[extropy-chat] why the vertebrate eye might not be suboptimal after all

[Damien Broderick]

My medico friend adds a more extension and persuasive rebuttal:

===============

...what would you consider to be suboptimal design of a visual organ? 
Putting the retina on backwards? Having the nerves and blood vessels 
run across the light-collecting surface of the retina? Having a blind 
spot? Having the risk of retinal detachments? Having an anterior 
chamber that can trap fluid and cause pressure damage? These are so 
obviously poor design features that I didn't think it necessary to 
defend them. What Ayoub has done, of course, is to pretend that these 
are optimal design features. There is no way they can be. No designer 
would *choose* these features. All of Ayoub's arguments are waving 
his hands. Which is why I raised that question at the start. What 
would you consider to be suboptimal design? I imagine that if you 
were to ask a thousand different camera designers how to build an 
effective eye-like structure from scratch, not a single one of them 
would incorporate any of those flaws. And I'm sorry for picking on 
the ID angle, but to me that was the only reason Ayoub came to his 
arguments in the first place. I don't know of anyone who thinks the 
vertebrate eye is an optimal design except creationists and IDists.

>>Note also that he called cephalopods "marine vertebrates".
>
>A typo, surely.

Yes, most likely a typo. But this was not an email written in haste, 
it was a referenced article that I presume he proof-read. And it's 
only 1500 words long. He then submitted it to a journal in 1996, 
admittedly a crap journal run by the ARN, but still a journal where 
presumably it was read by at least one editor. Then it was reproduced 
on the ARN's website. And from there it has been quoted many times by 
sympathetic souls (including that nicely quotable sentence) none of 
whom have corrected it or pointed out the error. So why, in ten years 
of opportunity to revise, has this sentence survived? The answer can 
only be that none of the people involved, from the author on, has 
been interested in critical appraisal of the text.

If you want a more pointed rebuttal of Ayoub's arguments, check out 
http://www-personal.umich.edu/~nesse/Articles/Nesse-MaladaptNatSel-QRB-2005.pdf 
which contains the following para:

The eye is also, however, the poster child
for the body's imperfections. It works well
when it works, but often it does not. Nearly a
third of us have hereditary nearsightedness,
and almost no one over 55 can read a phone
book unassisted (except for those who have
been nearsighted for decades!). The lovely
mechanism that regulates intraocular pressure
often fails, causing glaucoma. Then
there is the blind spot, a manifestation of the
abject design failure of nerves and vessels that
penetrate the eyeball in a bundle and spread
out along the interior surface instead of penetrating
from the outside as in the betterdesigned
cephalopod eye. Octopi not only
have a full field of vision, but they need not
worry about retinal detachment. They also
need neither the tiny jiggle of nystagmus that
minimizes the shadows cast by vessels and
nerves on the vertebrate retina nor the brain
processing mechanisms that extract the visual
signal from the nystagmus noise. In short, the
vertebrate eye is a masterpiece not of design,
but of jury-rigged compensations for a fundamentally
defective architecture.

Or there's 
http://www.pandasthumb.org/archives/2005/03/the_kraken_wake.html 
especially Ian Musgrave's contribution, which I've quoted hereafter:

The various anti-evolution commentators

[[Again, I note that of course I'm not interested in this as an 
*anti*-evolution argument; I'm a habitual contrarian wondering if 
what seems ill-adapted necessarily is. Yes, evolution retains plenty 
of solutions that barely do the job and that thereafter constrain the 
pathways that might have led to better means of doing a job. But 
we've also seen science far too quickly decide that some part of the 
body is merely vestigial and unnecessary; the whole "junk DNA" 
farrago is a prime example--some probably is conserved viral junk, 
but plenty isn't. DB]]

usually don't put in the time to understand the physiology of the 
retina in general, nor the comparative physiology of cephalopods and 
vertebrates. The RPE argument is a case in point. Most cephalopods 
have a transparent, non-pigmented epithelium that the photoreceptors 
are attached to. It is very hard to see it in the available online 
resources, but it is there. There is no reason it cannot do the 
recycling that the vertebrate RPE does (although this role is 
probably undertaken by processes from the supporting cells in the 
basement membrane). The absorption of excess light, as well as the 
nutrient function of the RPE/choroid, is undertaken by a layer of 
pigment cells and blood vessels in the basal membrane, that lies at 
the base of the light harvesting villi. You can see the blood vessels 
and pigments in <http://www.jcb.org/cgi/reprint/25/2/345>this paper 
on the octopus retina.

Tom Curtis wrote:
I notice that Ian did not deal with the nutrient issue that I 
discussed in most depth, and dealt differently (and better) with the 
over heating issue.

The nutrient issue follows directly from the heating issue. The blood 
vessels in the vertebrate eye are slightly distant from the terminal 
ends of the photoreceptor processes, and quite far from the cell 
bodies. The blood has to be pumped through at great velocity to cool 
down the choroid and to deliver nutrients to the distant retina. In 
the cephalopods the blood vessels are right next to the terminal 
parts of the photoreceptor process, the photoreceptor cell bodies and 
the pigment cells where it is needed. It is far more efficient than 
the vertebrate system for both cooling and nutrient delivery (this 
also takes care of Plunge's concerns).

Typically, most of the arguments for the "superiority" of the 
vertebrate "back-to-front" retina are irrelevant. Other commentators 
in this thread have pointed this out giving a number of examples. 
Vertebrate photoreceptors can detect a single photon, great, but so 
can cephalopod photoreceptors, and they are not covered with gunk 
that absorbs or scatters the incoming photons. Cephalopods occupy 
many niches, from shallow water tidal zones with high light 
intensities to the abyssal depths where every photon counts, some are 
ambush predators, and some are active hunting predators. Some see in 
black and white, some see in colour, some see polarized light (which 
vertebrates can't). Many have visual acuity equivalent to many 
vertebrates; cuttlefish have equivalent visual acuity to cats. All 
this without an invert retina. When Denton says

Denton wrote:
"that in redesigning from first principles an eye capable of the 
highest possible resolution (within the constraints imposed by the 
wavelength of light16) and of the highest possible sensitivity 
(capable of detecting an individual photon of light) we would end up 
recreating the vertebrate eye"

he is just plain wrong.

The pre-adaptation concept is nonsense. We are to expect that an 
intelligent designer will give the marine vertebrates, which are 
significantly more numerous in species and population than the 
terrestrial vertebrates, a poorly designed retina so that a very few 
percent of all terrestrial vertebrates can have supposedly superior 
vision? This is a definition of "good design" of which I was not 
previously aware.

[[again: ID is not the issue here]]

Tom Curtis wrote:
As to a cooling bath, a more efficient procedure (for humans at 
least) would be a UV filter in the lens, eliminating the damaging but 
useless radiation in the first place.

.

Not really, most of the damage is coming from heat generated by 
ordinary visual spectrum photons being absorbed by the pigments in 
the choroids (and superoxide radicals generated by the visual cascade 
itself), a UV filter would help, but would only reduce some of the 
damage. Cephalopods manage the cooling system better than vertebrates.

Plunge wrote:
Should we really claim that the human eye is NOT the best possible design?

Yes, because it isn't. The vertebrate eye works, and works rather 
well (one merely has to contemplate the visual acuity of the eagle to 
see that the "design" works well). But it is a suboptimal Heath 
Robinson "design" where the limitations of the original invert retina 
setup (which were irrelevant to amphioxus and the small chordates in 
which the vertebrate eye evolved) are worked around by kludges. It is 
like claiming that the misground Hubble mirror with its correcting 
lenses is the "best possible design" because it gives clear pictures.