I'm a little confused as to who is missing. After re-reading I'm thinking there's a missing 't after can and both 2GGPs are missing.
From my experience, the 1860 census was one of the most poorly taken and transcribed in U.S. history.
There are some areas where the census wasn't taken. These are "mostly uninhabited" places that were new to the union, and places that weren't states yet. In this case "inhabited" means by full American citizens. Some parts of the High Sierras in California might be one example (in the county my ancestors lived in 1860 was the first census taken there). Dayton, Indiana certainly would have been taken.
The census is mandatory, but there is little to no teeth in enforcing it. The mandate falls more to the census taker than the citizen. More, you didn't need to provide the information in order to end up in the census, so refusal did little good. The census takers would ask the questions of whoever was home (could even be a hired hand), a neighbor, or from personal knowledge in some cases. There might be occasions where the census taker got lazy, no one was home, the census taker assumed the house to be unoccupied, or a home was so remote that no census taker got out there.
Some things I've found about ancestors "missing" in the census in my own searches:
-A husband and wife were separated. She lived at home, he was in a boarding house hundreds of miles away. I had no reason to look for him there.
-The census was actually taken retrospectively. Meaning the census taker came after the census date, and asked about who was living there on the census date. Sometimes this leads to wrong answers. For examples, when people moved, were in hospital, or died near the census date sometimes they got recorded twice or not at all when they should have been.
-Using electronic search engines to locate an ancestor in the census:
The search engine is only as good as the data put in to the database being searched. From hand written records this requires transcription, allowing for human error. I've seen a name so badly butchered in a transcription you would never recognize it. Upon comparing the image and the transcription, I could see where the transcriber misread things, but the correct spelling was very clear. The only reason I found that particular one is because I was looking at the image page because another ancestor happened to live down the street.
The search engine is also only as good as the searcher. I've seen several occasions where researchers would stick to too strict guidelines. Spelling didn't count for much. Ages got messed up. A wife or mother in law might give wrong information. For example, if you're looking for John Smith born 1841 in Kentucky and you know he lived in Missouri for a while as a kid, you should also consider Jack Smythe born 1831 in Missouri as a possibility. Sometimes removing information from the search and looking at each return helps find the right one.
-Cannot confirm or deny this is the right ancestor....
Some of my ancestors went by their middle names, or got recorded by initials. Benjamin Frank Smith sometimes shows up as Benjamin, Frank, or B.F. Smith. Sometimes there isn't enough information to know if the record pertains to my ancestor or another B.F. Smith.
Some tricks:
If you know who else might have been in the house, like a child, look for them. Likewise, if the grandparents could have lived with a married daughter, look for their son in law.
If you know where your ancestor lived, get the images for the entire enumeration district or town and physically look at each image. This can be time consuming, but sometimes it's worth it. You could do this by microfilm, or ancestry.com allows you to move to images forward and back within collections.
Look at marriage records for your 2GGPs and their children. Were they already married in 1860? Were their kids married sometime between 1850 and 1870? This might help give you a place to start looking.
Post their names in a new question, some of us might take a crack at looking.