Society for Testing Materials. I have published numerous papers
on photography, spectroscopy, and analytical chemistry.
Lane arranged to have N230099 delivered to me at the laboratories
in Ipswich on February 10, 1952. He also, at my request, furnished
me for comparison purposes with a number of other old Woodstock
typewriters with serial numbers indicating ages both greater
and less than that of N230099; and I acquired parts of one
such old Woodstock at a typewriter store in Boston.
examination of Woodstock N230099 and comparison of it with
the comparison machines point definitely to the conclusion
that Woodstock N230099 is not a machine which has worn normally
since leaving the factory, but shows positive signs of having
been deliberately altered, in that many of its types are replacements
of the originals and have been deliberately shaped.
distribution and work on the solder holding the type to the
typebars on N230099 is different in major respects from that
observed on the comparison machines.
examination of the solder holding the type to the typebars
on N230099 showed that it differed significantly in its appearance
from the solder on the comparison machines.
view of the irregular manner in which it appeared that the
type on N230099 had in general been soldered on, it seemed
reasonable to suppose that the whole soldering job was done
sloppily, and that examination would disclose an abnormal
amount of solder distributed over the sides or skirts of the
type. Small samples of metal were therefore removed from the
sides of the type just below the hardened type face, but well
away from the bottom of the skirt. The samples were taken
by a dental drill, at a position shown by the small dimple
on the middle raised type-bar in Figure A-2. The types were
all carefully cleaned with organic solvents before they were
sampled. All forty-two types on N230099 were sampled, and
enough types were sampled on the comparison machines to yield
a statistically significant comparison figure. Analyses showed
that on N230099 one out of every three types definitely had
solder distributed over the skirt, whereas on the comparison
machines the average showed solder only on one type out of
seven (in the worst case, on one type out of six; in the best,
only on one type out of ten).
conclude, therefore, that the type on N230099 was not, in
general, soldered onto the typebars at the factory or by a
professional repair man.
used to attach type on heavily soldered type-bars on N230099
is of a different kind from that used to attach type on other
typebars on that machine and type on the comparison machines.
some of the type on N230099 appears to have been soldered
in a different manner from the other type and from the type
on the comparison machines, a spectrographic analysis was
made of the solder on a number of the typebars. Samples of
solder were taken from two types on N230099 that showed the
heavy incrustations of solder, and from one type that appeared
normal. These samples were compared with three parallel samples
taken from N233954, and one sample each from N223810 and from
a typebar of the kind used on serial numbers before 220000.
Spectrographic analyses showed that the solder on the A and
T types (heavy solder blobs) of N230099 contained somewhere
between ten and fifty times as much nickel as the solder from
the J type (normal appearing) on that machine, or the solder
from the five comparison keys from other machines. In addition
to these outstanding differences in nickel content, other
less marked differences in metallic content are apparent;
but full development of these differences would call for further
extensive and expensive analyses which have not been made.
data support the conclusion that the type on N230099 showing
heavy solder blobs was probably not put on at the Woodstock
of the types on N230099 contain elements apparently not present
in type metal used on Woodstock machines until serial numbers
beginning at a substantially later date.
forty-two types on N230099 were sampled on the side as described
in Point 1 above, and the samples subjected to spectrographic
analyses. The analyses show that the different types were
not all made from the same batch of metal. These analyses
were compared with similar analyses of forty keys from the
comparison machines and three keys of old pattern but from
machines of unknown serial numbers.
Aluminum, magnesium, vanadium, zinc, antimony, and cobalt,
and their combinations, are very minor constituents whose
presence or absence appear to be good criteria for showing
whether the metals used were identical. The analyses show
that the type metal on N223810 (before N230099) and on N233954
(after N230099) do not contain the critical constituents,
nor do any of the comparison machines of earlier serial numbers;
on the other hand, later comparison machines (i.e., starting
with N256269) do show the significant criteria.
is not uniform. Nineteen of its forty-two types show these
criteria; the balance of the types do not. Of these nineteen,
thirteen are among the twenty-nine typefaces showing peculiar
solder distributions; the remaining six types showing metallurgical
deviations are distributed through the solder-questioned types
and include one type which definitely looks like a factory-soldered
is no significant difference between the metallurgy of the
types which definitely show abnormal solder distribution and
the others. As a group, however, the type on Woodstock N230099
shows significant metallurgical deviations from the type on
comparison machines made at the same period.
of the surfaces of letters on N230099 display marks of mechanical
alteration of the surfaces.
I had completed the foregoing studies, I was authorized by
Mr. Lane to detach some of the typebars, so as to make it
possible to get a microscope close enough to the type face
to permit photographs of the type face metal at sufficiently
high magnifications to show surface detail . Removal of typebars
is a non-destructive process normally used by repairmen, and
is performed by
uncoupling the key levers and withdrawing the pivot wire on
which the typebars normally pivot. Typebars so removed are
not damaged or altered in any way and can readily be replaced
in their original positions on the machine.
I removed the typebars carrying the letters A, Y and T from
Woodstock N230099 and compared the type faces microscopically
with type from other machines. On the basis of this study
I am prepared to state that abnormal tool marks can be observed
on all three type faces, but outstandingly so on the small
Although the surfaces on the three latter comparison types
show differing degrees of corrosion, the corrosion being greater
in the older machines, the corrosion or roughening is relatively
uniform along the surfaces of each letter. The surfaces on
N230099 are non-uniform in their finish, and show the appearance
of surfaces of which parts have been worked over.
to before me this
day of March, 1952.