Bind Zones split by octet

Why are the reverse Zone files in Bind split by octets instead of subnets?

This is what I would expect to see:

subnets: 
  172.0.208/22
  10.0.2.0/25
  10.0.2.128/25host# ls /path/to/zone/files
208.0.172.in-addr.arpa
0.2.0.10.in-addr.arpa
128.2.0.10.in-addr.arpa

A query for 172.0.209.1 will be for a record at 1.209.0.172.in-addr.arpa, which isn’t under the 208.0.172.in-addr.arpa zone. This is also why you can’t have a reverse DNS delegation for a subnet smaller than /24, either; there’s no ability or logic in the DNS standard to “split” the delegations for these zones into smaller chunks.

DNS serves name data; it has no special logic built around serving data for subnets, so the standard for reverse zones conforms to the “DNS-looking” format of the dot-separated numbers.

Check more discussion of this question.

Related posts:

1. Why are these reverse lookup DNS containers grayed out?
2. In a 2 DC environment, should both DCs host AD integrated primary zones to ensure DNS redundancy?
3. BIND9 DNS for Round Robin routing
4. Ubuntu Bind9 Zones
5. How can I set up a local nameserver and modify DNS zones on it?