This is the analysis of the Trent / McMinnville UFO photos by Joel Carpenter (1959-2014), as placed on his website in 2004. It was removed not long afterward from its original web hosting and was long unavailable. However, it has always been available via the Wayback Machine of the Internet Archive, although not in a way that was easy to find, to read, or to study. I have re-assembled Carpenter's original McMinnville website here. I have changed only the links; all of the content is from Joel Carpenter.

I have also placed here Carpenter's 2012 paper titled, "Preliminary Notes on the Adamski Scout Ship Photos," although it is unrelated to Trent or McMinnville. So far as I know, there is no later version of this "preliminary" paper.

Robert Sheaffer, September 2014.


The McMinnville UFO Photos

A look at the geometry of the two photographs raises questions about their authenticity.

Dr Bruce Maccabee has web-published major papers on his long-term analysis of the photos:

His work is based heavily on the use of densitometric measurements -- electronic scans of the negatives that determine the precise illumination of components the scene. Densitometry can be used as a means of estimating the UFO's size and position by comparing its measured brightness to a range of estimated values for an object with a surface of a certain color and reflectivity over a range of distances from the camera. University of Colorado UFO project investigator William Hartmann suggested in his article on the McMinnville case in the Condon Report that the high brightness of the UFO's underside indicates that it could be a large distant object. Atmospheric haze between the camera and a large distant object could scatter light to the extent that the underside shadow would be partly filled in and lightened. The brightness of the object's underside, according to this argument, strongly implies that it is not a small nearby hoax object, which could be expected to have a much darker undersurface.

Maccabee's website contains very clear, high-resolution scans of the images that show the edges of the original frames (but note that as Maccabee explains, the original negatives have been cropped at some point. The right edge of Photo 1 is significantly truncated).

The colored lines below roughly indicate the camera's horizon line based on Maccabee's sketches.

Photo 1
Photo 2

In June 1967, during his on-site investigation, Hartmann took the photo below, apparently indicating Paul Trent's camera positions, as he estimated them, with the two small markers visible on the ground at the bottom of the frame.

Using the horizon lines described above, the approximate height of Trent's and Hartmann's camera's lenses can be determined by rough triangulation with the siding of the garage and house. (While the garage siding had apparently been changed between photos, the approximate height can be determined with reasonable accuracy, allowing for error due to ground variation, grass thickness and so on. The siding boards in the Hartmann photo are said to be 6.75" high.) Note that the lens was evidently just over 36 inches above the ground in photo 1 (red lines) and somewhat over 40 inches high in photo 2 (blue lines). Compare with the horizon of Hartmann's camera (green), evidently made with his own camera at eye level in a standing position -- about 70 inches above the ground. The dog in Hartmann's photo -- probably something like a medium sized German Shepherd mix -- provides a point of comparison as well. Obviously Trent's camera was positioned just a few inches higher than its head when Photo 1 was shot.

See: Enlarged view of Hartmann photo

Dr Maccabee displays this 1948 overhead photo of the Trent property on his website (left). Enlarged view (right) shows approximate locations of Trent's camera ("X") based on Hartmann's markers; truck in driveway (circled).

Maccabee provides this sketch of Trent's estimated position relative to the house and garage. Note that he places Trent considerably closer to the house and garage than Hartmann. Maccabee argues that the sightlines from the camera to the object cross some five and a half feet past the prominent overhead wires (where a hoax model would be most likely to be suspended), but if Hartmann's camera markers are closer to Trent's actual position, the sightline crossing would occur much closer to the wires

In the images below the approximate outlines of photos 1 and 2 are projected onto Hartmann's photo to help locate the image areas. The approximate projected positions of the UFO are indicated.

Photo 1  


Photo 2  

Note that the camera was close to the ground in each of Trent's shots, well below Hartmann's standing position. The camera is only about as high as the bottom ledge of the window on the house.

To position the lens at just over 36 inches above the ground, as needed to duplicate Trent photo 1 with the Roamer in the horizontal-format position, the photographer would need to adopt the position shown below if he were framing the shot through the viewfinder on the camera body.

See: The Universal Roamer: Trent's Camera

(Note: in the photos below the camera's upward angle is slightly exaggerated.)

Camera position, Photo 1, using viewfinder on camera body Camera position, Photo 1, attempting to use waist-level viewfinder

In order to use the camera's small waist-level finder while maintaining this approximate height, a posture similar to the image above right would have to be adopted (although the individual in these photos is 6' 1'' and Trent may have been shorter. The tripod was used to steady the camera). The waist-level finder is a small, crude assembly that is suitable for roughly framing objects like buildings or standing persons within a short distance from the camera. It's necessary to position the eye within 18" from the small finder just to begin to see its .45" X .45" image. Since the shutter release is not on the camera body as on a modern 35mm SLR camera -- it's out at the end of the bellows with the lens, and has to be pushed down and to the left (see camera article) -- the posture required to use this finder is awkward and not conducive to aiming at a distant moving object. (Left index finger is on shutter release under lens in right photo).

It's been suggested that the waist-level finder is most likely the one that Trent used to target the UFO. Here are some arguments against that theory:

1] An aspirin and a postage stamp illustrate the tiny size of the waist-level finder:


The reversed scene is just visible in the finder in this photo taken at a distance of about 10" from the camera. The UFO is essentially invisible.

2] The waist-level finder works by reflecting the image from the small lens from a piece of aluminum foil (actually not a prism) and through a convex lens on top of the assembly. In addition to reversing the image right-to-left, this introduces quite a bit of "fisheye" effect, meaning that as the camera is rotated and pivoted, the image in the finder moves in unexpected ways. This makes it difficult to quickly level the camera. In addition, because of its cheap construction, the "boresight" of the waist-level finder is not exactly aligned with the axis of the main camera lens, at least on this particular example.

3] If Trent used the waist-level finder while standing, he would have been forced to frame the UFO by bracketing it between the house and garage without directly looking at it. As a matter of practicality, he would have had to look up, then down, a number of times to do this, if only to be sure the UFO was still there. Why not simply raise the camera to his eye and use the viewfinder on the camera body? It would be just as steady, if not more so, in this position.

4] The shutter release is directly beneath the waist-level finder in landscape orientation, meaning that the user has to push the lever down and to the left without rotating the camera.

There seem to be two scenarios:

1] Trent did use the waist-level finder while standing because the object was a suspended model that wasn't going anywhere;

2] He was kneeling and was using the viewfinder on the camera body. This could apply comfortably to either the nearby model or distant object cases.

See: Concept for a suspension method

Computer Model

Using actual dimensions of the Trent garage measured by David Rudiak (the house was destroyed long ago) combined with distance and size data calculated by Dr Maccabee and Brad Sparks, a first-pass computer solid model of the Trent property was generated to provide another viewpoint on the scene.

Basic model with "measuring rods" (yellow) used to establish distance relationships within the scene, assuming that Hartmann's markers are accurate. Telephone pole position and height is approximate. Garage and house are positioned accurately based on known dimensions.

Shaded view of the same model. The height and position of Trent's camera are indicated by the blue and green rods (about 37 and 42 inches respectively)

The digital model emphasizes once again just how far into the back yard Trent moved in order to photograph the ostensibly distant object.


According to reconstructions of the incident by Maccabee and Brad Sparks, Trent moved into the back yard to take the photos because his wife, who was feeding rabbits at the time, had sighted the object from there and called him urgently. Sparks argues that Trent remained rooted at the spot and must have used the waist-level finder due to his concern about stabilizing the camera. If this theory is correct, Trent actually used the viewfinder that was least likely to permit him to quickly frame the object and produce a stable exposure. Instead of moving toward the object and shooting the photos from eye level in the unobstructed front yard, he shot the two photos up, from a very low level, from the back yard. For reasons explained above, it seems likely that he actually used the viewfinder on the body of the camera while kneeling. The overall geometry of the positions and the attributes of the camera suggest that he was attempting to frame a nearby object in such a way as to maximize the amount of sky around it and enhance its apparent altitude.

See: Suggestion on the identity of the object