Here, we present examples of in situ measurements of anvil cirrus from the CRYSTAL-FACE and TWP-ICE missions. As in other cirrus datasets, the small crystals are indicated by the measurements throughout the vertical extents and ages of the anvils sampled. We compare these measurements with cloud-resolving model simulations of deep convection/anvil cirrus systems as well as remote sensing retrievals of cloud-top effective radii. The in situ measurements, remote-sensing measurements, and simulations all indicate numerous small crystals near the tops of the anvils. However, the bases of the simulated anvils consist of only much larger crystals that precipitated from higher levels. We explore various processes not included in the cloud model that might produce the ubiquitous small crystals indicated by the measurements. We suggest that the most plausible explanation for the model/measurement discrepancy is shattering of larger crystals at the inlets of the aircraft instruments.