H-scan is a new ultrasound (US) tissue classification technology. Previous work has consistently found a strong correlation between H-scan US image intensity and US scatterer size in the medium being studied. However, there exists a subtle difference between theoretical H-scan US transmit pulses and that currently used in experimentation. In both simulation and experimental studies, we explored use of a traditional Gaussian transmit pulse shape and one that is more consistent with H-scan US theoretical considerations, namely, a Gaussian-weighted Hermite polynomial transmit pulse. For experimental studies, the shape of the US pulse was modified by using the arbitrary waveform generator functionality on our programmable US scanner (Vantage 256, Verasonics Inc) equipped with a L11-4v transducer. Simulated and experimental studies were conducted using homogeneous tissue-mimicking phantoms embedded with small spherical US scatterers of varying size (i.e. 15, 30 and 40 µm in diameter). Overall, these preliminary studies revealed an improved sensitivity of H -scan US imaging to changes in scatterer size after modifying the US transmit pulse to more closely match theoretical considerations in simulation and experimentation.