LETTER TO DR. MESELSON

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP87M00539R001001330001-0
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 18
• Document Creation Date: December 22, 2016
• Document Release Date: July 9, 2009
• Sequence Number: 1
• Publication Date: October 8, 1985
• Content Type: LETTER

Ah,
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
25X1
4e
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
Central Intelligence Agency
Dr. Matthew Meselson
Department of Biochemistry
and Molecular Biology
Harvard University
7 Divinity Avenue
Cambridge, MA 02138
0 8 OCT 1985
Dear Dr. Meselson:
Thank you for the article describing your investigations of bee feces.
While there is little doubt that some yellow spots collected in Southeast
Asia and alleged to have come from chemical weapons attack sites are in fact
bee feces, the identification of trichothecene toxins in other environmental
samples and in blood and tissues of victims cannot be denied. Our results
have been independently confirmed in multiple laboratories both domestically
and abroad. Therefore, on the basis of all the information available to the
Intelligence Community, we stand firm in our belief that trichothecene toxins
have been used in Southeast Asia and Afghanistan.
Sincerely,
/S/ John N. McMahon
John N. McMahon
Deputy Director of Central Intelligence
c gas
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
Distribution:
Orig - Addressee
1 - DDCI
1 - Executive Director
1 - Executive Registry
1 - DDI
1 - ADDI
1 - OSWR Chrono
1 - OSWR Action Group
1 - STD Chrono
1 - LSB Chrono
1-I
(2 October 85)
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
14LMOnANUUM F OK:
Director of Scientific a:;d Weapons
Research
Executive Assistant to the DDCI
STAT
have someone draft an acknowledgement
l
ease
Could you p
to this from DDCI. Assuming we disagree with conclusions,
we should say so.
Date 19 September 1985
Distribution:
0ri~L - Addressee
ER
- EA/DDCI
Approved For Release
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
DEPARTMENT OF BIOCHEMISTRY AND MOLECULAR BIOLOGY
HARVARD UNIVERSITY
7 Divinity Avenue
Cambridge, Massachusetts 02138
August 28, 1985
Mr. John McMahnn, neu~ y Director
Central Intelligence Agency
Dear Mr. McMahon,
For your possible interest, I enclose an article
describing our studies of "Yellow Rain", the material which
has been suspected of being a chemical warfare agent in
Southeast Asia. The conclusion of this work is that "Yellow
Rain" is a natural phenomenon, not a chemical warfare agent.
Sincerely yours,
ran af~ vw4
Matthew Meselson
Professor of Biochemistry
and Molecular Biology
C\ t5
I^- G*.0".
Av~~ 'i~t~T W
Ile S S
W ~
r-
1
G r, VV\?:e t aI \%^ .
STAT
STAT
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
an Article from , SCIENTIFIC
1 AMERICAN
SEPTEMBER, 1985 VOL. 253, NO. 3
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
unlikely to create new and anomalous
ones. To the extent that verbal slips
arise before items are chosen from the
lexicon, the view is viable. The other
view holds that a review process oc-
curs well after items are selected from
the lexicon but before articulation. In
other words, an impending utterance is
"edited" for linguistic integrity. This
view too is viable, particularly to the
extent that verbal slips arise after items
are chosen from the lexicon.
Which view best explains how the
system of speech production accom-
plishes its general avoidance of anom-
alous utterances? The question is not
settled, but many proponents of one
view or the other are moving toward
middle positions. In any case, it is clear
from the experimental work with slips
of the tongue that the speech-produc-
tion system concerns itself with more
than whether a given utterance ex-
presses the speaker's intended mean-
ing. It somehow acts to ensure more
generally that utterances are linguisti-
cally valid.
Another unsettled question is wheth-
er all natural verbal slips are the
result of message-option conflicts. Our
experimental work induces such con-
flict and yields abundant verbal slips,
but it does not rule out the possibility
of other types of slip. Indeed, a few
natural slips are hard to explain in
terms of message-option conflicts. It
is difficult to imagine what conflicts
could have led to slips such as "core-
gaty" (for "category"), "daygo plaints"
(for "dayglo paints") and "checking
cashes" (for "cashing checks"). Per-
haps an explanation would emerge if
one knew more about the context of
each such slip; perhaps not. Obvious-
ly, however, the possibility that unrec-
ognized sources of cognitive interfer-
ence underlie some verbal slips is
worth pursuing.
One thing that distinguishes human
language from other forms of biologi-
cal communication is that language
virtually always offers a wide variety
of alternatives for the expression of a
message. The speaker need not delib-
erate over the choices; a choice can be
made rather automatically and uttered
flawlessly, all within an instant. Stu-
dents of speech are far from under-
standing the process. To learn more
one can take advantage of what ap-
pears to be a fairly general principle:
Cognitive indecision over alternative
forms for a message sometimes results
in a slip of the tongue. By instilling
cognitive indecision, eliciting verbal
slips and examining the conditions that
precede and facilitate the slips, we
hope to learn more about the unique
qualities of human speech.
MEASUREMENT
with a QM 1 in crack propagation
?i ?' d
J.L. Humason, Technical Specialist, In his laboratory at Battelle Northwest,
monitoring a fatigue crack propagation experiment with a QM1 system which
Includes, on 3 axes, video camera and recorder, 35mm SLR and digital filar eyepiece.
Recently we had the privilege of visiting some of our customers with a
view to observing the ways in which they use our various special
systems. At Battelle Northwest we visited with Jack Humason who
was using a Questar? optical measuring system in his crack
propagation studies.
With the QM 1 system precise crack length measurements can be
made to establish crack length divided by crack opening displace-
ment gage factors. The QM 1 with a video system displayed a magnified
image of the crack on a monitor while a VCR recorded the entire test.
Tests were conducted at increasing constant load intervals, thereby
providing the crack growth rate measurements to be made for each
stress Intensity level.
The Questar image clearly showed the notch and the two mm
precracks in the metal sample. The crack progressed across the sample
as the stress was increased. At the higher stress intensities plastic
deformation occurred at the crack tip. The increasing size of the
plastically deformed region was clearly observed with the QM 1.
The Questar QM 1 system was also used to monitor the movement
of a LUDER's band migrating the length of an Iron tensile specimen.
And so for the first time, as a result of the depth of field and
resolution of the Questar optics, it was possible to see and record in
real time crack features and surface topography in detail. Tests of
this kind, whether In polymers, metals or composites, can be viewed
and taped for future study with a Questar system.
In many other applications complete systems are supplying the
solution to difficult questions of procedure, often defining areas
that previously could not be seen with any instrument. We welcome
the opportunity to discuss the hard ones with you. Call on us-
we solve problems.
QUESTAR
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0 127
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
Yellow Rain
A yellow substance found on rocks and leaves in Southeast
Asia is alleged to bean agent of chemical war. The material
is indistinguishable from the feces of indigenous honeybees
by Thomas D. Seeley, Joan W. Nowicke.
Matthew Meselson, Jeanne Guillemin and Pongthep Akratanakul
ince the late 1970's reports of
chemical warfare have emerged
from Laos and Kampuchea. The
allegations have come from refugees
and soldiers opposed to the Laotian
and Kampuchean governments that
are supported by Vietnam, and they
soon prompted investigations by the
U.S. Government. In June, 1979, two
officials of the Department of State
conducted 22 interviews in Thailand
with refugees from Laos, using a medi-
cal questionnaire prepared by U.S.
Army experts in chemical warfare.
The records of the interviews tell of
bombs and rockets delivered by air-
craft, which were said to have caused a
variety of medical symptoms and
many deaths.
The State Department investigators
were given samples of the alleged
chemical agent-pieces of vegetation
spotted with a yellow substance-
which were sent to the Army's Chemi-
cal Research and Development Center
(cRDC) in Aberdeen, Md., for chemical
analysis. Four months later Army phy-
sicians held further interviews with
refugees who said they had witnessed
chemical warfare in Laos, and they too
received samples, which were subse-
quently transmitted to the CRDC. Again
the samples were yellow spots a few
millimeters in diameter, said to have
been sprayed by an aircraft. In these
early interviews and in subsequent in-
terviews with refugees from Laos the
deposits of the presumed chemical-
warfare agent are almost always de-
scribed as yellow; they have come to
be known as yellow rain.
The diversity of the reported medi-
cal symptoms led the Army interview-
ers to conclude that several chemical
agents had probably been employed: a
nerve gas, a riot-control agent and a
chemical that causes internal bleeding.
Nevertheless, the Army's chemical
analysis of pieces of vegetation with
and without yellow spots, of yellow
materials scraped from rocks and veg-
etation and of water-more than 50
samples in all-turned up nothing. No
known chemical-warfare agent could
be detected by even the most sensitive
techniques.
On September 13, 1981, the scientif-
ic impasse seemed to be broken. Secre-
tary of State Alexander M. Haig went
before the Berlin Press Conference
with a dramatic announcement: "For
some time now, the international com-
munity has been alarmed by continu-
ing reports that the Soviet Union and
its allies have been using lethal chem-
ical weapons it Laos, Kampuchea
and Afghanistan. We now have phys-
ical evidence from Southeast Asia
which has been analyzed and found
to contain abnormally high levels of
three potent mycotoxins-poisonous
substances not indigenous to the re-
gion and which are highly toxic to
man and animals."
The physical evidence to which Sec-
retary Haig referred was a sample of
vegetation from Kampuchea, reported
to be contaminated with minute quan-
tities of three fungal toxins called
trichothecenes. The toxins were re-
ported not by the Army but by a labo-
ratory at the University of Minnesota
to which the Government had sent the
sample. Trichothecene toxins, which
are produced by species of the fun-
gal genus Fusarium, sometimes con-
taminate cereal grains, and in animals
they are reported to cause skin le-
sions, vomiting, diarrhea and gastro-
intestinal bleeding. The detection of
the toxins was the smoking gun the
State Department relied on to charge
the U.S.S.R. with waging or abetting
chemical warfare.
Such actions would constitute vio-
lations of two international arms-
control treaties: the 1925 Geneva
Protocol, which bans the use but not
the possession of chemical and bio-
logical weapons, and the 1972 Bio-
logical Weapons Convention, which
bans even the possession of biological
weapons, including toxin weapons. Al-
though Laos and Kampuchea are not
parties to the Geneva Protocol, the
U.S., the U.S.S.R. and Vietnam have
ratified the agreement. All the rele-
vant countries, including Kampuchea,
Laos, the U.S., the U.S.S.R. and Viet-
nam, are parties to the Biological
Weapons Convention.
In this context the U.S. accusation is
an extremely serious charge. The Gov-
ernment's evidence for the charge,
however, is ambiguous. In particular,
analyses by the Army have never de-
tected trichothecene toxins-or any
other chemical-warfare agents-in any
samples from sites of alleged chemi-
cal attack in Kampuchea or in Laos,
which puts the earlier reports of their
presence in serious doubt. Moreover,
our own investigations lead to an alter-
native explanation for yellow rain. We
have good physical and biological evi-
dence that yellow rain is the feces of
Southeast Asian honeybees.
T he evidence cited by the U.S. Gov-
ernment in support of its conclu-
sions can be arranged in three main
categories: the interviews with alleged
witnesses of chemical warfare, the re-
ports of trichothecene toxins in sam-
ples and the numerous descriptions
and samples of the yellow material it-
self, collected from the alleged attack
sites. A fourth category of evidence,
secret intelligence reports, is not avail-
able for independent evaluation.
The accounts of chemical warfare
come primarily from interviews with
Hmong refugees from Laos. The
Hmong are a highland people some of
whom constituted a secret army main-
tained by the U.S. Central Intelligence
Agency in Laos during the Vietnam
war. Beginning with the collapse of
U.S. support in 1975, many of the
Hmong fled Laos for Thailand; thou-
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 sands of them have subsequently set-
tled in the U.S., where they have been
accepted as political refugees. Some of
the Hmong who remained in Laos con-
tinued to resist the Laotian communist
government and the occupying Viet-
namese forces. Since 1978 Hmong ref-
ugees from Laos have reported numer-
ous chemical attacks, which allegedly
began in 1977 or before and continued
at least until early last year.
Various investigators, including rep-
resentatives of the State Department,
the Army, the Canadian government
and the United Nations as well as
American volunteers, have conducted
more than 200 interviews with alleged
witnesses, most of whom were Hmong
refugees who described suspected
chemical attacks in Southeast Asia.
Secretary of State George P. Shultz
summarized the information collect-
ed between 1979 and mid-1982 in a re-
port to Congress: "Usually the Hmong
state that aircraft or helicopters spray
a yellow rain-like material on their vil-
lages and crops." In an earlier report to
Congress, Secretary Haig refers to a
"reported symptomology of victims
which commonly included skin irrita-
tion, dizziness, nausea, bloody vomit-
ing and diarrhea and internal hem-
orrhaging." According to Secretary
Haig's report, it was this constellation
of symptoms that led in 1981 to the
tests for trichothecenes toxins.
One of us (Guillemin) has examined
the records of 217 such interviews, in-
cluding 193 that were conducted with
Hmong refugees, all done between
YELLOW SPOTS on vegetation in the forest of the Khao Yai Na-
tional Park in Thailand closely match the samples and the descrip-
tions of an alleged chemical-warfare agent known as yellow rain.
According to reports published by the U.S. Department of State
that summarize interviews with alleged witnesses to chemical war-
fare, yellow rain has been sprayed by aircraft, rockets and bombs
January, 1979, and August, 1983. The
descriptions of the color of the alleged
chemical deposits remain consistent
throughout the interviews, but the ac-
counts of the nature of the alleged at-
tacks and the medical symptoms fol-
lowing them vary widely.
More than 85 percent of the people
interviewed who specify the color of
the deposits of the alleged agent say it
was yellow. As for the method of at-
tack, about 40 percent of the respon-
dents cite a specific type of aircraft:
various propeller-driven airplanes, jets
and helicopters are all mentioned.
Only about a third of the respondents
note any particular kind of system for
disseminating the alleged chemical
agent, and again the accounts vary: the
reports cite rockets, bombs, sacks, air-
in attacks against insurgents and civilians in Laos and Kampuchea,
causing sickness and death. The authors present evidence that yel-
low rain has the same natural origin as the spots on the vegetation
in the photograph have: they are the feces of Southeast Asian hon-
eybees. The bees build as many as 100 nests in a tree and make
massive cleansing flights that leave a swath of yellow, fecal spots.
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
craft sprays and artillery fire. Notwith-
standing the many samples of yellow
rain that have come out of Laos and
Kampuchea, no chemical munition
or fragment has ever been obtained.
Sixty percent of the respondents re-
port deaths. Nevertheless, the set of
symptoms described as common in the
Haig report is rarely seen. For exam-
ple, only 8 percent of the respondents
report having bloody vomiting, 10 per-
cent report having bloody diarrhea and
21 percent report having rashes or
blisters. If all the interviews are count-
ed in which a symptom is cited, wheth-
BAN VINAI
REFUGEE CAMP
KHAO YAI
NATIONAL PARK
BANGKOK
?
er in the respondent or in others, the
frequency of each symptom is still less
than 25 percent. Only eight of the 217
people interviewed reported the three
symptoms in combination, either in
themselves or in other alleged victims.
Remarkably, the frequency of report-
ed illness is as high among respondents
who describe arriving at a site after an
attack as it is among respondents who
were allegedly exposed directly.
One cannot dismiss the accounts of
sickness and death, but one must
be aware of the ambiguities in the in-
REGIONS OF ALLEGED CHEMICAL ATTACKS are shown on the map in color. Most
of the allegations have come from Hmong refugees in the Ban Vinai refugee camp in Thai.
land. In the forest of the Khao Yai National Park three of the authors (Akratanakul, Mesel-
son and Seeley) found swaths of honeybee feces that closely resemble samples and descrip-
tions of yellow rain. The three were caught in a fecal shower in the village of Khua Moong.
terviews before interpreting them as
evidence for chemical warfare. One
main weakness in accepting the reports
in the interviews at face value is the
difficulty of distinguishing phenomena
that are merely associated with one an-
other by the respondents from phe-
nomena that are causally interrelated.
According to the interview reports,
aircraft, yellow deposits, sickness and
death were all observed on many occa-
sions. Whether some of these phenom-
ena caused the others, however, must
be open to doubt. Indeed, as we shall
discuss below, there is strong evidence
that aircraft had nothing to do with the
appearance of the yellow deposits and
that the yellow material is not harmful
to people.
There are other reasons to be skepti-
cal about the interview reports. Al-
most all the interviews were done with
refugees in camps who were selected in
advance because they said they had
been victims or witnesses of chemical
attacks. Randomly chosen refugees
from the same villages, who might
have provided cross-checks, were not
sought out. Both the respondents and
their interpreters were aware that the
purpose of the interviews was to gather
information about chemical warfare,
and no controls were employed to
make sure they did not try to accom-
modate their responses to the catego-
ries and expectations of the Western
investigators. Even the interviewers
themselves were not free of uninten-
tional bias. Their questioning often
presumed the existence of chemical
warfare, and they did not probe for
alternative explanations. Solid conclu-
sions about the occurrence of chemical
warfare cannot be drawn from the evi-
dence in the interviews.
There are several earlier cases in
which sickness and death in Southeast
Asia may have mistakenly been attrib-
uted to unusual materials from the sky.
One example is reported in a 1972
study of the effects of herbicides in
Vietnam, which was conducted for the
Department of Defense by the Nation-
al Academy of Sciences (NAS). Accord-
ing to this study, Vietnamese Monta-
gnards interviewed in refugee camps
attributed diarrhea, vomiting, skin
rash, fever, dizziness, the coughing of
blood and many deaths to the spraying
of herbicides on or near their villages.
Exposure to each of three different
herbicides was reported to cause sick-
ness and death, although none of these
herbicides would be expected to have
such severe effects. Moreover, a si-
multaneous study by the NAS showed
that lowland Vietnamese exposed to
the same herbicides rarely claimed
such serious effects. It is likely that the
reports of sickness and death among
130
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
DATE OF ATTACK
DISSEMINATION
SYSTEM
CHEMICAL
AGENT
VOMITING
DIARRHEA
CHEST
PAIN
RASHES
OR BLISTERS
BLEEDING
DEATHS
OCTOBER 1977
ROCKET
YELLOW-GREY CHEMICAL
+
25
1978
BOMB
YELLOW CLOUD
SOME
FEBRUARY 1978
UNSPECIFIED
YELLOW RESIDUE
+
+
+
500
FEBRUARY 1978
BOMB
YELLOW RESIDUE
+
7
MARCH 1978
UNSPECIFIED
YELLOW DROPS
+
5
SPRING 1978
UNSPECIFIED
YELLOW SPLOTCHES
+
+
+
+
+
0-2
MAY 1978
4 BOMBS
YELLOW DUST
18
MID-1978
6 ROCKETS,
RED GAS
SOME
JUNE 1978
ROCKET
YELLOW-RED RESIDUE
+
(+)
(+~
10-30
JUNE 1978
UNSPECIFIED
YELLOW GAS
+)
(+
90
OCTOBER 1978
8 ROCKETS
YELLOW-GREY FOG
ABOUT 150
OCTOBER 1978
4 ROCKETS
YELLOW CLOUD
8,000
OCTOBER 1978
ROCKET
RED GAS
0
NOVEMBER 1978
UNSPECIFIED
YELLOW AND BLUE GAS
+
+
80
NOVEMBER 1978
ROCKET
YELLOW GAS
29
NOVEMBER 1978
BOMB
YELLOW RAIN
2
1978 AND 1979
UNSPECIFIED
YELLOW RAIN
+
40
APRIL 1979
SACKS
RED-BROWN RESIDUE
+
4
APRIL 1979
UNSPECIFIED
YELLOW RAIN
20-30
APRIL 1979
UNSPECIFIED
YELLOW-BROWN RAIN
+
+
+
+
3
APRIL 1979
UNSPECIFIED
YELLOW SPOTS
(+)
+
(+)
2
MAY 1979
UNSPECIFIED
YELLOW RESIDUE
+
+
+
+
SOME
EARLY INTERVIEWS with Hmong refugees who said they had
witnessed chemical warfare in Laos are summarized in the table.
The interviews were done in June, 1979, by officials of the State De-
partment. Symptoms reported by the refugees are designated by a
plus sign; if the symptom was reported only in people other than the
Montagnard refugees can be traced in
part to endemic diseases and in part
to hearsay and exaggeration. Medical
symptoms and deaths attributed to
yellow rain may have a similar genesis.
In principle, chemical analysis of
samples collected from the sites of
alleged attacks could lead to firm
conclusions about the occurrence of
chemical warfare. In support of its
conclusions the State Department has
often cited reports of trichothecenes
in environmental and biomedical sam-
ples. The U.S. Army and two universi-
ty laboratories have tested a combined
total of about 100 environmental sam-
ples from alleged attack sites in Laos
and Kampuchea for trichothecenes.
Trace amounts of trichothecenes have
been reported in six of these samples,
all collected in 1981 and 1982. Fur-
thermore, the trichothecene T-2 or its
metabolite HT-2 have been reported
in the blood, urine or tissues of 20 peo-
ple, all said to have been exposed to
chemical attack in 1981, 1982 or 1983.
The Army has not examined any of
the biomedical samples, and so it can
provide no confirmation for the posi-
tive test reports. There is a serious con-
flict, however, between the Army's re-
sults for the environmental samples
and the ones cited by State.
All the positive reports for tricho-
thecenes have come from the two uni-
versity laboratories. Chester J. Miro-
cha of the University of Minnesota
respondent, the plus sign is in parentheses. The summaries are ar-
ranged chronologically according to the date of the alleged attack.
The interviewers were given samples of the alleged chemical agent
from two of the attacks, namely the fourth and the last entries in
the table. These samples were yellow Fpots on pieces of vegetation.
tested six environmental samples from
alleged attack sites, sent to him from
the CRDC by way of the Armed Services
Medical Intelligence Center at Fort
Detrick, Md. Mirocha reported that
five of the six samples were positive for
trichothecenes; they include a sample
of vegetation, a water sample and
three samples of material scraped
from rocks and vegetation. Mirocha's
analyses were the earliest ones done
for trichothecenes, and they included
the analysis on which Secretary Haig
based his charge. Joseph D. Rosen
of Rutgers University analyzed one
sample, a yellow material obtained by
the television-news organization of
the American Broadcasting Company
(ABC Television News), and he report-
ed that the sample was positive.
On the other hand, since late in 1982
more than 80 environmental samples
from alleged attack sites in Laos and
Kampuchea have been analyzed for
trichothecenes by the Army laborato-
ry, and not one of them has been
found to contain the toxins. There is
little doubt about the Army's ability
to detect trichothecenes: control sam-
ples intentionally contaminated with
trichothecenes have consistently yield-
ed positive test results. Moreover, like
the six environmental samples report-
ed as being positive for trichothecenes
by the university laboratories, most
of the samples analyzed by the Army
were vegetation, water or yellow ma-
terials scraped from rocks and leaves.
About 50 of the Army's samples were
collected in 1981 and the rest later.
One of the environmental samples
that Mirocha reported as being posi-
tive has also been tested by the CRDC.
The sample is a yellow material
scraped from rocks in Laos in 1981,
and according to Mirocha's results, it
carried the highest concentration of a
trichothecene toxin reported for any of
the samples: 143 parts per million of
the toxin T-2. About a year after Miro-
cha's analysis the CRDC tested part of
the same yellow material from which
Mirocha's sample had been taken. The
Army found no trace of T-2.
Such gross divergence in the test
results for trichothecenes-six out
of seven positive, 80 out of 80 nega-
tive-cannot plausibly be explained by
statistical errors in sampling. Instead it
raises a number of serious and still un-
answered questions: How long would
the toxins be stable and detectable in
the relevant samples? Could the posi-
tive test results somehow be caused by
experimental artifacts? Can one be as-
sured of the authenticity and the in-
tegrity of the samples? Without an-
swers to these questions the analyses
of the samples cannot be accepted as
evidence that chemical warfare was
waged with trichothecene toxins.
The third category of evidence cited
in support of the chemical-warfare
theory consists of the frequent descrip-
tions and many samples of the yellow
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
substance said by the refugees and
presumed by the Government to be a
chemical-warfare agent. According to
various Government reports, includ-
ing the reports to Congress by secretar-
ies Haig and Shultz, the substance is a
"yellow rain-like material" that falls to
form "sticky yellow spots" that soon
dry to a powder. Since 1979 dozens of
samples have been given to American,
British, Canadian and other officials.
In January, 1982, investigators at
the British Chemical Defense Estab-
lishment in Salisbury, England, found
that samples of yellow rain contain
large amounts of pollen. Soon after-
ward the same discovery was made in-
dependently by workers at Mahidol
University in Bangkok and at Agricul-
ture Canada in Ottawa. The findings
have been confirmed for, samples from
at least 30 alleged attacks in Laos and
Kampuchea, including samples given
to the U.S. investigators in 1979. Ac-
cording to Emory W. Sarver of the
CRDC, "most of the samples that are of
STERCULIA-
LIKE
MACARANGA
DENTICULATA
DILLENIA
PENTAGYNA
ELAEOCARPUS-
LIKE
ELAEOCARPUS-LIKE
yellow rain are fairly dry and they
have a high level of pollen grains in
them." To the best of our knowledge
all the samples of the yellow material
examined under the microscope have,
without exception, been found to con-
sist primarily of pollen.
The reasons for adding pollen to a
chemical-warfare agent are obscure,
and they have not been adequately ex-
plained by the proponents of the chem-
ical-warfare theory. In a briefing held
in November, 1982, and distributed by
the U.S. Information Agency, Gary
Crocker of the State Department notes
that the particles of pollen are not
windborne but rather are "the type of
thing a honeybee would take from
flowers." The physical introduction of
pollen into the alleged chemical agent
would then presumably require that
pollen gathered by honeybees be har-
vested, mixed with fungal toxins and
dispersed from weapons. Neither the
logistics of the enormous harvesting
operation required to account for the
MACARANGA
DENTICULATA
ti FRAGMENT
\ Of POLYAD
1.~?
quantities of pollen that would be
needed nor the significance of the
kinds of pollen found in the samples
is addressed by that hypothesis. We
shall have more to say on this point.
In the same 1982 briefing Sharon A.
Watson of the U.S. Armed Forces
Medical Intelligence Center suggested
a role for the pollen in the chemical
agent: "The agent, as it comes down, is
wet, and at this time the primary expo-
sure appears to be through the skin....
But as the agent dries, a secondary aer-
osol effect can be caused by kicking up
this pollen-like dust that is of a particle
size that will be retained in the bron-
chi of the lung." Watson's explanation
is faulty on two counts. First, a rela-
tively large amount of energy is need-
ed to form an aerosol from a con-
gealed deposit. Second, the samples of
yellow rain examined in the laboratory
show no tendency to disperse.
The abundance of pollen and the
lack of a plausible military explana-
POLLEN GRAINS make up the bulk of the material both in yel-
low rain deposits and in honeybee feces. The scanning electron mi-
crograph on this page shows pollen in the sample of yellow rain
from Laos, obtained by ABC Television News; the scanning elec-
tron micrograph on the opposite page shows pollen from the feces
of the honeybee Apis dorsata. Identifications and classifications of
the pollen are given in the key maps. There are three types of pollen
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 lion for its Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001.Oeentrated in rain-
yellow rain has a natural origin and led one of us (Meselson) assembled a con- like spots on rocks and leaves? The
us to obtain samples of yellow rain for ference in Cambridge. Mass.. to dis- puzzle led Ashton to approach one of
independent examination. The sam- cuss the evidence for chemical warfare us (Seeley). who is an expert on the
pies, which were made available to us
by the Canadian government, include
leaves and pebbles, each spotted with
one or more yellow deposits. They
were given to Canadian diplomatic
personnel in Thailand by Hmong refu-
gees, who said they had gathered the
samples in late March or early April,
1982, at the sites of two chemical at-
tacks in Laos. We have also examined
a sample of yellow material obtained
by ABC News and said to have been
scraped from vegetation by Hmong
soldiers at a site of an alleged chemical
attack in Laos in March, 1981. The
ABC News sample is the same one that
was analyzed by Rosen, who reported
it to contain 48 parts per million of T-
2, in addition to other trichothecenes.
In April, 1983, Carl Kayseri of the
Massachusetts Institute of Technolo-
gy, Stewart Schwartzstein of the Insti-
in Southeast Asia, with particular em-
phasis on the source and composition
of yellow rain. The conference partici-
pants included experts in anthropolo-
gy, botany, chemical warfare, chemis-
try, medicine and mycology as well as
officials from the U.S. Army and the
State Department. Peter M. S. Ashton
of Harvard University made a crucial
observation at the meeting. Prelimi-
nary analysis published by the Austra-
lian Department of Defense in Can-
berra showed that the plant families
represented in the yellow rain pollen
could be identified with certain fami-
lies strongly represented in Southeast
Asia. Ashton pointed out that the flow-
ers of these plant families are frequent-
ly visited by bees.
If yellow rain has a natural origin,
Ashton's observation raised an impor-
tant question: How could the pollen
visible in the electron micrograph of yellow rain that match grains
seen in the micrograph of bee feces: the Elaeocarpus-like grains,
which are the smallest grains and the most numerous; the Sterculia-
like grains, which have a reticulate surface, and the grains of Dii-
lenia penlagvna. which have a clumpy, irregular surface. The scan-
ning electron micrographs were made by one of the authors (No-
wicke); the magnification of each micrograph is 950 diameters.
behavior and ecology of honeybees.
Seeley noted that the Government's
description of yellow rain is an accu-
rate description of the fecal droppings
of honeybees. Like yellow rain, the fe-
ces take the form of small, yellow, pol-
len-filled spots that dry to a powder.
To test the hypothesis that yellow
rain is the feces of Asian honey-
bees we began a series of comparisons
between yellow rain and bee feces.
Fred Dyer, then a graduate student at
Princeton University, was in India at
the time studying Asian honeybees.
Dyer sent us fecal deposits of Apis ce-
rana and Apis dorsata, two of the three
Asian species of honeybees. Yellow
rain is of course most often described
as yellow, but according to witnesses,
its color also varies from pale yellow
through shades of yellowish brown
DILLENIA ELAEOCARPUS-
PENTAGYNA LIKE
STERCULIA-LIKE MACARANGA-
LIKE
ID
MACARANGA-
LIKE
ELAEOCARPUS-
LIKE
DILLENIA
PENTAGYNA ELAEOCARPUS-
P~-a LIKE /1 DIPTEROCARPUS
STERCULIA-
LIKE
STERCULIA-LIKE
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
SAMPLES OF YELLOW RAIN
FECES OF HONEYBEE
APIS DORSATA
HONEY FROM
CHIANG
POLLEN TYPE
CHIANG
KHAO YAI
KHAN
ABC
LEAF 3,
LEAF 3,
LEAF 4,
LEAF 4,
LEAF 5,
LEAF 5,
ROCK
MAI
NATIONAL
,
THAILAND
NEWS
SPOT A
SPOT B
SPOT A
SPOT B
SPOT A
SPOT B
,
THAILAND
PARK
AQ
UIFOLIACEAE
ILEX
O
0
0
0
0
0
0
COMPOSITAE
O
Q
DILLENIACEAE
DILLENIA HOOKERI
0
0
0
DILLENIA PENTAGYNA
DIPTEROCARPACEAE
DIPTEROCARPUS
O
ELAEOCARPACEAE
ELAEOCARPUS-LIKE
O'
OO
O'
EUPHORBIACEAE
MACARANGA DENTICULATA
?
Q
QQ
?'
?
MACARANGA-LIKE
O'
FAGACEAE-LIKE
OO
0
0
GRAMINEAE
ICACINACEAE
APODYTES
0
OO
0
0
O'
0
MELASTOMATACEAE
0
Qi
Q
?
0
Q
0
STERCULIACEAE
STERCULIA-LIKE
Q
OO
O MINOR COMPONENT, LESS THAN
5 PERCENT OF POLLEN GRAINS
IN SAMPLE
QQ 5-50 PERCENT OF POLLEN GRAINS
IN SAMPLE
MORE THAN 50 PERCENT OF POLLEN
GRAINS IN SAMPLE
POLLEN TYPES in samples of yellow rain from Laos and in samples of honey and honey-
bee feces from Thailand are listed in the table. Pollen types not identified in the samples are
not included in the table. The ABC News sample of yellow rain is one of six environmental
samples reported to contain trichothecenes, which are toxic substances produced by cer-
tain fungi. It was collected by a Hmong soldier at the site of an alleged chemical attack in
1981. The leaf and rock samples were collected by a group of Hmong at two sites of alleged
attack in 1982. The six leaf samples represent two spots on each of three leaves. Spot B on
leaf 3 is made up almost entirely of Fagaceae-like pollen, a type that is absent from spot A
on the same leaf. The remaining leaf spots include similar pollen types but in quite differ-
ent proportions. Such diversity of pollen from spot to spot would not be expected from a
manmade spray. The rock sample was obtained from rocks with yellow spots. The samples
of bee feces were made by pooling several spots scraped from leaves by the authors. All the
pollen types identified in yellow rain are from plant families common in Southeast Asia.
The table shows these types are also present in honey and in the feces of Apis dorsata, which
demonstrates that the pollen types in yellow rain are gathered by indigenous honeybees.
PRESENCE OF PROTEIN in pollen grains is indicated by the dye
Coomassie Brilliant Blue. Protein is digested out of the grains by
enzymes in the intestinal tract of the bee, and so pollen grains that
have passed through a bee do not stain in the presence of the dye.
The pollen grains in the photomicrograph at the left were gathered
and stored (but not eaten) by the honeybee A. dorsata; they stain a
deep blue. In the photomicrograph in the middle are pollen grains
from the feces of A. dorsala that were also treated with the dye; they
do not stain. In the photomicrograph at the right are pollen grains
in the ABC News sample of yellow rain. None of the grains in the
right photomicrograph are stained by the dye, just as one would ex-
pect if the yellow rain sample is made up of bee feces. The photo-
micrographs were made by Phillip M. Rury of Harvard University;
the magnification in all three photomicrographs is 520 diameters.
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
and reddish brown. Our own samples
of yellow rain confirm this distribution
of colors, and the distribution matches
that of the Asian honeybee feces we
received from India.
As we noted above, the average di-
ameter of the spots of yellow rain that
are reported to fall on alleged attack
sites is about three millimeters, and
they range from two to six millimeters
across. The average diameter of the
spots of yellow rain made available
to us by the Canadian government
is 3.2 millimeters, and the standard
deviation of the distribution is one
millimeter. The measurement is in-
distinguishable from the average di-
ameters we measured for the fecal de-
posits of the two honeybee species we
received from India.
Both honeybee feces and samples
of yellow rain include high concentra-
tions of pollen: approximately a mil-
lion pollen grains per milligram. The
pollen makes up about half of the vol-
time of the material, and it is held in a
coherent mass by an amorphous ma-
trix that is only partially soluble in wa-
ter. Bee hairs and bits of fungi are mi-
nor components both in samples of
yellow rain and in bee feces.
We also tested the pollen grains in
yellow rain for the presence of protein.
It is known that when pollen passes
through the digestive system of the
bee, the contents of the pollen, includ-
ing the protein, are digested out of it.
On the other hand, pollen that does not
pass through the bee retains its protein
intact, and such pollen stains a deep
blue in the presence of the dye Coo-
massie Brilliant Blue. We found that
freshly gathered pollen, pollen taken
from stores in the nests of bees and
pollen from honey all stain deeply,
which indicates a high protein content.
In contrast, most of the pollen grains
in bee feces are not stained by the dye.
What is significant, the pollen found in
our samples of yellow rain was not
stained either, just as if it had been di-
gested by bees. It would seem that in
order to accept the chemical-warfare
theory of yellow rain in the face of this
evidence one would have to imagine
an enemy so devious that its chemical
weapon is prepared by gathering pol-
len predigested by honeybees.
T he most detailed evidence for the
origin of yellow rain 'is derived by
analyzing scanning electron micro-
graphs of the pollen it contains. Pollen
grains carry the male genetic material
for all plants that reproduce from seed.
Almost all such grains have apertures,
or thin, preformed areas in the grain
wall, which allow the sperm nuclei to
be released. The size of the grain, its
surface sculpture and the shape and
number of apertures on its surface can
be highly specific for the taxon, or
plant group, from which it comes. In
combination such features can make it
possible to distinguish pollen from dif-
ferent genera within the same plant
family and, in some cases, to distin-
guish pollen from different species.
within the same genus. Since botanists
recognize more than 200,000 species
of flowering plants, pollen analysis can
sometimes give specific information
about the source of the materials that
contain the pollen.
What happens to the distinctive ap-
pearance of the pollen grains when
honeybees eat them? Typically the pol-
len is first gathered by older bees and
stored in the nest. The young adult
bees then eat it, whereupon most of the
interior of the grain, including fats and
protein, is digested by the bee. The ex-
ine, however, which is the outer shell
of the grain, is indigestible and passes
into the feces. The morphological
characteristics relied on to identify
pollen are largely unaffected by the
bee's digestive system.
One of us (Nowicke), assisted by
Janice Bittner, also of the Smithsonian
Institution, has analyzed the pollen in
the Canadian samples and in the ABC
News sample of yellow rain. The work
yields three important conclusions.
First, all plant taxa that have been
identified from the pollen in yellow
rain are common in Southeast Asia,
and their habitat is compatible with
ecological conditions near the sites of
the alleged chemical attacks. Second,
many of the pollen types found in sam-
ples of yellow rain match the types
found in honeybees, in honeybee feces
collected in Thailand and in samples
of honey collected in the mountains
of northern Thailand and along the
Thai-Laotian border. The presence of
yellow rain pollen types in bees and
bee feces validates the implication of
our protein-stain experiment: the pol-
len was indeed gathered by Southeast
Asian honeybees.
Third, no two spots of yellow rain
that have been examined, not even ad-
jacent spots on the same leaf, have the
same mixture of pollen types. Instead
there are wide variations from one
spot to another. Such diversity in the
composition of the pollen from spot to
spot is characteristic of honeybee fe-
ces, but it would not be expected from
a manmade mixture. Thus laboratory
examination of the yellow rain sam-
ples, including the ABC News sample
that reportedly contains trichothecene
toxins, has provided detailed evidence
that yellow rain is honeybee feces.
At this stage in our investigation it
still remained an open question wheth-
er honeybee defecation could account
for one of the central claims made by
the refugees, namely that the alleged
chemical-warfare agent falls in light
showers like rain. We knew that hon-
eybees do not normally defecate in
their nests; instead they do so in flight.
Indeed, beekeepers in temperate cli-
mates are familiar with the massive
defecation flights of the European
honeybee Apis mellifera on the first
warm days of spring. The behavior is
attributed to the bees' need to defecate
after the long period of enforced con-
finement during cold winter weather.
In the Tropics, however, such synchro-
nized cleansing flights were not neces-
sarily expected, and none had been re-
ported in the scientific literature.
Ater proposing that yellow rain is
the feces of Asian honeybees, we
learned of a report published in China
that described massive showers of bee
feces in northern Jiangsu Province in
September, 1976. The local popula-
tion could not determine the' cause
of the phenomenon, and so it was
brought to the attention of the geology
department at Nanjing University. Ac-
cording to the report by Zhang Zhong-
ying and his colleagues, the showers
must have been extraordinary: they
generally lasted for several minutes,
and they deposited yellow spots rich in
pollen over areas of from .2 hectare to
six hectares. The spots ranged in diam-
eter from two to six millimeters.
It is significant that no one who wit-
nessed the showers reported seeing
any bees overhead. The deposits were
only later identified as honeybee feces
through examination in the laborato-
ry. The bee species was not identified,
but the bees may not have been native:
A. mellifera had been introduced into
the region for the commercial produc-
tion of honey some time before the
cleansing flights were noticed. Never-
theless, the occurrence of massive
cleansing flights in September showed
that they need not take place only af-
ter a long period of cold weather.
To determine whether honeybees in-
digenous to the Tropics of Southeast
Asia also make such massive cleansing
flights, three of us (Akratanakul, Mes-
elson and Seeley) undertook a field
study in Thailand in March, 1984. The
bees we observed were mainly A. dor-
sata, one of the species of Asian honey-
bees and one of the two species whose
fecal deposits we had already studied
in the laboratory. The worker bees
of A. dorsata are the largest workers
among the Asian honeybees, and they
usually build nests that hang from the
limbs of tall trees in forests and vil-
lages. Typically the population of a
nest is between 30,000 and 50,000
bees, and there is often more than one
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
nest in a single bee tree. Exceptional
trees have as many as 100 nests.
W e examined 10 nesting sites of A.
dorsata: nine were in trees and
one was under the eaves of a building.
We found leaves, rocks or both spotted
with honeybee feces at all the nesting
sites. On horizontal surfaces the spots
were circular: their average diameter
was 3.2 ? .9 millimeters. The color of
the spots on young leaves ranged from
white through yellow to shades of
brown and brownish red, although yel-
low was strongly predominant. The
texture of the spots varied from waxy
to powdery. At two nesting sites the air
was particularly hot and dry, and there
were no spots on the younger leaves of
the vegetation. The absence of spots
showed that the defecation flights had
stopped at least several weeks earlier,
even though at one site the bees were
still foraging. We found recently fallen
feces, including moist, sticky deposits,
at all the other eight nesting sites.
We devoted our closest study to a
site about 800 meters above sea level
in the forest of Khao Yai National
Park, where two of us (Akratanakul
and Seeley) had investigated the be-
havior of honeybees in 1979 and 1980.
The park includes 2,200 square kilo-
meters of largely undisturbed ever-
green forest in the mountains 120 kil-
ometers northeast of Bangkok. The
nesting site was a dead dipterocarp
tree; hanging from a limb of the tree
were three 'large nests of A. dorsata
about 20 meters above the ground. A
swath of yellow-spotted vegetation
about 40 meters wide extended from
near the base of the tree out to a dis-
tance of about 160 meters. The long
axis of the swath followed a partial
opening in the forest canopy, which
ran downhill toward a valley about
five kilometers away where the bees
were probably foraging. Directly un-
der the nests and out to about 20 me-
ters from the tree the density of the
spots in the swath was low, but it then
increased sharply to a density of about
100 spots per square meter. The densi-
ty remained at roughly this level out to
about 120 meters from the tree before
it finally began to decline. Our counts
represented only two days' accumula-
tion of spots because they were done
following ap unseasonably heavy rain-
storm that had washed away most of
the older deposits.
APIS DORSATA, the giant Asian honeybee, is seen on the surface of a nest. Between 30,-
000 and 50,000 bees inhabit a typical nest, and often more than one nest is found in a tree.
YELLOW RAIN and honeybee feces leave deposits that are indistinguishable to the eye.
The spots on the leaves in the photographs at the left are the deposits made by yellow rain;
the leaves were turned over to Canadian government officials in Thailand in April, 1982,
by Hmong refugees. The refugees said they had collected the leaves at the site of a chemical
attack in Laos the month before. The spots on the leaves in the photographs at the right are
the fecal deposits of the Southeast Asian honeybees Apis cerana (upper right) and A. dorsata
(lower right). The spots vary on the average from three to five millimeters in diameter.
T he mere observation of a fecal
swath does not determine whether
the deposition takes place as a distinct
shower or as an intermittent spattering
that occurs over a relatively long time.
A shower would of course be compati-
ble with the refugees' description of
yellow rain. To resolve the question we
placed six large sheets of white paper
in various exposed places between 40
and 140 meters from the nesting site,
near the center axis of the swath. We
examined the sheets of paper periodi-
cally one morning between 7:00 A.M.
and 12:30 P.M. Sometime between 9:00
and 9:35 A.M. there was a fecal shower.
All six sheets registered the shower,
and the average density of spots was
29 per square meter. If our samples
were typical of the entire swath, the
fecal shower must have covered at
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
least 6,000 square meters and deposit-
ed more than IOO,(X)O spots.
We found that each spot on the sam-
pling sheets included copious quan-
tities of pollen. There were from
100,000 to a million pollen grains in a
typical spot, just as there are in spots of
yellow rain. We made scanning elec-
tron micrographs of the pollen in 10 of
the spots and found there were differ-
ent and sometimes widely varying
mixtures of pollen in each spot. Two of
the pollen types, which for conve-
nience we call Elaeocarpus-like and
Sterculia-like, match grains we had al-
ready seen in samples of yellow rain,
including the ABC News sample. In-
deed, most of the pollen types we iden-
tified in yellow rain are also found in
samples of honey and feces of A. dor-
sata from Thailand.
During the shower recorded in Khao
Yai National Park we were outside the
swath, and so we could only conclude
that the shower had lasted for no more
than 35 minutes. Later we were actual-
ly caught in a fecal shower. We were
visiting a region known for bee trees in
which an unusually large number of
nests are suspended. In the village
of Khua Moong, about 20 kilometers
south of Chiang Mai in Thailand. we
examined the area around two such
trees, one bearing about 30 nests and
the other more than 80, hanging from
20 to 50 meters above the ground. As
we observed the second tree through
binoculars from a clearing about 150
meters away, we saw a lightening in the
color of several nests. Hundreds of
thousands of bees were suddenly leav-
ing their nests. Moments later drops of
bee feces began falling on and around
the three members of our party. About
a dozen spots fell on each of us. We
could neither see nor hear the bees fly-
ing high above us.
The shower began at 5:17 P.m. and
lasted for approximately five minutes.
The density of the spots on the hood
and roof of our parked Land Rover
was 209 per square meter. The fresh
deposits were sticky, and they varied in
size and color much like the spots we
collected at Khao Yai National Park.
Our observations showed that showers
of honeybee feces do indeed occur in
the Tropics of Southeast Asia; more-
over, the showers and spots closely re-
semble the showers and spots said to
be caused by yellow rain.
We next sought to learn whether
the Hmong refugees might mis-
take bee feces for an agent of chemical
warfare. To investigate the question
we went to the Ban Vinai refugee
camp, where most of the interviews
with witnesses of the alleged chemical
attacks have been conducted. One of
us (Akratanakul) speaks Lao and so we
were able to question 16 groups of
people we encountered at random in
the camp.
We showed leaves spotted with the
feces of A. dorsata to each group and
asked them to identify the spots. Thir-
teen of the groups concluded they did
not know what the spots were, al-
though some people said they had seen
such spots before. One group of nine
people and one group of six told us the
spots were ke,ni, their term for the al-
leged poison. The remaining group in-
cluded three men, one of whom identi-
fied the spots as insect feces. No one
else we encountered came as close as
this man to a correct identification.
After some discussion among them-
selves, however, the three men agreed
the spots were kemi.
Our interviews with Hmong refu-
gees from Laos indicate the Hmong do
not generally recognize honeybee. fe-
ces for what they are. Moreover, some
of the Hmong identify bee feces as the
alleged agent of chemical warfare.
We conclude that yellow rain is the
feces of honeybees, not an agent
of chemical warfare. This conclusion
has emerged from many independ-
ent sources: from detailed laboratory
comparisons of samples of yellow rain
and bee feces, from field observations
of the behavior of bees and from inter-
views with Hmong refugees. Bee fecal
deposits account for all the consistent-
ly reported features of the deposits left
by yellow rain, including their color,
size and texture, their deposition in
showers and their high pollen content.
They also account for the results of
our detailed pollen analysis and other
laboratory tests.
A single clear discrepancy between
yellow rain and bee feces, such as a
mismatch between the average diame-
ters of the two classes of spots, would
naturally have forced us to reconsider
our hypothesis. No such discrepancy
has been found. In contrast, to support
the hypothesis that yellow rain is a
chemical-warfare agent one must in-
voke an entire series of unsupported
suppositions. The chemical-warfare
theory even fails to explain such strik-
ing properties of yellow rain as the
presence and the variety of pollen in
the samples.
It cannot be proved that some kind
of chemical warfare has not taken
place. The evidence for it, however,
from interviews with alleged witnes-
ses as well as from the chemical anal-
ysis of samples, is ambiguous and con-
flicting. We are reasonably confident
about the origin of the alleged chemi-
cal agent itself, the yellow rain: it is a
phenomenon of nature, not of man.
A deferise
against cancer
can be cooked up
in your kitchen.
There is evidence
that diet and cancer
are related. Some
foods may promote
cancer, while others may "
protect you from it.
Foods related to low-
ering the risk of cancer
of the larynx and esoph-
agus all have high
amounts of carotene, a
form of Vitamin A
which is in canta-
loupes, peaches, broc-
coli, spinach, all dark
green leafy vegeta-
bles, sweet potatoes,
carrots, pumpkin,
winter squash, and
tomatoes, citrus fruits and
brussels sprouts.
Foods that may help reduce the
.risk of gastrointestinal and respira-
tory tract cancer are cabbage,
broccoli, brussels sprouts, kohl-
rabi, cauliflower.
Fruits, vegetables and whole-
may help lower the
risk of colorectal
cancer.
Foods high in fats,
salt- or nitrite-cured
foods such as ham,
and fish and types of
sausages smoked by traditional
methods should be eaten in
moderation.
Be moderate in consumption
of alcohol also.
A good rule of thumb is cut
down on fat and don't be fat.
Weight reduction
may lower cancer
risk. Our 12-year
study of nearly a
million Americans
uncovered high
cancer risks partic-
ularly among people
40% or more overweight.
Now, more than ever, we
know you can cook up your
own defense against cancer. So
eat healthy and be healthy.
AMERICAN CANCER SOCIETY?
- grain cereals such as oat-
meal, bran and wheat
No one faces
137
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
 Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0
THE AMATEUR
SCIENTIST
What forces shape the behavior of water
as a drop meanders down a windowpane?
by Jearl Walker
windowpane speckled with rain
1A offers two subtle puzzles in
the physics of fluids. How do
the drops cling to the glass? When wa-
ter runs down the glass in a stream,
why does it often meander instead of
going straight?
The clinging of drops is often attrib-
uted to the surface tension of water,
yet under a common definition of sur-
face tension the drops should not cling.
Meandering is often attributed to con-
tamination on the glass, yet contami-
nation seems unlikely to account for
the normally regular pattern of a me-
andering stream.
Imagine a small drop of water on a
solid horizontal surface such as a piece
of glass. If the drop does not spread
over the surface as a film, it forms
a roughly hemispherical, slightly flat-
tened bead. The shape is determined in
part by the mutual attraction of the
water molecules; that force acts to
minimize the surface area. The sur-
face is effectively a stretched, elastic
membrane.
Usually the tendency to minimize
the surface area is described in terms
of tension on the surface of the drop.
Imagine a line running across that sur-
face. The surface tension is represent-
ed by forces that pull perpendicularly
to the line, causing it to be in a state of
tension. The surface tension of the
water is defined as the ratio of the
force pulling on one side of the line to
the length of the line.
The perimeter of the contact,.area
between the drop and the solid surface
is called the triple-phase line because
of the conjunction of water, air and
solid. The angle at which the water
touches the solid at the triple-phase
line is called the contact angle. It is
measured between the solid and a tan-
gent to the water surface. If the contact
angle is less than 90 degrees, the water
tends to spread over the solid and is
said to wet it. If the angle is more than
90 degrees, the water pulls itself into a
bead and does not wet the solid.
In the early part of the 19th century
Thomas Young, who is remembered
for his pioneering work on optical'in-
terference, stated that the size of the
contact angle is set by the tendency
toward equilibrium of three tensions
pulling on the triple-phase line. The
tension of solid and water pulls in one
direction along the interface between
the solid and the water in what is prop-
erly called an interfacial tension. The
tension of solid and air pulls in the op-
posite direction. The surface tension of
the water pulls along a tangent to the
surface of the water.
The horizontal component of the
pull from the water's surface tension
is, say, rightward and its size depends
on the cosine of the contact angle. Ac-
cording to Young's argument, if the
triple-phase line is in equilibrium, so
that the drop is neither spreading nor
contracting, the net horizontal force
on the line must vanish. A specific val-
ue of the contact angle results in a
balance of the horizontal forces on
the line.
Young's argument is simple, but
no one has verified it experimentally.
Moreover, measurements of the con-
tact angle of water on a solid surface
138
Approved For Release 2009/07/09: CIA-RDP87M00539R001001330001-0