Tom R. has a far superior answer to why the moon appears larger at the horizon.
Maybe it’s because he’s from Chicago…
But anyone who observes the moon rising or setting over a body of water KNOWS that it does not appear larger because one compares it to houses and trees around it that are smaller.
That is a very lame explanation.
For reference, the column in question is: Why does the moon appear bigger near the horizon? (January 25, 1985). Tom R.'s explanation is:
Cecil disagrees.
It’s the same effect as this optical illusion. Is this difficult to understand?
As Cecil reiterated to Tom R, try the paperclip experiment yourself, and see if the moon is the same size relative to the paperclip when it’s overhead, versus near the horizon. The size of the paperclip at arm’s length isn’t going to change. And if the size of the moon does not appear to change, then the impression that it is bigger is an optical illusion, not refraction, sorry.
The trouble with the assertion that the moon’s apparent large size at the horizon results from comparison to objects on the horizon is that it fails to explain the apparent larger size of the moon when viewed against a horizon that is featureless, like the ocean, or Lake Michigan. The same illusory size difference holds, making the comparison illusion inappropriate as an explanation for the illusory size.
That’s why I tend to prefer the explanation that deals with the idea that the sky overhead isn’t really a spherical bowl. That explanation doesn’t rely upon comparison to horizon objects.
I don’t believe it’s settled that the illusion holds even on a featureless horizon. Granted, I’ve never been one to notice the illusion much at all, horizon features or not, but I do recall reading that viewing the moon along a featureless horizon is one way to prove that the size difference is illusory.
Powers &8^]
The distance from your eyes to the paperclip is going to change.
Depends on how you hold the paperclip and position your body.
Stand up to look near the horizon. Lie on your back to look overhead.
Having lived near an ocean, I can assure you it looks bigger on the featureless horizon.
Now, perhaps it looks BIGGER bigger with trees and chit there as well. 
Really?
It doesn’t wholly matter about the features on the horizon - just that it’s the horizon. The brain’s fairly primitive comprehension of the sky’s perspective assumes up in the sky = (relatively) near and horizon = far, and since the moon’s size is constant, interprets the moon on the horizon as being huge.
I would add though that I think comparison features do enhance the effect: I’ve been in the Himalayas, and the high moon against a very tall peak does look bigger than normal; whereas the moon over a landmass on a seaward horizon looks fucking enormous.
I think it’s just great that something that was a puzzle to Ptolemy is still being thrashed around. One of these days we may actually get the reason!
It’s a refracting lens effect of looking through ever-thinning levels of atmosphere at an angle. In fact you can see the sun and moon when they are actually below the horizon and they are seen to rise faster than they really do. On rare occasions, when thesun sets it expands into a great red oval and as it goes below the horizon, the top turns green and shoots up into the sky. It’s called a green flash and I’ve seen it only once.
You would think that photographic evidence would put an end to the silliness.
No, it isn’t, as any of the experiments listed in this thread as well as the linked article will demonstrate. It’s an illusion: the moon is not actually optically larger at the horizon.
Jerseyman, the moon does get optically distorted right as it passes over the horizon. That effect is not contributing to the Moon Illusion that occurs a few degrees higher in the sky.
Contrapuntal, yes, really. To be more precise, the variance in size attributable to the distance from the eyes of the paperclip is negligible if you consistently assume the same position. Holding the arm above your head as opposed to in front of your head might have some slight change in size, but that can be negated by holding the paperclip in front of your head in both cases, and changing body position so it is oriented the same toward the moon, and different with respect to the ground. I.e. lie on your back to look overhead, stand up to look at horizon.
Besides, a bigger effect here is the ability to focus on really close and distant objects in succession. I have tried this numerous times with the moon, the hard part is comparing the objects that are at different distances and thus different focuses.
Like everything else NASA does, that photo is fake. 
I, for one, have trouble understanding the photo. It looks to me like the moon gets really big at the end.
I’m jealous; I’ve never seen a green flash and I’ve looked every time while on the shore at sunset. I understand there has to be zero haze in the atmosphere – no dust at all – so it’s most often seen far out at sea.
I’ve read that if you turn your back on a rising moon, bend over and look at between your legs, it does not look larger. I’ve never had the courage to try, though.
BigT said:
The final image of the moon has a much longer exposure than all the other images, which helps brighten the foreground and also overexposes the final image, giving the big bright washed-out glow. The other moons are short length exposures at a regular pacing to show that the visual size does not change as the moon moves higher in the sky.
I suspect that BigT knows this and was being silly. 