Comparisons were made between the results of testing muscle spindle primary afferents with low-frequency, large-amplitude sine waves of muscle stretch and with repeating ramp-and-hold stretches. Three different situations were explored, namely, activation with succinylcholine (SCh), generation of various fusimotor outflow patterns by brainstem stimulation and fusimotor activation by stimulation of single ventral root gamma-axons. With ramp-and-hold stretches the effects of bag1 fibre contraction were thought to be generally better represented by increases in the dynamic response to stretch (delta DD) than by increases in dynamic index. The equivalent effect on the sine wave response was an increase in the magnitude of the fitted sine (delta Mag). Static effects were best measured by increases in the initial frequency for ramps (delta IF) and by increases in the fitted centre frequency (delta CF) for sines. There was good agreement between sine and ramp testing in the case of SCh effects, but much less consistency during brainstem stimulation. Sine wave testing can be regarded as less efficient than ramp testing in distinguishing between static and dynamic actions because dynamic fusimotor stimulation causes equal rises in CF and Mag, but fourfold larger rises in DD than in IF. Ramp responses with combined dynamic and static fusimotor stimulation showed increases of both IF and DD as expected, but whilst sine wave responses always showed increase in CF, they often showed no increase in Mag. Comparisons of the effects of SCh, central stimulation and fusimotor stimulation showed that it is important to recognize the different effects of chain and bag2 intrafusal fibre activation as components of static fusimotor excitation. Stimulation of driving static axons (chain activation) raised the IF and CF often with 1:1 or 1:2 driving during the shortened phase of ramps and sines. However, the stretch response in both cases was either unchanged or reduced. Stimulation of a non-driving static fusimotor axon (bag2 activation) caused a marked rise in CF for sine waves and a fall in Mag, but a modest increase in ramp stretch sensitivity (DD) with no increase in dynamic index. A way is described for measuring flattening of sine wave responses as a possible way of detecting chain fibre activation. The consequences of these findings are discussed in relation to previous reports of the use of sine wave testing for assessing fusimotor mechanisms.