Portable instruments such as ocean bottom seismographs and the PASSCAL recorders often use rugged, portable geophones. The desire to use such sensors for relatively low-frequency work has raised questions about the limits of their sensitivity. The lower and upper frequency limits of performance of seismic sensors are determined by the sensor's mass, period, and Q, and by the amplifiers used with those sensors. We have tested Mark Products 1 Hz, 2 Hz, and 4.5 Hz velocity transducers against Streckeisen seismometers in order to examine the limits of their performance in measuring ground noise, particularly at low frequencies. Among the velocity transducers, only the 1 Hz Mark Products L-4 sensor provided good resolution of the 6-sec microseism peak. For this sensor, the lower limits of sensitivity was at approximately 0.06 Hz, although this depends on the amplifier used and the noise level at a given site. The amplifiers examined included conventional, low power, and commutating auto-zero operational amplifiers. It was found that the noise levels of the amplifiers intersected the ground noise level at frequencies ranging between 0.06 and 0.2 sec, depending on the amplifier and the exact circuit design. Measurements indicated that by modeling the amplifier noise for a given circuit correctly, the performance of an amplifier can be predicted with a high degree of accuracy, obviating the need for actual circuit construction to determine performance in the field. Given the very steep slope of the ground noise spectrum between 0.05 and 0.1 Hz and the rapid fall off in a seismometer's output below its resonant frequency, it would require a lowering of amplifier noise by more than an order of magnitude to be able to resolve ground noise at frequencies lower than 0.05 Hz using relatively small geophones such as the L-4. To resolve ground noise at lower frequencies, it is necessary to use a seismometer with a displacement transducer to sense the mass position, such as Guralp or Streckeisen sensors.