In the previous post, defects in the eyes were used to help identify a homozygous merle. That is often a strong indicator, but there is one situation where that is not always helpful. Collies, and some of the closely related breeds, have a number of issues with their eyes that are unrelated to the merle gene. Compare the normal eyes of the merle Shetland Sheepdog, above, to the eyes on the merle Rough Collie below.
Like the Great Dane in the previous post, this dog has an eye that appears to be too small and set incorrectly. Her left eye, which is blue, also has a distorted pupil similar to the one seen on the Dane. It is more obvious when viewed from the front.
Although it might look like she is looking to the side in this shot, her pupil actually skewed over toward that corner, giving her a cross-eyed look.
She is not a double-merle. Her pattern is typical for a single merle with moderate white irish patterning. Whatever is wrong with her eyes, it is probably separate from her merle coloring. Her one blue eye makes the problem more noticeable, but chances are she would have had issues whatever color she happened to be.
And that is why distortions in the eyes on collie breeds are not necessarily proof that the dog is homozygous for merle. Fortunately for identification purposes, most double-merle Collies are quite dramatically white so they are unlikely to be mistaken for a heterozygous merle.
But dogs like this one also point to the reason why using double-merles in Collie breeding programs is a bad idea, even when someone else made the ethical compromises necessary to create the dog in the first place. Because two copies of the merle mutation damages the eyes, there is no way to know if a homozygous merle breeding animal had eye problems unrelated to the merle coloring. Dismissing eye problems with the assumption that the heterozygous offspring will not be affected could be a mistake, because there is no way to be sure that the homozygous parent has otherwise normal eyes.