Quote:

---

Neitz, Jay & Jacobs, Gerald H. (1986). "Polymorphism of the long-wavelength cone in normal human colour vision." Nature. 323, 623-625.

As for numbers, it won't take a long search of available literature to bring up some study, but suffice it to say that there is some variation in the color sensitivity of different people. In any case, there is enough evidence to consider the reports of sightings of color in bright emission nebulae (even red) to be possible.

---

The paper gives strong evidence that there are variations in the L-cones of humans. More precisely it shows elements that support the thesis that there are two genetic types.
A color matching experiment was carried out in which two monochromatic sources, respectively green at 546 nm, and red at 690 nm, were to be adjustend in intensity so that their mix looked the same color of a 600 nm yellow monochromatic sources.
The amount of red and green needed to match the yellow was recorded. The distribution is bimodal. In one case the proportions of green and red were 0.47 to 0.53 and in the other mode 0.54 to 0.46 (approximately). That was considered as the evidence of two types of red cones. The researchers carefully considered other possible causes producing the same results (differences in transimissivity, photopigments etc, and experiments were set up which ruled out those possibilities).
Since the gene for the L-cones stands in the X-chromosome, an additional experiment was carried out with females who, having two X-chomosomes, may happen to have only type (a), type (b) or both. The distribution confirmed the expected occurence of a, b, and a+b.
Thus researchers concluded that there appears to be two types of L-cones. They also estimated peak sensitivity wich resulted at 556 and 559 nm respectively (3 nm apart).

Thus, the sensitivity of L-cones needed a stimulus of 0.47/0.53 = 0.89 for the first type and 1.17 for the second type to see the same yellow. The overall ratio in L-cones sensitivity is therefore 1.17/0.89=1.32 which means a mere 0.30 magnitudes.

Is that "much" "enough"? If the average threshold for color detection is between 17 and 18, can we conclude that somebody can see colors at surface brightness of 20-22 based on those findings?

Edited by Mauro Da Lio (12/06/07 05:07 PM)

Post Extras