Step 1: Recall how quadratic probing works. On a collision at index \( h(k) \), it checks slots \( h(k) + i^2 \) for \( i = 1, 2, 3, \dots \) instead of the next slot.
Step 2: This spreads probes out, so it avoids the long runs of filled slots that linear probing creates. That problem is called primary clustering, and quadratic probing fixes it.
Step 3: But there is a catch. Any two keys that hash to the same home index follow the exact same probe sequence. So they still pile up along one path. This is called secondary clustering.
Step 4: Option A is wrong because it does handle collisions. Option B is not a standard term. Option D is not a defining drawback. The correct answer is secondary clustering, option C.