(first published in the MUFON Symposium procedings, 2003; updated to 2005)
The February 28, 1904 sighting, by witnesses aboard the U.S.S. Supply, of three bright objects
moving rapidly through the sky is compared with what would be expected if they had been meteors.
Information provided by the Monthly Weather Review and by the ship's log is compared with theoretical
expectations for meteors. This comparison indicates that the objects were not meteors because they
(a) were too close to the earth (about a mile high or less), (b) made a large change in flight
direction, and (c) were seen for too long for the amount of sky traversed.
There are meteors, and then there are REMARKABLE meteors! But just how remarkable must a meteor be
in order to be classified as a UFO? Here is an over 100 year old sighting that stretches the meteor
explanation to its limit... and beyond!
THE REMARKABLE REPORT
The March, 1904, issue of the Monthly Weather Review (MWR) contains a letter entitled "Remarkable
Meteors." The letter was written by Lt. Frank H. Schofield, who was just beginning a long and important
career with the Navy. Some twenty years later he would be promoted to rear Admiral and ten years after
that would be the Commander in Chief of the U. S. Fleet.
The article in the MWR begins as follows: "We are pleased to report the following communication
from the U.S.S. Supply at sea off the coast of California." The letter from Schofield follows:
"1) I have the honor to report that three somewhat remarkable
meteors were observed from this ship at 6:10 AM (Greenwich Mean
Time, 3 hours, 12 minutes) February 28,1904, in latitude 35 deg. 58
min. north, longitude 128 deg, 36 min west." (Note: This is about 300
hundred miles west south-west of San Francisco.)
"2) The meteors appeared near the horizon and below the clouds,
traveling in a group from northwest by north (true) directly toward
the ship. At first their angular motion was rapid and color a
rather bright red. As they approached the ship they appeared to
soar, passing above the clouds at an elevation of about 45 deg.
After rising above the clouds their angular motion became less and
less until it ceased, when they appeared to be moving directly away
from the earth at an elevation of about 75 deg. and in direction
west-northwest (true). It was noted that the color became less
pronounced as the meteors gained in angular elevation."
"3) When sighted the largest meteor was in the lead, followed by the
second in size at a distance of less than twice the diameter of the
larger and then by the third in size at a similar distance from the
second in size. They appeared to be traveling in echelon and so
continued as long as in sight."
"4) The largest meteor had an apparent area of about six suns. It
was egg shaped, the sharper end forward. This end was jagged in
outline. The after end was regular and full in outline."
"5) The second and third meteors were round and showed no
imperfections in shape. The second meteor was estimated to be twice
the size of the sun in appearance and the third meteor about the
size of the sun."
"6) When the meteors rose there was no change in relative position;
nor was there at any time any evidence of rotation or tumbling of
the larger meteor."
"7) I estimated the clouds to be not over 1 mile high."
"8) The near approach of these meteors to the surface and the
subsequent flight away from the surface appear to be most
remarkable, especially so as their actual size could not have been
great. That they did come below the clouds and soar instead of
continuing their southeasterly course is also equally certain, as
the angular motion ceased and the color faded as they rose. The
clouds, in passing between the meteors and the ship completely
obscured the former. Blue sky could be seen in the intervals
between the clouds."
"9) The meteors were in sight for over two minutes and were
carefully observed by three people, whose accounts agree as to
details. (The Officer on Deck) sent a messenger to me who brought an
untelligible message. When I arrived on the bridge the meteors had
been obscured for about one-half minute."
The number of details in the report indicate that Lt. Schofield carefully interviewed the witnesses
shortly after the sighting and then compiled the above report. According to the report the three "meteors"
traveled "in echelon" (a steady flying formation of one in front, the next to one side and a bit behind,
and the third also to one side and a bit behind the second). They were seen underneath the 1 mile high
cloud layer and then their altitude increased so that they passed through the cloud layer and appeared to
move directly away from the ship, which might have been nearly radially away from the earth. Figure 1
illustrates, in two views (top view looking down and side view looking westward), a possible path of the
"meteors" based on the information in the report. Of course, there are numerous possible paths that one
could construct from the rather inexact information about their motion but any suggested path must be
consistent with a reasonable interpretation of the description. The sketch shows the objects approaching
the ship from the north-northwest, traveling in echelon, and below cloud level. Then the objects curved
upward and, while maintaining their echelon order, passed through the cloud layer and traveled away from
the earth. Ultimately they traveled directly away from the ship in a direction west-northwest and at an
angular elevation of about 75 degrees (nearly straight upward).
If Schofield's letter is taken as literal truth, then these "remarkable meteors" were not meteors for
three main reasons. First, according to the letter the objects were traveling for a period of time (nearly
two minutes?) below the clouds. Meteors can, of course, penetrate the atmosphere to altitudes beneath the
clouds, but by the time they reach such low altitudes they are no longer glowing and they are falling
downward, not traveling in a horizontal trajectory. Had these been meteors below the clouds they would not
have been seen as brightly glowing objects with angular sizes comparable to or greater than that of the
sun (1/2 degree). In fact, they probably would not have been seen at all (unless they hit the ship).
Second, according to the letter, before the objects disappeared their forward motion diminished and
finally ceased and they then appeared to move "directly" away from the ship. If, to an observer, the forward
motion of a typical meteor seemed to cease it would be because it burned up or because it penetrated the
atmosphere enough to slow down below the speed needed to make the atmosphere glow. In neither of these
cases would it appear that the meteor was moving away from the observer before its brightness decreased to
zero. In the rare case of a grazing or "skipping" meteor (see below), the brightness would decrease and
eventually it would stop glowing as it enters thinner atmosphere at higher altitudes before leaving the
earth entirely. In this case, also, the brightness would decrease before the forward motion apparently
ended, in contrast to the description provided by Schofield: the forward motion stopped before the brightness
decreased to zero.
The third reason is that meteors travel from near the horizon to nearly overhead in a time much shorter
than the reported time of "over two minutes," for reasons described below.
Thus, if Schofield's description is accurate these objects were not meteors. End of discussion.
(Then, what were they? Beginning of another discussion...)
But, what if Schofield's description was not accurate? What if they were not below the clouds after
all? What if their forward motion did not actually stop? What if they were seen for less than two minutes?
What if the only accurate part of the report is that the objects apparently glowed like meteors? We must
assume that this part of the report, at least, is accurate, because if the objects hadn't appeared to the
witnesses be meteors they likely would not have made the report, and certainly they wouldn't have called
the objects "remarkable meteors."
The fact is that, if we ignore (a) the apparently low flight path, (b) the apparent cessation of
forward motion and (c) the reported 2 or more minute duration, the report could be consistent with the
meteor explanation. Even the "echelon" flight pattern might fit the meteor explanation if the left-to-right
distances between the meteors were small. There have been numerous observations of meteors traveling in what
appears to the witness(es) to be a flight pattern. This occurs as a meteor enters the atmosphere and heats
and breaks into pieces. The pieces then travel together. Sometimes individual pieces burn out or deviate
from the initial trajectory or appear to tumble. However, based on the description one can assume that in
this case the individual pieces stayed together in whatever "formation" they initially acquired.
Hence, if we assume that Schofield's report contains major errors of observation (regarding the actual
flight path, altitude and duration), then these objects could have been parts of a single meteor and there
would then be only one remaining aspect of the report to explain, namely the reported large apparent size.
It does seem strange that a normal meteor or even a bolide (a large meteor, a "fireball" meteor) would be
bright enough to appear to have an angular size of "six suns." Meteors range in size from tiny (millimeters)
to perhaps centimeters in size (very rarely meters in size) and they are seen glowing (or causing the
atmosphere to glow brightly) at altitudes above 30 miles. Therefore the actual angular size of a meteor is
extremely small compared to that of the sun (the angular size of 1 cm at 30 miles distance is about
1 cm/4,800,000 cm = 0.000000208 radians, whereas the angular size of the sun is about 0.0088 radians, or about
42,000 times larger than the angular size of a 1 cm meteor at 30 miles; 1 radian = 57 degrees). Even though
the angular size of a typical meteor is extremely small, a meteor may seem sizeable to the observer because the
visually apparent size of a small bright light source, i.e., the apparent angular size, increases with the
luminous flux or "luminosity" of the source. For example, stars appear to have different sizes, with the
brighter stars - e.g., Vega or Sirius - seeming to be larger than the dimmer stars, even though the angular
sizes of all stars are vanishingly small (even smaller than the actual angular size of a meteor at 30 miles).
As another example, consider that a bright light bulb that is only a few inches in size might appear to be
larger than a foot in diameter when seen from a great distance (consider the apparent sizes of car headlights
seen in the distance). In one experiment a steel sphere 5 mm in diameter, heated to incandescence (bright white)
and viewed from about 1/2 mile away seemed to be as large as the moon's diameter (1/2 degree), i.e., as large
as a 23 ft sphere at a distance of 1/2 mile (I. L. Smith, Journal of the Royal Academy of Science, Canada,
Vol. 8, pg 109, 1914). The increase in apparent size with brightness results from optical properties of any
imaging system including the eye. Therefore, if the lead object was a meteor that seemed to be six times
larger than the sun then it must have been EXTREMELY bright! In fact, EXTREMELY, EXTREMELY bright....
Note that the witnesses claimed the color was red. (If this color was a result of temperature, then the
temperature was lower than the temperature needed for incandescent white, 5000 - 6000 K.)
Nevertheless, for "dyed in the wool" skeptics and those who would take the "easy way out," this is the
end of the discussion. The witnesses saw meteors and misperceived the altitude, the flight dynamics, the
duration and the apparent size. What they saw were three parts of a rather large meteor which had broken into
pieces before they saw it. (Airplanes are ruled out as a possible explanation... there weren't any in 1904!)
However, I am skeptical of glib explanations proposed by the skeptics. I ask, is it justifiable to simply
reject, as witness errors, the three major characteristics that make these "meteors" so remarkable? Let us
compare the reported details with what would be expected if they actually were meteors.
1) As the objects traveled toward the ship they appeared to be under the clouds.
The witnesses claimed that the objects were initially below cloud level. If they actually were meteors
then this claim must be based on a "radical misperception?" Is it reasonable to believe that such a radical
If the objects were meteors they would have been far above the clouds, probably more than 25 miles (32 km)
up (see further discussion below). Therefore the only way they could have been seen is through holes or gaps in
the cloud cover. Schofield's statement that blue sky could be seen between the clouds indicates that there must
have been enough light from the approaching sunrise to make the sky visible so that they could, in fact, see
the clouds. Therefore one may assume that the witnesses would have noticed if the objects appeared to pass
behind a cloud because the light from the objects would be temporarily blocked.
According to the report, initially the objects appeared to be "near the horizon and below the clouds."
Meteors might appear to be lower than the clouds, if they were seen under a cloud cover that ends long before it
reaches the visual horizon (the distance at which the curvature of the earth cuts off the line of sight).
In this case, with an altitude of 1 mile, the visual horizon is about 90 miles away. If, for example, the
cloud cover in the direction of the assumed meteors ended 50 miles from the ship, then the angular elevation
of the objects would have been less than arctan(1/50) = 1 degree; if the cloud cover ended 20 miles from the
ship the angular elevation would have been less than 3 degrees. The point is that at the time of the initial
sighting their angular elevation must have been only a few degrees.
As the objects approached the ship the angular elevation increased. (This is true whether the objects were
lower than cloud height or high above the clouds.) Not until the angular elevation reached about 45 degrees did
the objects appear to go above or behind the clouds. If the objects were meteors approaching the ship, then they
must have been visible continually until they reached this angular elevation, otherwise the witnesses would have
seen them going behind clouds and would have realized they were above the clouds. The cloud which first blocked
the direct view of the hypothetical meteors would have been at an angular elevation of about 45 degrees and,
hence about 1 mile horizontally from the ship. (Consider a 45 degree right triangle with the cloud at 1 mile up,
and 1 mile away (horizontally), leading to 1.4 miles radially from the ship). We can therefore conclude that, if
they were meteors, then there was a long gap or "slot" in the cloud cover that ran from about 1 mile from the ship
(horizontal measure) out to the horizon, a distance of perhaps 50 miles, in the direction of the path of the oncoming
"meteors." If the sky were nearly clear of clouds this conclusion might be acceptable. So we can ask,
what is the liklihood that there was a long clear area in the direction toward the objects?
The ship's log, a copy of which is in Figure 2, has provided useful information in this regard. According
to the log, at 0600, there were "stratus, moving from north, coverage 9." the log also shows that the sky coverage
had been in the range of 8 to 9 for the preceding two hours. Stratus clouds are horizontally layered clouds
caused by atmospheric cooling over large horizontal areas. "Coverage 9" indicates that the sky was about 90%
covered by the cloud layer which was coming from the north and moving southward. This amount of cloud coverage is
incompatible with the suggestion that there was about a many mile long gap or "slot" in the cloud cover through
which the hypothetical meteors could have been seen.
The report includes the statement that "The clouds, in passing between the meteors and the ship completely
obscured the (meteors)." But this statement, because of its location in the narrative, clearly refers to the
conditions after the angular elevation of the objects exceeded about 45 degrees.
Conclusion: the witnesses did not misperceive the altitude of the objects. The weather report indicates
that there was a cloud layer coming from the north with 90% coverage of the sky at the time of the sighting.
Hence, if the objects had been meteors it is likely that they wouldn't have been seen at all, or at most briefly
through holes in the clouds. But one can deduce from the way the report is written that the objects were seen
continually from the horizon to a distance estimated to be about 1 mile (horizontally) from the ship, a viewing
distance which is incompatible with the reported 90% cloud cover. Hence the combination of the witness statement
and the weather report provides strong evidence that the objects were, in fact, below the altitude of the
clouds. This, by itself, rules out any normal meteors.
2) When the objects reached an angular elevation of about 45 degrees they
appeared to soar upward through the clouds, then the forward motion ceased and
the color "faded."
According to Schofield's report, "As they approached the ship they appeared to soar, passing above the
clouds at an elevation of about 45 deg." and "That they did come below the clouds and soar instead of continuing
their southeasterly course is also equally certain, as the angular motion ceased and the color faded as they
rose." I assume that the phrase "color faded" means that their brightness decreased after they rose above the
If they were meteors they wouldn't have actually "soared" while the brightness decreased. Instead, they
would have maintained a nearly constant high altitude and simply disappeared behind the clouds or burned out.
Meteors would have actually gotten closer to the ship and possibly gotten brighter, had they continued in a
straight line from the angular elevation of 45 degrees to the angular elevation of 75 degrees. If they had
been meteors at, say, 30 miles altitude, then at the 45 degree elevation the radial distance from the ship would
have been about 1.4 x 30 = 42 miles. When the hypothetical meteors reached the 75 degree elevation they would
have been about 31 miles (radial distance) from the ship. Hence they would have appeared brighter, not dimmer,
and would not have given the impression of soaring upward and fading (unless, of course, they were burning out).
The actual (as opposed to observed) forward motion of meteors does not stop unless the meteor burns up.
Before the meteor burns up its brightness generally fades rapidly, so, to the observer, it would appear that
the forward motion continued until it burned out. Generally this burnout is very fast (in a fraction of a
second to a second or two). In rare cases large meteors have been known to travel in slightly curved arcs around
the earth, maintaining altitudes above 30 miles or so (about 50 km), and then to leave the atmosphere and continue
traveling away from the earth (grazing or "skipping meteors"). Observers who see such a meteor fading in
brightness as it leaves the atmosphere would see the forward motion continue at high speed. The meteor would
not appear to slow down and apparently stop its forward motion before it faded out.
However, the report indicates that the witnesses saw exactly this, the forward motion slowed to a stop
before the brightness faded out. Could they have been wrong? Does it seem reasonable that they would
misperceive the burnout of meteors at high altitude and interpret this as low altitude bright objects that
rose upward and faded? This doesn't seem likely but, before reaching a conclusion, let us consider more
carefully what they reported.
According to the report, the angular elevation increased rapidly at first and then slowed down, reaching
a maximum of about 75 degrees: "At first their angular motion was rapid and color a rather bright red. As they
approached the ship they appeared to soar, passing above the clouds at an elevation of about 45 deg. After
rising above the clouds their angular motion became less and less until it ceased, when they appeared to be
moving directly away from the earth at an elevation of about 75 deg. and in direction west-northwest (true).
It was noted that the color became less pronounced as the meteors gained in angular elevation." The report
reiterates this last observation: "That they did come below the clouds and soar instead of continuing their
southeasterly course is also equally certain, as the angular motion ceased and the color faded as they rose."
The statement that the color (brightness) faded as the objects "soared" is consistent with the inverse
square law of apparent brightness of a light: the apparent brightness (luminous intensity, actually) would
diminish in proportion to the inverse square of the distance between the ship and the objects (double the
distance, decrease the apparent brightness to 1/4, etc.).
Could the witnesses have been incorrect in their description of what happened just before the objects
faded out? Perhaps they could have been incorrect in one or two of the details, such as appearing to rise
upward through the clouds or appearing to fade out rather then suddenly disappearing. However, it is difficult
to imagine that they could have incorrectly perceived all four of these easily observed and interconnected
events near the end of the sighting: (a) the apparent rising upward through the clouds, (b) the apparent
slowing down of the rate of angular elevation increase, (c) the apparent stopping of the forward motion at 75
degree elevation, (d) and the apparent decrease in brightness, while slowing and "soaring," that gave them the
impression that the objects were moving away from the earth. These 4 events taken together are consistent
with objects which initially traveled along a horizontal track near the earth's surface (lower than cloud level)
and then abruptly turned "upward and to the right" (from the point of view of an observer traveling with the
the objects), reversing their southward motion somewhat, so that they would travel a bit northward and directly
away from the ship (and from the earth).
Conclusion: the witnesses were probably correct in their observations. Objects which make direction changes
as large as reported and travel almost radially away from the earth are not meteors.
3) The objects were observed for several minutes.
Schofield wrote, "The meteors were in sight for over two minutes and were carefully observed by three people,
whose accounts agree as to details." This is, indeed, a long duration for meteors and, when combined with the
observation that the objects were not observed as they traveled from horizon to horizon but rather as they traveled
from the horizon to nearly overhead, immediately rejects meteors. This is because meteors travel in (nearly)
straight lines through the atmosphere at such high speeds that they typically would go from horizon to horizon in
under 2 minutes (see below). Therefore they would typically go from horizon to overhead in less than 1 minute,
not "over two minutes." Again we must ask, could the report and/or the observers have been wrong? It is well
known that people often misjudge time durations when observing dynamic events. (A typical misjudged time statement
is like this: "It seemed to have lasted an eternity even though it must have been only a few seconds to a minute.")
In order to determine how accurate Schofield's report of the duration I did, in 1976, something no one had ever
done before (at least no one had ever reported doing before): I looked up the ship's log at the National Archives
to see if there might have been some mention of the sighting in the log. Figures 3 is a copy of the portion of
the ship's log that shows the events of the day before, during and after the time of the sighting.
For the date of the sighting I found the following entry at 0800 hours (8:00 AM):
"0400 - 0800 Cloudy to fair; light breeze from WSW; at 0600 wind shifted to SW;
steaming on course NE(1/4)E; executed morning orders; steam 125 lbs., revolutions
64.6. At 0610 three large bodies appeared in the sky traveling from NW(1/2)W.
The largest one egg or pear shaped, with sharp point and ragged edge to full body
aft. In size it appeared to be six times the size of the sun. The next one was
round and about twice the size of the sun. The third one was round and about
twice the size of the sun. They were in echeleon (sic) when first seen and were
below the clouds and traveling fast and rising to directly overhead. They were
dull red in color and were in sight about three minutes. The largest body would
cover all of them. When first seen were like an airship."
Clearly the log leaves out many of the details that Schofield reported, after interviewing the witnesses,
but it confirms Schofield's report of the long duration, the apparent large size, the color and the initially low
altitude (below the clouds). The phrase "rising to directly overhead" is a short version of the description of
climbing through the clouds and then the forward motion ceasing and the objects disappearing while they were
at a high angular elevevation.
Comparing the report with the log it appears that Schofield "underplayed" the duration by writing "over
two minutes" rather than "about three minutes." Perhaps he was aware that the reported duration of the observation
was extremely long for meteors and, to be as consistent as possible with normal meteors, underreported the duration.
It seems, therefore, that the rather long duration is a fact of the observation and not an error. It could mean
that the time for the objects to travel from the horizon to overhead was as long as 3 minutes. This is clearly
far beyond the expected duration for normal meteors, as I will now demonstrate by calculating the longest duration
time possible for meteors, namely the observation duration of "grazing meteors," also called "skipping meteors."
These are meteors that both enter and exit the earth's atmosphere. The question is, what would be the maximum
sighting duration if the observer were in the optimum location for the seeing the meteor from horizon to horizon?
Given this answer, then, if the observer were below the end of the visible path (as the meteor leaves the atmosphere),
as suggested by the witnesses claims that the objects faded out while nearly overhead, the observation time would
be about half that of the maximum duration.
The visible duration of a grazing meteor can be estimated by imagining a nearly straight line path of a
meteor above the curved surface of the earth. (The geometric model is illustrated in Figure 4.) The atmosphere
is curved around the earth, with the greatest density at the surface and a nearly exponential decrease in density
with increasing altitude. The meteor enters the atmosphere and approaches the earth on a tangential path. I assume
it starts to glow (actually it starts to make the air glow) at an altitude called Hi. It penetrates the atmosphere
to some minimum depth called Hm, and then, still traveling in a (nearly) straight line, increases its distance from
the earth while traveling out of the atmosphere. It stops glowing at a final height, Hf. The actual path would
actually be slightly curved around the earth, but the radius of curvature would be several times that of the radius
of the earth, so the actual path can be usefully approximated by a straight line, at least over the distance from
Hi to Hf. Use of a curved-path in this illustration would seriously complicate the analysis and provide little
further information. It may seem to be "cheating" to base calculated speeds and observation times on a straight
path, but in 1975 I exchanged letters with Dr. David Meisel, then the Director of the American Meteor Society on
this very issue. I sent him some of my calculations and he responded "Your velocity analysis using a linear
trajectory is a reasonable approximation."
[Of course, the "ultimate" curved path for a non-crashing object is parallel to the surface of the earth,
i.e., an orbit with radius of curvature =(6328 km + Hm), where Hm is the minimum height of the object in km.
No grazing meteor track would have a radius of curvature this small because if it did, it wouldn't be "grazing."
Instead, it would be captured,i.e., a satellite of earth. :) ]
First I estimate the maximum possible duration of meteor visibility. Assume the meteor started to glow
visibly at an altitude as high as Hi = 100 km (60 miles), then penetrated the atmosphere deeply to about Hm = 40 km
(25 miles) and then continued along a substantially straight path away from the earth, becoming invisible as it
reached Hf = 100 km altitude. A slightly lower altitude, 20 miles, is mentioned in the Encyclopedia Britannica
as an absolute minimum altitude for large, bright meteors called "fireballs," to be visible. (At lower altitudes
they slow and cool below visibility.) However meteors which penetrate to down to 20 miles from the surface either
explode or land. Grazing meteors don't penetrate as far. Therefore, to calculate a visible duration which is
longer than the longest likely duration (i.e., an upper bound on the visible duration) I use, as lowest altitude,
40 km. The geometry is illustrated in Figure 4 and the pertinent equations for right triangles are presented in
Figure 5 (Re = radius of the earth). With one side of a right triangle having a length of Re + Hm = 6328+40 =
6368 km and the hypotenuse having a length of Re + Hi = 6328+100 = 6428 km, we find that the path length through
the atmosphere from Hi to Hm is (6428^2 - 6368^2)^(1/2) = 876 km (544 mi). Since Hf = Hi in this model, the path
length from Hm to Hf will be the same so the total length of the straight path would be about 1,750 km (1,090 mi).
How long would it take a meteor to travel this distance? Meteors come from long distances away from the
earth and travel at high speeds in orbit around the sun. As they approach the earth they tend to speed up.
Meteors enter the atmosphere at speeds greater than 11 km/sec (6.8 mi/sec, the orbital capture speed). However,
to calculate an upper bound on the duration of visibility, i.e., a time longer than any actual duration of
visibility, I use the orbital speed. The corresponding duration of visibility would on the order of
1,750 km/11 km/sec = 160 sec. this is considerably over two minutes. However, it must be understood that this is
unreasonably long because it is based on an unrealistically low velocity and because it assumes an unrealistally
low minimum altitude with the consequent unrealistically long path through the atmosphere. Of course, in order
for an observer to see the meteor this long he would have to be below the point of closest approach and then watch
it from "birth to death," i.e., the observer must see it start to glow and then be able to watch it as it goes from
horizon to overhead to the opposite horizon. The observer would first see the meteor start to glow at an angular
elevation about 3 1/2 degrees above the horizon and a distance of about 880 km. He would last see it at the same
angular elevation and distance but above the opposite horizon. As it approached the angular elevation would
increase at first slowly and then more rapidly as the meteor passed directly overhead. The forward motion would
appear to decrease and then cease. Also, it would be brightest directly overhead, not at an angular elevation of
If the observer were below the end of the glowing path, i.e., directly below Hf, he would first see it at
his horizon, already glowing, and at a distance of about 880 km. It would appear to increase in angular elevation
while increasing in brightness (as the distance decreased) and finally fade out overhead as the meteor leaves the
atmosphere and cools. The forward motion of the meteor would appear to be greatest just as the meteor dimmed
(the exact opposite of what was reported). The duration of observation would be half that of the observer who was
beneath the center of the path, i.e., about 80 seconds.
An alternative hypothesis to calculate a long observation time, also unrealistic, is that a meteor might
follow a curved path around the earth at constant altitude at which the glow is brightest. Suppose a meteor
penetrates the earth's atmosphere to a depth of 40 km and then travels at that constant altitude of 40 km while an
observer, standing below the center of the meteor's path, watches it from horizon to horizon. In this case the
path length would be the arc of a circle of radius 6368 km. The angle formed by the arc would be 2 x inverse
cos(6328/6368) = 12.9 degrees. Hence the curved path length would be (12.9/360)x (2 pi) x 6368 = 1,430 km! This
is less than distance calculated before using the straight line approximation so the duration would be less,
namely, 130 sec. (If one assumed a circular arc at higher altitude the path length would be longer but also the
meteor brightness would be less because the atmosphere would be less dense.)
The above results are based on physically unrealistic assumptions about skipping meteors. However, the
general conclusion that the longest realistic meteor visibility would be less than 160 seconds can be checked
against the famous August, 1972, fireball meteor sighting which has been used in the past to justify the claim
that meteors can be seen for long times. This was a grazing meteor which passed through the upper atmosphere
of the earth, glowed, and then continued its travels after having suffered a severe deviation (a curvature around
the earth) from its initial track in space (an ellipse with the sun as a focal point). Calculations based on
measurements by infrared sensors on a military satellite, and presented in Sky and Telescope Magazine (July, 1974),
indicate that it became visible to the infrared sensor when it was about 76 km high, that it penetrated the
atmosphere to a depth of about 58 km, and then became invisible to the satellite again at about 102 km. The
infrared sensor was able to "see" the meteor for 101 seconds. According to the article in Sky and Telescope the
length of the path of visibility (to the satellite) was about 1,500 km (930 mi). The average speed in the path
was therefore 1500/101 = 14.85 km/sec (9.2 mi/sec), which is larger than the orbital speed used in the calculations
above. (Note: according to the Sky and Telescope article, the 1972 meteor approached the earth from behind at a
relative speed difference of about 10 km/sec, but then accelerated as it got closer and was traveling at about
14.8 km/sec on a track nearly parallel to the surface of the earth at its closest approach.)
Some people have used this 100 second observation by the satellite as evidence that meteors can be seen from
the ground for nearly as long as two minutes. However, no ground observer could have seen it that long, even if he
were in the optimum viewing location (below the center of the path of the glow) because this duration is based on
infrared measurements. Before the meteor entered the atmosphere it was very cold. The friction of moving at high
speed into the atmosphere heated up the hot gas (plasma) around the meteor, first to a temperature that was visible
to the infrared satellite sensor, and then to an even higher temperature that was visible the naked eye. As the
meteor left the atmosphere this process reversed: first the temperature of the hot gas (plasma) decreased to the
point that it was too cool to be visible to the naked eye and then it continued to decrease, eventually becoming
too cool to be detected by the infrared sensor. In other words, the meteor became visible to the eye after it was
visible to the satellite and it became invisible to the eye before it became invisible to the satellite. Hence the
length of the path of visibility (to a human) was less than 1,500 km and so the duration of sufficient brightness
to be seen with the naked eye was less than the 100 seconds "seen" by the satellite. Exactly how short the path of
visibility was I do not know. However, for the purposes of using the straight line model of the meteor path to
estimate the maximum possible duration of visibility to a human I have assumed that the meteor would have been
visible from Hi = 75 km altitude ("turn on") to Hf = 100 km ("turn off"). I have also assumed penetration to
Hm = 58 km altitude.
Using the above numbers the straight line model predicts a distance of (6403^2 - 6386^2)^(1/2) = 470 km from
Hi to Hm and (6428^2 - 6386^2)^(1/2) = 730 km from Hm to Hf. The total path length of visibility to a human was
therefore about 1160 km. If an observer had been able to watch it from the beginning to the end of its glow he
would have seen it for about 1160/14.8 = 78 seconds. A observer below the point of closest approach would have
first seen the meteor when it was about arctan(58/470) = 7 degrees above the horizon, he would have seen it pass
overhead and disappear when it was about arctan(58/740) = 1.2 degrees above the opposite horizon.
This maximum duration can be compared with the numerous witness accounts of the meteor. There were many
observations over 10 seconds but most of the observations were for 40 seconds (estimated) or less. The longest
estimated duration was reported by an astronomer and meteor expert who was on vacation in the Grand Teton area
of Wyoming. He first saw it 10 degrees above the horizon. It did not go over his head, but reached a maximum
elevation of about 20 degrees and then receded and disappeared over his local horizon. He estimated that he had
it in sight for about 1 minute (letter by astonomer Luigi Jacchia in Sky and Telescope, Oct. 1972). The famous
film of this fireball seems to last "forever," but, in fact, lasts "only" 26 seconds. No observer reported
seeing this meteor from the first moment it was visible to the last. However, it is reasonable to guess that,
had there been such an observer he could have seen the meteor for maybe 70 - 80 seconds (but certainly not as
long as 100 seconds; only the satellite could have seen it for that long). This is consistent with the 78 second
duration based on the straight line path calculation presented above. The consistency of the calculated results
with an actual observation suggests that the previous estimate of an upper bound to the maximum possible duration
of visibility, 160 seconds from horizon to horizon, is correct for the unusual or "impossible" case of a meteor
that starts to be visible at 100 km altitude, penetrates to 40 km altitude and stops being visible at 100 km.
Based on this calculation, the time it would take for a meteor to travel from the horizon to directly overhead
would be less than 80 seconds, perhaps about a minute.
In an attempt to approach the Remarkable Meteor duration of nearly three minutes to go half the way across
the horizon (or nearly six minutes to go from horizon to horizon), one might ask, couldn't a meteor "turn on" at
a higher altitude and thereby increase the path length and the calculated time? The answer to this is that the
atmosphere gets thinner with increases in altitude and above 100 km it simply to thin to for a relatively large,
relatively slow meteor (the type that would make a bright fireball) to cause a sufficient glow to be seen.
Alternatively, one might ask, why not decrease the assumed the minimum altitude and thereby increase the path
length and the duration? The answer to this has been given before: as the altitude decreases the meteor enters
continually more dense atmosphere where is slows down and actually cools off becoming invisible. Large meteors
penetrating to low altitudes explode or land.
Conclusion: the report was probably correct in stating that the duration of the sighting was "over 2 minutes."
Objects viewed for over two minutes while they travel from horizon to overhead are not meteors.
The above discussion shows that the meteor hypothesis is strained, at the very least, by the observation
that these objects appeared to travel beneath the clouds (especially considering the 90% cloud cover), strained
even further by the observation that the angular motion ceased as they appeared to travel directly away from
the ship, and finally it is strained to the breaking point by the reported duration of over 2 minutes to go from
near the horizon to overhead.
REMARKABLE SIZE, SHAPE AND SPEED
A size estimate can be calculated based on (a) the observation that the largest body had an apparent diameter
of "six suns" and (b) the assumption that the distance to the objects was about 1.5 miles when the angular size
was "six suns."
Obviously both of these numbers are disputable. The apparent angular size of the first object was probably
somewhat larger than the actual angular size because it was bright, and bright objects appear larger than they are.
The distance has to be just a guess since the report does not state at what time during the sighting the angular
size was "six suns." Here I assume that it was when the objects were closest, that is, when they were passing
through the cloud layer at about 45 degree elevation with the cloud layer at 1 mile altitude. Hence the slant
range would have been about (1 mile) x (tan 45) = 1.414 miles which I have "rounded upward" to 1.5 miles. The
angular size of the sun is about 1/2 degree or about 0.0088 radians, so the object appeared to have an angular size
of about 0.053 radians. Multiplying the angular size in radians by the distance gives the "actual" size as measured
transverse to the line of sight (the size as projected onto a plane surface that is perpendicular to the line of
sight). In this case the result is 0.053 x 1.5 x 5280 = 420 ft (128 m). The actual size could have been smaller
if the angular size were overestimated or larger if the range were underestimated. If we use the third "meteor"
as a better example, described as perfectly round and the size of a "single" sun, the calculation yields about 70 ft
(about 20 m).
As pointed out above, size estimates of bright objects could be exaggerations resulting from brightness.
This applies to sources of tiny angular size (stars, for example), but does it apply here? As anyone (with normal
vision) who has looked at the sun or moon can attest, the apparent size does not change much with brightness when
the light source is already "large" (compare the sun with and without partial obscuration by clouds or sunglasses;
its apparent size changes only a little). The "growth-with-brightness" rule has only a slight affect on light
sources which are much larger than "point" sources, i.e., light sources which are large enough (in angular size)
to be resolvable to the eye (e.g., sun, moon). The report does not say that the brightness was so great as to
make the objects difficult to look at, as the sun is difficult to look at. Under the circumstances (early morning)
we may assume that the witness' eyes were dark-adapted. Hence very bright lights could have even been painful to
look at, would have left afterimages, etc. Yet, despite the level of detail in Schofield's report, there is
nothing that indicates optical difficulty in looking at the objects. Instead, we find some details about the
shapes: "The largest meteor had an apparent area of about six suns. It was egg shaped, the sharper end forward.
This end was jagged in outline. The after end was regular and full in outline. The second and third meteors were
round and showed no imperfections in shape." That the observers were able to see details of the shapes and outlines
("jagged," "regular and full") suggests that they were easily able to look at the objects and note details of
appearance. This strongly suggests, then, that the angular sizes were not a result of brightness, but rather are
indicative of the actual physical sizes of the objects.
How fast were they actually traveling? It is not known how far away they were when first seen, but if they
were at cloud height, about 1 mile altitude, and if they were about 1 degree above the horizon, then they were
about 50 miles away (the distance of a point 1 mile up and 1 degree above the horizon) at the beginning of the
sighting. Also assume that they got to within 1.5 miles of the ship before "soaring," and that they traveled
this distance in 2 minutes = 120 sec = .0333 hours. Then their speed was approximately 0.40 mi/sec (0.64 km/sec)
or 1,440 mph (2316 km/hr). This is fast, but not meteoric speed, which is more than 10 times faster. If the
objects had been farther away when first seen and/or had reached the 45 degree elevation in less than 2 minutes
then they would have been going even faster. Had they been closer when first seen then the speed would have
been lower. Obviously this speed calculation can do no more than "get us into the ballpark."
Although there is always a problem with correctly interpreting witness statements (and removing the witness'
own interpretations so that the true description "shines through"), the analysis provides three main reasons for
rejecting the meteor explanation. These reasons are (1) the objects were seen below the 90% cloud cover, whereas
meteors would be far above clouds, (2) they made a large direction change from initially traveling along a
horizontal path toward the ship to traveling along an upward path, directly away from the ship ("soaring" at an
elevation of 75 degrees), whereas meteors would continue along in a (nearly) straight line, and (3)they were seen
for over two minutes (perhaps as long as 3 minutes), whereas meteors traversing the distance from horizon to
overheard would be seen for less than a minute. Any of these reasons would be sufficient to reject meteors or
fireballs as the explanation.
NOTE: Some people have confused the maximum duration of individual meteors and fireballs with the duration of a
particular event which involved many meteors and fireballs. In 1913 there was a "meteor procession" studied
first by Professor C. A. Chant of Canada. This procession involved many meteors traveling in an elongated group.
Calculations based on sightings suggest that the group of meteors and fireballs in this procession traveled more
than 6,000 miles around the curve of the earth. The total duration of meteor and fireball sightings at any
particular location along the path of the procession was estimated at about 5 minutes. Any given meteor in this
group could have traversed a long path partly around the earth and thus it could have been visible to a series of
observers along its path. However, any one observer at a particular location under the path saw any particular
meteor for much less than 5 minutes. According to astronomer William Pickering, writing about this event in
Popular Astronomy (Vol. 30, pg 632, 1922), individual fireballs in this procession were seen for 40 seconds or
less by observers at fixed locations.