The Blair Cuspids: The Second Photo

Since we published the web pages on Frame LO2-61H3, a photograph showing some stratnge-looking shadows and the objects that cast them, we've learned from Fran Ridge of Lunascan that NSSDC had informed him there was a second photograph, LO2-62H3, that showed the "Cuspids" in Mare Tranquillitatus. LO2-62H3 was taken 2.2 seconds after LO2-61H3, and the Cuspids moved to oppposite edges of the camera's field of view in the time between the two photographs. Other than that, there is little difference between the viewing and lighting geometry of the photographs. The most obvious and interesting difference between the photographs is the appearance of the largest of the apparenly anomalous objects, referred to previously as Cuspid 5. Figure 1 below shows the object as it appears in the two photographs.

FIGURE 1 The image on the left is from 61H3, and the one on the right is from 62H3. Both were scanned at the same resolution 2700DPI from the 8X10 inch negatives, with one pixel equal to about 1/4 the quoted resolution for the Lunar Orbiter camera at a 3:1 contrast ratio.

In Frame 61H3, part of the object is missing where it has been cut off at a framelet boundary (misalignment of framelets was a problem with many of the Lunar Orbiter images).

In Frame 62H3, it can be seen that there is a rounded lobe on the right side of the object that matches the one on left, suggesting that the object has a mirror-image symmetry. In 62H3's complete image of the object, the boundary between its base and the lunar surface appears to be unusually linear. This apparent base line is between the two white arrows pointing toward each other in Figure 1. The apparent regularity of the objects' base and mirror-image symmetry of its top would seem to be very surprising features in a common lunar boulder.

The substantially greater width of the shadow near the Cuspid in 62H3 has little effect on the shape of the profile of the object previously derived from compressing the shadow's image from Frame 61H3 because the profile's width is influenced more by the width of the shadow farther from the object, which is the full width of the object in the first photograph and is not cut off by the framelet boundary. FIgure 2 shows the profile resulting compression of the Cuspid 5 shadow in Frame 61H3 after first moving the two adjacent framelts apart by the amount indicated to be missing in 62H3. The slope illumination angle of 6.9 degrees used to produce this profile is in the center of the range of angles previously estimated from the apparent penumbra. As will be shown on the next page, that angle is probably the best estimate of the illumination angle because it is supported by photometric analysis.

FIGURE  2  Profile of Cuspid 5 created from the shadow of the object from LO2-61H3 taking into account the area missing due to framelet overlap for a 6.9 degree sun elevation above the slope on which the shadow is cast. The image on the right is the same as the one on the left, with the additions of outlining to indicate the apparent edges of the object.

Due to some unavoidable blurring caused by the resizing algorithm used to compress the shadow in Figure 2, it is difficult to see exactly where the shadow boundary is. That might lead you to wonder whether I've been a little too imaginative in how I drew the outline of the Cuspid on the version to the right in Figure 2.  However, you should be able to satisfy yourself that the above profiles are reasonable given the sun angle of 6.9 degrees by doing the fallowing:

  1. Download this GIF file (65K) of the 2700DPI close-up of the Cuspid 5 shadow with the gap already inserted at the framelet boundary to compensate for missing part of the image due to the overlapping framelets.
  2. Measure the length of the shadow in the GIF. (You may have to adjust the brightness and contrast to see the shadow's boundary clearly).
  3. Measure the length, L, of the shadow from where it joins the object casting it to its tip.
  4. Measure the width, W, across the shadow perpendicular to its long axis at various points.You may have to increase the contrast on this unenhanced image and draw lines along the shadow boundary so that you can measure across the part missing at the gap.
  5. Multiply L/W (from Steps 3 and 4) by the tangent of 6.9 degrees (the sun angle) to get the ratio of the object's height to its width at that point.
You should find that the height-ot-width ratio for the object, is approximately 1 near the base, but about half way above the base, the ratio is more like 1.25, or a height 25% greater than width.This object is no slender "spire", but neither is it a shapeless boulder.Boulders tossed randomly seldom come to rest in such an unstable position. The outlined image of Figure 2 should be a good approximation of the object's profile as viewed from the lunar surface looking toward the sun, assuming a sun angle of 6.9 degrees. As I will discuss on the next and probably final Blair Cuspid page, photometric analysis strongly suggests that this is more than a mere guesstimate.

Too Many "Cuspids"?

Figure 3 from Frame 62H3 shows a kilometer-wide crater a few hundred meters to the north of the main group of objects. It can be seen that there are quite a few objects casting sizable shadows in and around this crater. Many of the shadows appear similar to those of some of the smaller Cuspids that William Blair noticed were positioned at the apexes of isoceles and equilateral triangles, an arrangement he thought suggested artificiality. There is nothing to suggest that any of these other objects are anything besides large rock fragments. There is certainly no indication of any pattern to their positioning. Unless some criteria can be established that distinguishes the debris strewn randomly around this crater from Cuspids of similar appearance, the significance of the Cuspids' spatial arrangement appears doubtful.

FIGURE 3 A one-Kilometer wide crater to the north of the Cuspids from LO2-62H3. This image was scanned at 337 DPI from the negative. The main group of Cuspids are a few hundred meters to the right of the position of this image.

A Baby Tooth?

The arrow in Figure 3 points to a shadow of a different shape from those of the others around this crater. The object casting this shadow is perched on the eastern edge of the crater. As shown in Figure 4 below, the shadow is needle-like, suggesting that the object casting it could be similar in shape to the largest of the Cuspids, #5, although this object is much smaller.

FIGURE 4 "Cuspid 6," the Crater Cuspid. Object and shadow on the eastern rim of the crater of Figure 3.The image was scanned from LO2-62H3 at 2700 DPI. This shadow and the part of the crater it is near were not visible in LO2-61H3 because they were off the edge of the photograph. The image has been rotated 90 degrees counter clockwise here from its orientation in Figure 3 so that the entire shadow fits on the monitor. Note the bright object at far left of the image, which is casting the shadow.

The shadow is obviously being cast downslope into the crater, which might lead the astute observer to suspect that the shadow's length has been greatly exaggerated by the slope in the way previously explained. However, the slope is much less along the crater rim in the direction the shadow points than it would be if the shadow were pointing toward the crater's center, where it is deepest. This couuld mean that the object casting the shadow might still have a significant height.

But this "maybe" and "might be" sort of argument will never lead to a reasonable estimate of what the shape of the object actually is. The shadow is too small to detect any indication of a penumbra, which was previously used for a rough estimate for the height of Cuspid 5. We have learned, however, that the Lunar Orbiter negatives do contain sufficient photometric information for estimates of the terrain slopes. A preliminary analysis not only suggests that "Cuspid 6" does have an anomalous morphology, but the same technique supports the estimate previously made for the shape of Cuspid 5. An image displaying the probable profile of this new object will be shown on the next page, but first you'll have to wade through some discussion of basic photometric principles before you get to it.

Cuspids Page 5: Photometric Analysis of the Slopes

Cuspids 3

Cuspid Page 2

Cuspid Pag 1

Back to VGL Home Page