Class C addresses represent the smallest allocatable unit in the classful addressing hierarchy, designed for the numerous small organizations that would never need thousands of IP addresses. With only 254 usable hosts per network, Class C was intended for small businesses, branch offices, and individual departments.

Yet this apparent precision created its own problems. Organizations needing 300 addresses faced an impossible choice: accept a single, inadequate Class C (254 hosts), or manage the routing complexity of multiple discontiguous blocks. This 'too small, too many' paradox made Class C both the most numerous allocation category and the most administratively burdensome.

The first-octet range of 192 to 223 encompasses over 2 million potential Class C networks—a vast pool that proved essential for the Internet's democratic expansion, even as its limitations drove the CIDR revolution.

Class C addresses are identified by their leading three bits being '110'. This three-bit prefix creates the precise mathematical boundaries that define Class C behavior:

Binary anatomy of a Class C address:

Mathematical derivation of the range:

• Minimum first octet: 11000000 = 192
• Maximum first octet: 11011111 = 223

Therefore, any address with first octet between 192 and 223 is definitively Class C. The second and third octets provide 16 additional bits to distinguish among the 2 million+ networks in this class.

The default subnet mask: Class C addresses carry an implicit mask of 255.255.255.0 (or /24 in CIDR notation). This is the most common subnet mask in networking—so common that /24 has become nearly synonymous with 'a network' in casual discussion.

Class C's address space spans from 192.0.0.0 to 223.255.255.255. Despite having the smallest per-network capacity, Class C represents half of the routable IPv4 address space by sheer network count:

Network identification example:

Class C's defining characteristic—only 254 usable addresses—created a dual-edged reality. For truly small organizations, it was perfectly sized. For medium organizations, it was maddeningly insufficient.

The allocation efficiency paradox:

The multiple-block management nightmare:

Organizations requiring more than 254 addresses had to request multiple Class C blocks. Unlike a single Class B (which provided contiguous 65K addresses), multiple Class Cs often arrived from different parts of the address space:

• 203.0.113.0/24 (first allocation)
• 198.51.100.0/24 (second allocation, months later)
• 192.0.2.0/24 (third allocation, from different registry)

This created severe operational challenges:

RFC 1918 designated 192.168.0.0/16 for private networking—the range that became synonymous with home and small office networks worldwide. Despite being called 'Class C private space,' it actually spans 256 contiguous Class C networks.

Why 192.168.x.x became universal:

The ubiquity problem:

While 192.168.x.x's universality is convenient, it creates challenges:

• VPN conflicts: Connecting to a corporate VPN while your home uses 192.168.1.0/24 and the office also uses 192.168.1.0/24 causes routing conflicts
• Network mergers: Acquiring companies with identical 192.168.x.0/24 addressing requires renumbering
• Guest device assumptions: Devices may expect gateway at .1 or .254 based on home network patterns

Enterprise networks often avoid 192.168.x.x entirely to prevent conflicts when employees work remotely.

If Class B caused routing table concerns, Class C created a genuine crisis. With over 2 million possible networks, each requiring a separate routing entry, backbone routers faced memory exhaustion as Class C allocations proliferated.

The problem scale:

In the early 1990s, backbone routers had 4-16 MB of memory. If even 10% of Class C networks were allocated and announced, routers would exhaust memory purely on routing tables. This wasn't theoretical—routers were failing as Class C allocations accelerated.

The CIDR rescue:

CIDR (RFC 1519, 1993) transformed Class C management through supernetting and provider-based allocation:

1. Supernetting: Instead of receiving scattered Class Cs, organizations received contiguous blocks from their upstream provider (e.g., 203.0.112.0/22 = four contiguous Class Cs)

2. Aggregation: Providers announced summary routes covering all customer allocations, dramatically reducing global routing table size

3. Hierarchical allocation: RIRs allocated large blocks to providers, who sub-allocated to customers, enabling aggregation at each tier

Several portions of Class C space are reserved for special purposes, reducing the publicly allocatable range. Understanding these reservations prevents configuration errors and clarifies documentation examples:

Using documentation ranges correctly:

When writing network documentation, training materials, or code examples, always use TEST-NET ranges rather than randomly chosen addresses. This prevents:

• Accidental connectivity to real networks during testing
• Legal/policy issues if examples resemble operational networks
• Confusion if examples later appear in production logs

While Class C networks are small (254 hosts), they can still be subnetted for segmentation, security, or address conservation. Subnetting a /24 creates even smaller networks—a practice common in modern enterprise and cloud environments.

Common Class C subnet divisions:

Class C terminology and concepts permeate modern networking, even though CIDR has technically superseded classful addressing. The /24 boundary inherited from Class C remains the most common network size:

The /24 minimum routing requirement:

Most Internet backbone operators have policies rejecting routes more specific than /24 to prevent routing table explosion. This effectively makes /24 (a single Class C) the smallest independently routable block on the public Internet.

Implications:

• Organizations cannot announce /25 or smaller to the global Internet
• For smaller allocations, traffic must route through a provider's aggregate
• IPv4 addresses in blocks smaller than /24 have reduced utility and market value

Class C completes the allocatable portion of the classful hierarchy—small networks designed for the numerous organizations that needed only modest IP capacity. Yet its limitations proved as instructive as its utility. Let's consolidate the key concepts:

What's next:

With Classes A, B, and C (the unicast allocation classes) complete, we'll explore Class D (Multicast) addresses—a fundamentally different addressing paradigm not for host identification but for group communication. You'll understand how addresses in the 224-239 range enable efficient one-to-many transmission without unicast replication.